The Perception of The Visual World-GibsonJJ PDF

THE I'ERCEPTION OF THE
YISUAt
By James J. Gibson
KOUGHTON MIFFLIN COMPANY BOSTON

The Riverside Press, Cambridge
PDF compression, OCR, web optimization using a watermarked evaluation copy of CVISION PDFCompressor
UNDER THE EDITORSHIP OF Leonard Carmichael
Secretary, Smithsonian Institution;
formerly President, Tufts College, and
Director, Tufts Research Laboratory of
Sensory Psychology and Physiology
Copyright, 1950
By James J. Gibson
All rights reserved including the right to re-

produce this book or parts thereof in any form.
THE RIVERSrDE PRESS
Cambridge, Massachusetts
Printed in the U.S.A.
tdftor's Introduction
Artists and philosophers as well as physicists, physiologists, and

psychologists have long been interested in isolating the factors which
make possible man's visual world. Physicians who are interested in the
eye and its diseases, illuminating engineers, photographers, designers
of photographic and motion picture equipment, and many others in pure
and applied science are also concerned with certain aspects of this
complex subject.
The present book represents the culmination of nearly a quarter of
a century of study of visual phenomena by its able author. He has ap
proached the problem in an eclectic manner. In its pages the point of
view of the student who is being introduced to the subject is never for-
gotten.
The author emphasizes the fact that a fundamental condition for
seeing is an array of physical surfaces which reflect light that is then
projected on the retina. He further gives new emphasis to the importance
of considering the retinal images of each eye as involving steps and
changes in gradients of light.
The student will find in this volume an interesting discussion of
the old and difficult problem of the nature of visual depth. The author
also deals with the constancy of the characteristics of perceived ob
jects in relation to geometric space and many other related topics.
Throughout the book theories of perception are carefully evaluated.
Certainly the present volume can be recommended to all artists and
vi EDITOR'S INTRODUCTION
scientists who are interested in learning about the nature of man's

visual world.
The study of perceptual problems is central in psychological theory.
The study of perception has sometimes been called "psychologists'
psychology" because professional students of man's mental life most
clearly recognize the basic nature of perceptual problems. Today, when
more than ever before many special applications of psychology are at-
tracting the attention of students in this field, it is fortunate that Dr.
Gibson has prepared this new and stimulating volume. It will again
direct the interest of serious students of psychology to the basic prob-
lems of perception, especially as they are related to this greatest of
man's distance senses.
Leonard Carmichael
reface
The principal subject of this book is the visual perception of

space. The essential question to be asked is this: How do we see the
world around us? The question is at once a theoretical one, a factual
one, and a practical one. The theories to be considered have to do with
the history of philosophy and psychology. The facts come from psychol
ogy, physics, and physiology. The applications extend to art, aviation,
photography, and mountain-climbing. This book, however, is not a
historical survey of the problem, nor a record of the existing facts, nor
a discussion of the applications. The intention is to formulate a con
sistent approach to the problem a way of getting new facts and making
new applications. The construction of a theory is most useful when the
theory is ttvu1nerab1e, that is to say, when future experiments can but
do not disprove it. A strenuous effort has been made to keep the proa'
positions of this book explicit enough to be potentially incorrect. Need
less to say, the author hopes that they will comprehend the facts and
will predict the results of future experiments.
A theoretical approach is called for because the perception ot
what has been called space is the basic problem of all perception. We
perceive a world whose fundamental variables are spatial and temporal
a world which extends and which endures. Space perception (from
which time is inseparable) is not, therefore, a division of the subject
matter of perception but the first problem to consider, without a solu-
tion for which other problems remain unclear. That a solution is 1ack
ing, most psychologists would agree. The existing theories to account
for the spatial and temporal character of our perceptions are not very
satisfactory.
viii PREFACE
This book has a great deal to say about the physical stimuli which
are the correlates of perception, but relatively little to say about the
activities of the sense organs and the brain which are also the correlates
of perception. The writer has elected to study psychophysics rather than
psychophysiology because he believes that it offers the more promising
approach in the present state of our knowledge. This is not to minimize
the importance of physiology. Such books as Bartley's Vision: 4 Study
of its Basis (6) are essential to an understanding of the complete
process. What we lack, however, ¡s an application of the psychophysical
methods to perception.
A psychophysics of perception may sound to some readers like a
contradiction in terms. This book undertakes, however, to justify and
make possible such a science. For many years, experimental evidence
has accumulated about the effect of the observer's attitude on percep-
tion, the influence of culture on perception, and the roles of past ex-
perience and of sensory organization in perception. All these experi-
ments, however revealing, leave out of account the simple question of
the relation of the stimulus to perception. Until this question is settled
the other evidence will be hard to evaluate.
Several recent currents of psychological thought have influenced
the writing of this book: the ideas of Gestalt psychology, of American
functionalism, and of what might be called dimensionalism. The
twentieth century scientists to whom I am most in debt are Kurt KoiTha,
Leonard T. Troland, and Edwin G. Boring. The hypotheses I have
adopted were precipitated by research in the field of military aviation,
carried out during the war.
Every book is a collaboration of its writer with others. The
hardest collateral labor that went into these pages was performed by
Eleanor J. Gibson, my wife, whose scientific conscience is stricter than
my own and to whom the reader ought to be very grateful1 This book is
for her, with thanks and affection. The text has also been combed by
Leonard Carmichael, editor of psychological books for Houghton Mifflin
and one of my earliest teachers, with so much insight and erudition that
I can never repay him. At an early stage of the manuscript it was care-
fully read by S. Rains Wallace who made the kind of detailed and
penetrating comments that only a genuine friend is capable of. I am
PREFACE
grateful likewise to a number of other friends and colleagues who have

gone over large or small parts of the manuscript: Robert B. MacLeod,
John Volkmann, T. A. Ryan, Edwin G. Boring, Wolfgang Kohier, Hans
Wallach, Annalies A. Rose, Fritz Heider, R. T. Sollenberger, H. E.
Israel, Mervin Jules, and Oliver W. Larkin. Thanks are especially due
to Frederick N. Dibble, who has worked with me for several y ears in
testing experimentally some of the hypotheses to be described and who
helped formulate them. Finally, my debt must be acknowledged to
Robert M. Gagne and the co-workers of my wartime research unit who
performed the feat of behaving like scientists in a military community.
This is a book intended to interest anyone who has ever acquired a
sense of the awe'4nspiring intricacy of vision. No realm of inquiry of»
fers more strange and wonderful discoveries.
JAMES J. GIBSON
IIII TO LOOK AT THE ILLUSTRATIONS IN THIS BOOK
Many of the pictures are intended to give

an impression of depth or distance. This
effect will generally be clearer and more
vivid if you will close one eye, look at
the center of the picture, and hold it
somewhat closer than you are accustomed
to. You may have to wait a few seconds
for the full effect to occur. This rule ap-
plies to the photographs and drawings but
not to the cross-sectional diagrams.
::
Editor's Introduction
V
Preface
I
Why Do Things Look as They Do?
1
2
Theories of Perception
12
3
The Visual Field and the Visual World
26 I
4
The Formation of Retinal Images
44
,1
5 y,,
A Psychophysical Theory of Perception
59
6
The Stimulus Variables for Visual Depth and Distance - Momentary Stimulation
77
xii
CONTENTS
7
The Stimulus Variables for Visual 1)epth and Distance - The Active Observer
i 17
8
The Problem of the Stable and Boundless Visual World
145
9
The Constant Sizes and Shapes of Things
163
Io
Geometrical Space and Form
188
II
Meaning
197
12
Learning
214
13
Spatial Perception and Spatial Behavior
223
R eferences
231
Index
239
'p he
I Perception of the Visual World
r or this end the visive sense seems to
have been bestowed on animals, to wit,
that by the perception of visible ideas
they may be able to forsee the damage or
benefit which is like to ensue upon the
application of their own bodies to this or
that body which is at a distance; which
foresight how necessary it is to the
preservation of an animal, everyone's
experience can inform him."
George Berkeley, Bishop of

Cloyne, 1n Essay Towards
a 1'Vew Theory of Vision,
1709
a
Why Do Things Look as They Do?
The Theoretical Approach . . . The Initial Hy.

potheses of a "Ground Theory" of Space Percep.
tiori . . . . Sensation and Perception
This is a book about how we see. There fly an airplane, or even leap through the
are, as everybody knows, a number of con- air at the top of a circus tent. He can
ditions which have to be fulfilled before match colors and draw representations of
anyone can see: there must be light to see things. He can design and build machines,
by; the eyes tnust be open; the eyes must and he can change the appearance of the
focus and point properly; the sensitive film environment almost to suit himself. Or, as
at the rear of each eyeball must react to another possibility, he can simply sit and
light; the optic nerves must transmit im- look at the scenery.
pulses to the brain. Just so long as one This last, in a way, is the most as-
of these conditions is not fulfilled, the tonishing performance of all, for the view
seeing person is blind. People who have of a room or a countryside which one gets
not thought about the problem find it dif- when he simply looks at it in a receptive
uicult to realize that sight depends on mood has great scope and, at the same
such a complicated chain of circumstances, time, the most minute detail. The number
for seeing does not 'fee1 like" that. It of items that can be described in such a
fee1s as if" things were simply there. view is enormous. What is most astonish-
Nevertheless, such is the case. ing is that it is in every detail a nervous
Normal sight is an astonishingly good process. The panorarnais utterly and en-
guide for getting about and doing things. tirely a performance of the living organism.
A seeing man can walk without colliding If the brain is injured in a particular way,
with obstacles. He can use tools as fine a partial blindness results, and the
as a jeweler's needle and as large as a kind of blindness is related to the parti-
steamshovel. He can read print, or look at cular injury. If the optic nerve or the
pictures, or identify faces. He can dis- retina of the eye is damaged in some part,
criminate objects which resemble one an- sight suffers damage in a precisely cor-
other even at a considerable distance. All responding way. The simplest experiment
these a blind person cannot do. A seeing is to close one's eyes and reflect on the
man can climb a cliff, drive an automobile, fact that the visual panorama vanishes.
2 THE PERCEPTION OF THE VISUAL WORLD
The problem of visual perception has a this book will also be concerned with
long history. For hundreds of years men them. For example, how do we see the
have felt the need for some explanation of form or shape of a thing? This question
why things are seen. Among the many is not at all easy to answer. The search
puzzles to which the problem leads, perfor an answer takes one so far afield that
haps the oldest and most general of al] it provided, some thirty years ago, the
is this:fhow can one account for the rich- basis for a new approach to psychology -
ness of sight considering the poverty of the theory of Gestalt psychology. For
the image within the eye? Vision depends another example, how do we see the motion
on this retinal picture. But what an in- of a thing? Still more fundamental, how
adequate thing the image seems to be when do we see a thing - the mere object as
compared with the resu]t! (The visible distinguished from its general background?
scene has depth, distance, and solidity; Probably this last reduces to two ques-
the image is flat How can vision depend tions: first, how can we see an outline as
on the pictures in the eyes and yet produce separated from its background, and second,
a scene which extends to the horizon? how do we see a solid surface? There are
The physical eivironment has three dimen- many other such questions, not easy to
sions; it is projected by light on a sensi- formulate scientifically because they are
tive surface of two dimensions; it is per- all more or less interrelated. Why do
ceived neverthe]ess in three dimensions. things have location, that is, how can we
How can the lost third dimension be re- see where they lie? How do we see fine
stored in perception? detail, and what are the limits of this
This is the problem of how we perceive acuity? Why do things look right side up?
space. The question is put in terms of the Why does the world always appear level
geometrical dimensions of height, width, even when wehe down?
and depth. In a sense, this book is about There are also a whole set of practical
space perception. The plan of these chap- problems which depend on the so]ution of
ters, however, is to end with the problem the theoretical problems. How can men
of abstract space rather than to begin with see to fly airplanes and drive automobiles?
it. The space to be considered first is not What does the artist see when he paints a
a void with three ]ines intersecting at picture? Why is a photograph so astonish-
right angles but the space of rooms, ingly like the scene at which the camera
streets, and regions, and the space of men was pointed? How far must the movies
who walk, drive, or fly an airplane. The inevitably fall short of natural seeing? Can
puzzle of the third dimension can be much vision be improved by training? What is
better understood if we first examine the visual education and how may it be used
scenes we actually see and the ones which to advantage in school and college? These
are of practical importance for human practical problems will be touched upon,
behavior. but it is fruitless to look for their solution
The problem of how we perceive space without first laying the groundwork of a
implies a good many other problems, and scientific theory.
- i
w___. ' 1- , - 'w
íf
4' L;
J
'd1'
* i i '
p
A ' S ; ;:q
.
¿* ..p
, Lp' 4p
"
p
. . p
.p
-, 4.r
4pP
:-I p'
e'
P( P3p
*
p..
(
p
1 p
tp - p
p
__p
p p' '
:
p
,9:- ,-
p
iPt? :.e
.
pp p
'p
* ' ' p,
r' -
p,
a' s I, P 'PiS _
FIGURE 1. A Scene for Analysis
There is another problem with which we objects. accepted terminology, our

In the
sha]1 be indirectly concerned since upon problem is that of perception, not of sensa-
our knowledge of it everything else de- tion.
pends how do we see light and how can All these problems can really be summed
we perceive color? Light and color are, up in a single general question: Flow do
in a way, the raw material of vision. The we get the experience of a concrete visual
perception of an object in space would be world? The visual world can be described
impossible if we were not sensitive to the in many ways, but its most fundamental
light reflected from the object and to the properties seem to be these: it is extended
brightness and hue of this light. There in distance and modelled in depth; it is
is a vast accumulation of evidence about upright, stable, and without boundaries;
brightness and hue. Nevertheless, this it is colored, shadowed, illuminated, and
evidence is not enough to provide answers textured; it is composed of surfaces, edges,
to the other questions, inasmuch as the shapes, and interspaces; finally, and most
seeing of an object is an ability quite important of all, it is filled with things
different from the seeing of abstract color. which have meaning. If we could account
Seldom or never dbes one see a color as for the perception of these properties of
such. This is primarily a book about the visual 'world, we should at least be
well on the way to explaining the whole contour, shape, adjacent location, illumi
panorama of visual experience. nation, and shading. The list is tentative
Examìne the scene reproduced in Figure and incomplete but it illustrates the kind
1. It represents an uninteresting dry river of problems with which the contemporary
bed of the sort common in the Southwest, study of space perception is concerred.
bordered by a tall growth of bushes, with There are, of course, other properties of an
two men standing in the foreground. It will actual scene which do not show up in a
serve as an example of what is meant by a photograph but are nevertheless important
concrete visual world. Let us consider it for space perception. Chief of these are
abstractly without regard to the familiar the stereoscopic impressions dependent on
meanings which can be applied to it. The vision with two eyes, and the vivid quali-
bottom part of the picture (the "ground") ties of depth which occur when the head is
looks solid whereas the upper part (the moved. The contributions of these im-
t1sky") does not. The solid part looks pressions to the perception of space have
generally level and this surface appears to been known for a long time but they are
extend to a great distance. Actually it is not, as is sometimes believed, the ex-
a compound of visual surfaces (ground, clusive basis of our perception of a three-
bushes, men) separated by contours. One dimensional world.
of the most prominent contours is the In contrast with the substantial world
horizon. The various surfaces have the represented in Figure 1, 1er us imagine the
quality of texture, sometimes fine and perception obtained by an observer in the
sometimes coarse, although the sky does nearest possible approach to empty visual
not have this quality. Some of them have space. Assume that his environment con-
closed contours or shapes and they are sisted wholly of atmosphere without any
located with reference to the ground. Parts opaque objects. He could live for some
of the ground appear to be illuminated and time at the center of such a sphere of air
other parts shadowed. Most abstractly of although, without the gravity of the earth,
all, the whole scene is composed of a he could not maintain a posture or change
pattern of light and dark, that is, an enor- his location. Suppose that this environment
mously complicated mosaic of grays, is illuminated by external sources but that
blacks, and whites, with variations (not his world of air is so large as to diffuse
represented in the photograph) of yellow the light evenly, as our familiar sky tends
and brown, dusty green, and vivid blue. to diffuse the light of the sun. If he opens
Granting that the picture, although it his eyes he can see, and the question is
fails in some ways to look like the actual iv/tat will he see?
scene, is quite similar to it, what makes There are experiments which yield a
it such a good substitute? Analysing it, reasonably sure answer to the question.
the properties which give it the appearance The light which stimulates the retinas of
of concrete visual reality seem to be just his eyes will be homogeneous (67), that
those listed; surfacequality, solidity, is, the same at all points. 11e can turn
horizontal character, texture, distance, his eyes in any direction but they will
WHY DO THINGS LOOK AS THEY DO? 5
not focus or converge and he cannot fixate The hypothetical man at the center of a
or look at anything for there is nothing to sphere of pure air is even further instruc-
fixate on. Ile will see luminosity or color tive. Although he would presumably have
but it is the kind of color which Katz has no impressions of far or near, and no sense
named film-color as distinguished from of his surroundings as being either flat like
surface-color (61). It is unlocalized in the a picture or modeled like a sculpture, that
third dimension; its distance is indeter- is not all he would lack. Almost certainly
minate. The sea of light around him might he would have no impression of up and
vary from bright to dark and from one hue down. Since the pull of gravity on his
to another but the quality of color would body and the resistance of his legs against
be neither that of a surface on the one hand the substratum are wholly lacking,he would
nor would it be extended in depth on the have no equilibrium and could not maintain
other. It is neither near nor far. The space a posture. He would feel as if he were float-
he sees is certainly not two-dimensional ing. Although he could look toward or away
in the sense of being flat but it is also not from his feet and could see his right hand
three-dimensional in the sense of being and his left hand, these acts would probably
deep. Assuming the atmosphere to be ,have lost much of their normal meaning of
cloudless, without dust or fog-particles, it up or down, right or left, and he wtuld ex-
has no texture, no arrangement, no con- perience a profound and complete disori-
tours, no shapes, no solidity, and no hori- entation. He could thrash about but could
zontal or vertical axes. The observer not change his position in phenomenal
might as well be in absolute darkness so space, and in fact he would have no posi-
far as he can see anything. The results tion in a visible environment. Pis sense
of this hypothetical experiment suggest, of the vertical and horizontal directions
then, that what an observer would perceive (ordinarily given by the stimuli for his
in a space of air would not be space but postural reflexes and. by the main lines of
the nearest thing to no perception at all. his retinal images of the horizon and of
The suggestion is that visual space, un- trees, tables, and rooms) would be wholly
like abstract geometrical space, is per- lacking. Since he would have no axes of
ceived only by virtue of what fills it. reference for his space it is questionable
'Besides the experiments of Katz on film- the illumination until nothing was seen but
color (61) there are also the results of Metz- film or fog. The conclusion of these experi-
ger on homogeneous light stimulation over the ments was that a visual surface depends on
total visual field, the Ganzfeld (81). Taken the perception of tm1crostructure, that is,
together, they suggest the above results for a the minute inhomogeneities of reflected light
hypothetical observer floating in air. Katz which give it a texture or grain. These re-
studied taperture_colors the appearance suits are to be contrasted with the theory of
of a hole in a surface behind which is another Bühler that space might be given by a hypo-
surface too distant to yield the perception of thetical "air-light" - a direct seeing of the
a surface. Metzger studied the appearance of atmosphere dependent on molecular particles
a uniform surface which filled the whole field ( 16). This supposition has never received con-
of view. He made it homogeneous by reducing firmation, and Bh1er himself abandoned it.
whether he could be said to perceive even experiments failed to clarify the practical
an abstract geometrical space.2 problems of how a man lands an airplane.
Many tests were devised but none of them
The Theoretical Approach
Ar the beginning of World War II there
predicted a prospective flier's success or
failure at this task. Many suggestions for
was a sudden need to understand the
perception of depth and distance as they training were made but none of them made
the performance substantially easier.
applied to aviation. The critical task of
estimating distance froni the ground when a Toward the end of the war it began to be
flier is landing an airplane was particularly evident to psychologists working on prob-
important. Research was begun and lems of aviation that the usual approach
studies of space-perception multiplied, to the problem of depth-perception was in-
based on what psychologists already knew correct. Experiments needed to be per-
about it from the great experiments of the formed ')Utdoors. The stimuli to be
judged ought to be those of a natural
19th century. A list of the clues or cues
for the perception of the distance of an environment. A hypothesis with a vast set
object had resulted from these experiments of new implications (new at least to the
and this list had gained acceptance. The writer) began to assert itself - the possi-
cues were classified as binocular or
Ç
p hility that there is ltera/iy no sucii thing
(iS a perception of space 14)IthOlIt the per-
monocular according to whether they de-
pended on the use of two eyes dr one eye. ception of a c')n tiflUOUS bach ground sur-
The typical means of experimenting was face. This hypothesis might be called
a "ground theory" to distinguish it from
to employ a stereoscope, or a depth-per-
ception apparatus, or a dark room in which the t'air theory" which seemed to under-
points of light or similar isolated stimuli lie the earlier research. A few experiments
were performed by the writer and his colla-
appeared. The points, lines, or objects
whose distance was to be judged usually borators before the war ended using out-
door situations, photographs, and motion
appeàred against a homogeneous back-
ground. The fact was, however, that these
pictures, in which a level ground was
always visible (39).
2There have been almost no experiments The basic idea is that visual space
which study the effect of eliminating corn-
pletely the force of gravity on the perceptions should be conceived not as an object or an
of a human observer. They are needed, since array of objects in air but as a continuous
the rocket-passenger outside the earth's gravi- surface or an array of adjoining surfaces.
tational field will meet just this condition, and
it is no longer fantastic to be concerned with The spatial character of the visual world is
the problem. A man falling freely toward the given not by the objects in it but by the
earth satisfies the condition, but volunteers
for such an experiment are rare. In any event, background of the objects. It is exemplified
there has been no instance in which a man by the fact that the airplane pilot's space,
without postural stimulation has also been
presented with absolutely homogeneous visual paradoxical as it may seem, is determined
stimulation. A free-falling parachutist can by the ground and the horizon, not by the
always see the horizon. The description
above is therefore speculative, although air through which he flies. This conception
consistent with such evidence as exists (42). leads to a radical reformulation of the
,.
/' -
FIGURE 2. The Look of the World from the Air
stimuli or cues for depth and distance. we need is a new theory rather than new
Instead of investigating the dierences in evidence. The "air theory" of visual
stimulation between two objects, the ex space is actually inconsistent with a good
perimenter is led to investigate the varia- many experimental results. But, as
tions in stimulatiGn corresponding to a Conant has remarked of the history of
continuous background. This shift of science, "a theory is only overthrown by
emphasis has a great many implications, a better theory, never merely by contra-
and these will be explored in the ensuing dictory facts" (24).
chapters.
This "ground theory" of visual space The InitialHypotheses of a "Ground
is the organizing scheme of the present Theory" of Space Perception
book. The classical problems and facts What are the main principles of such a
of perception will be considered, but not theory? Since they determine the plan of
in the order in which they were dis- the book, it might be well to summarize
covered and not under the usual headings. them at the outset. Their explanation and
If our scientific conception of space per- factual status will be given in later chap-
ception was inapplicable to aviation, what ters.
I Th e re s s io n s o f a
e L em en ta ry ¿riz p is a gradual change along an axis of the
visual world are those o f surface (md edge. retinal image, an increase or decrease, for
These are the fundamental sensations of instance, in the density of the texture of
space, the stimuli for which need to be the image. Likewise,a stimulus ought to
discovered. They are elementary, how- be discoverable for an edge or contour and
ever, not in the sense that atoms or units for the impression of depth at a contour.
are supposed to be elementary, but only Perhaps this is a jump or discontinuity in
in the sense in which a variable or quality a gradient of the retinal image.
of something is essential to understanding The policy of searching for a stimulus
it. These candidates for the status of variable with which some quality of experi-
sensations are very different from the ence may prove to be in correspondence is
elementary impres sions of location assumed the policy which underlies psychophysical
by the traditional approach to space per- methods in psychology (40). It is the first
ception. Such elements were arranged step in the explanation of experience.
from right to left, from up to down, and Some would argue that there is no real ex-
from near to far according to the abstract planation of perception until the physiolo-
coordinates or dimensions of geometry.) gical mechanism s have been discovered,
The impression of a Continuous surface but this is a matter of preference. There
may account for visual space conceived as are laws relating perception to ph sical
a background. The impression of an edge stimulation as well as laws relating it to
may account for an outline or figure against physiological processes. Explanation is a
the backgroundthe Ufigureground pheno- matter of lawfulness, although there are
menon"and together vith the surface en- different levels of explanation. The level
closed may account for the perception of to be aimed for in the present book is a
an oLject. psychophysical theory, not a physiological
2 1 Tu ere is (1 IW(1 ys s on e varia hie in theory.
stimulation (however difJicult it ma) be to 3. The s ti)'ÎÌUÌlLS-L'ari(Zb!C u'i thin ti, e
discover and isolate) ?1'IliC/l corresponds to retínal image to ioliicli a ¡)ropert of visual
a pro)erty' of the Sfl(LtiUl ,orld. This space corrccponds need be Orti) a corre-
hypothesis says that even complex per- l(ltt of that property, not a copy of t t.
ceptual qualities must have stimuli. It is The qualities of solidity and depth, for
an extension of the principle of psycho- instance, do not have any replica in the
physical correspondence to visual percep- two-dimensional retinal image but they
tion-.-the principle which has served so may very well prove to have correlates
well in the study of sensation. ; there. An assumption will be borroved
This rule suggests that a "stimulus" from geometry which states that when a
can be found for the impression of a sur- three-dimensional physical world is pro-
face. Probably it is ma textured retinal jected optically, the slant and shape of its
image. A stimulus ought to be discoverable surfaces undergo a mathematical trans-
also for the quality of distance or depth formation in the projection but that they
over a Continuous surface. Perhaps this do not on this account vanish or disappear.
WHY DO THINGS LOOK AS THEY DO? 9
There is a naive theory of perception to ferentiated visual image is variously

the effect that the outer world somehow named. how do we perceive form, pattern,
gets into the eye. Almost the first prin- configuration, order? Why is vision or-
ciple the beginning student learns is ganized, structured, detailed, precise?
that nothing gets into the eye but light. The greatest achievement in the theoreti-
This third assumption can be sharpened by cal struggle with this problem has been
saying that, in a special sense, the outer reached by Koffka in his Principles of
world does get into the eye. It implies Ge.stzlt Psychology ((7). An attempt will
that at least the surfaces, slopes, and he made iiI Chapter 5, however, to follow
edges of the world have correlates in the a different theoretical path and to suggest
retinal image specifically related to their that a so-called pattern of stimuli is it
objective counterparts by a lawful trails- self a stimulus. The term pattern is vague
formation. If this is correct, the problem and unanalysed. The mathematical con-
of the restoration of the lost third dimen- ception of order, as exemplified by the
sion in perception is a false Noblem. number-series, is more exact. An effort
There is another naive theory of the will be made to show that a few simple
visual process to the effect that a retinal variables of pattern - texture, contour,
()i(t1Lre iS transmitted to the brain by the and density of texture - are definable as
optic nerve. In a more sophisticated form variations of adjacent order in the retinal
it is tempting even to the visual scientists, image.
although it leads to difficulties. According Tlìe experimental study of what was
to the first part of the hypothesis, however, called ifl/U)fli ugen eit' or di(ferentiation of
there is no need for a picture-theory of the retinal ímage has mostly been carried
psychophysical correspondence since per- out under a different name and with quite
ception may be a correlate, not a copy, of a different intention. lt has been called
the image. If the image is neither a the study of visual aruit'. A great many
replica of the world nor a picture for the experiments have been carried out on
perception but a complex of variations, acuity, but in them only a few kinds of
it may prove easier to trace its specific inhomogeneity have been studied: the
correspondence to both. separateness of two adjacent spots or
4. The in/tornogençities of the retinal two parallel bars, the gap in a broken ring,
image can be analysed by the methods uf the impression of a single line, a grating
TI U 777 ge o 'n L' try in to a
b e r tu e o ry an (I TI? () (IC 171 of dark and light bands, and the familiar
set of variables analogous to ¿lie variables letters of the acuity test. It can be argued»
of physical energy. This says, in effect, that these are artificial rather than natural
that the order or pattern of the retinal types of stimulation. An attempt to connect
image can be considered a stimulus. It is acuity, or "resolving power," with the
the most debatable and least developed of more general idea of a differentiated,
the hypotheses being summarized. The patterned, or textured image will be made
problem of the abstract nature of a dif- in Chapter 6.
lo THE PERCEPTION OF THE VISUAL WORLD
5. The problem of how we perceive the commonly ascribed to it, the best hope of
visual world can be divided into two prob- understanding these defects is first to
lems to be considered separately, first, the examine the respects in which perception
perception of the substantial or spatial is adequate and exact.
world and, second, the perception of the The method of investigating adequate
world of useful and significant things to i impressions of a substantial or spatial
which we ordinarily attend. The first is i world is the psychophysical experi-
the world of colors, textures, surfaces, ment. This is, essentially, a procedure
edges, slopes, shapes, and interspaces. of isolating and then systematically vary-
The second is the more familiar world with ing a feature ofthe physical stimulus for
which we are usually concerned, a world an observer who makes judgments of
of objects, places, people, signals, and ttmoreor "less," or otherwise shows that
written symbols. The latter shifts from h discriminates the variation. Although
time to time depending on what we are this method has been very little used in
doing at the moment, whereas the former the study of perception (as distinguished
remains a more or less constant back traditionally from sensation) there is every
ground for our experience, and a sort of reason to think that it can be applied (40).
support for maintaining posture and for The attempt to do so can be called a
moving about. The world of significant psychophysical approach to the study of
things is too complex to be attended to perception. It involves searching for son-ic
all at once, and our perception of it is feature of the physical stimulus with which
selective. Certain features stand out to set up an experiment.
prominently, others are neglected. lt is The method usually employed in the
sometimes said that our perception is past for the study of perception is funda-
distorted and falsified by this fact. This mentally different from that of the psycho-
kind of perception can be called physical experiment. lt was a policy of
whereas the first kind can be called literal. searching for discrepancies rather than
Before one can fully understand schema- correlations between the stimulus and the
tic perception one must understand literal perception. Assuming that sensation is
perception since it provides the funda- dependent on stimulation but that percep-
mental repertory of impressions for all tion is not, the policy of the experimenter
experience. This is primarily a book about has been to isolate and study these dis-
literal perception, therefore, and only crepancies. A favorite device for en-
secondarily a treatment of schematic per- hancing them has been the tachistoscope
ception. The discussion of the meaning- which presents an image to the observer
ful visual world is deferred until the enti for only a fraction of a second. The method
and does not pretend to be complete. is one of "impoverishing" the stimulus, or
Although it is true that everyday per- reducing the optimal conditions for literal
ception tends to be selective, creative, perception which characterize the psycho-
fleeting, inexact, generalized, stereo- physical experiment. Brief exposure, low
typed, and to have all the other defects so illumination, many stimuli in succession,
WHY DO THINGS LOOK AS THEY DO' ii
and the use of indefinite, ambiguous, or ceptions were built up out of elementary
equivocal stimuli have all been employed sensations, a more defensible idea has
in this kind of research. The resu1tjm been gaining ground: that of variables
body of facts is very large, but only a or dimensions of all experience, per-
small part of it can be considered iii the ceptual as well as sensory . Such vari-
present book. ahies as the texture and slant of a sur-
face are, no doubt, a far cry froiiì the
Sensation and Perception variables of hue, brightness, and satura-
tion of color. Rut, if it is no longer to be
The approach outlined above is not assumed that the mind constructs the sur-
consistent with the usual meanings of face out of bits of color, the ualitics of a
the terms .censatioïi and perception. surface need to be analysed as the quali-
Obviously these terms will have to be ties of color were analysed many years
either discarded or redefined. The second ago, and the first problem is to search for
hypothesis implies that perception, at variables of the retinal image with which
least of the type called t1itera1,'' is these qualities might prove to be in cor-
primarily dependent on stimulation rather respondence.
than on meaning or mental elaboration. A substitute for the distinction between
This li rotlesis contradicts the traditional sensation and perception will be offered in
conception that, whereas sensation de- Chapter 3, a substitute intended to retain
pends only on immediate stimulation, per- what is verifiable in the classical dis-
ception depends also on past stimulation, tinction and eliminate what has been
or memory. theoretically misleading. We can attend
The rejection of this distinction is either to color-impressions or to oect-
neitlìer so novel nor so radical as it may i mpress ions, generally speaking. Intro-
sound. Although a generation ago it was spection of the first sort yields an ex-
still possible to suppose that sensations perience of the visual field. Introspection
and perceptions were essentially different, of the second sort, called 'phenomenolo-
the discoveries of Gestalt psychology have gical," yields an experience of the visual
overthrown the logical basis for the dis- 14'O rid. Both these kinds of experience
tinction. The seemingly vast diiference must be accounted for if we are to under-
between a sensation and a phenomenal stand vision, but the latter is the subject
object has been slowly vanishing in recent of this book. how can we see the world
years. Instead of the doctrine that per- '' h. do things look as they do?
Theories of Perception
The Distinction between Sensation and Perception

. . . . Nativism and Empiricism . . . . Extensity
and Location . . . . Form or Shape in Two Dimeno
sions . . .e Depth and Distance . ..
The
Theory of Cues . . . . Gestalt Theory . 'e The
Fact of Perceptual Constancy . . . Summary
j The traditional explanation of vision is

that perceiving things depends on first
this evidence, and we may start by
inquiring how the distinction between
having sensations. Sensations are sup- sensation and perception arose.
posed to be the raw material of human ex-
perience and perceptions the manufac- The Distinction between Sensation and
tured product. Sensations are only colors, Perception
sounds, touches, odors, and tistes; ob- Around the latter part of the seventeenth
jects and space depend upon perception. century, the imagination of men began to
A certain hue, a feeling of warmth, and a be stirred by the theory that all human
smell of smoke are not things in them- knowledge comes through the senses and
selves. Only when they are combined in from no other source. in short, we learn
a perception do they mak?e us experience a our ideas instead of discovering them iiiì-
fire. The eyes furnish us with an array of planted in our minds by Cod. It follows,
colors, the ears with a flow of sounds. for example, that every man can acquire
That is all they can co, and the rest of his ideas for himself, and that he himself
experience is a matter of combining, is the best judge of their truth. The doc-
ordering, and uniting the sensations into trine was given a special impetus by John
things and events. The play of light with- Locke in 1690 in An Ess(ly Concerning
in the eye can give us color but not things. Ilurnaii L'nderstanding. The mind at birth
Things are a product of a mental capacity - -
is a blank page a tabula rasa on which
called perception. experience writes its record. If knowledge
This explanation has so much age and could exist in mind only by way of sense,
respectability that there is a temptation to it was obvious that the sensory capacities
forget that it is only a theory. As a matter of man needed to be carefully investigated.
of fact it is not consistent with a great Since vision was the principal sense, schol-
deal of accumulated evidence in psychol- ars began to concern therAselves with the
ogy. It would be worth while to consider optics of seeing, and to note what they
12
THEORIES OF PERCEPTION 13
themselves could see under controlled sense the external object at a distance -
conditions. But here they encountered a the object "itself." The paradox was
difficulty. The visual sense was simply that the latteris nevertheless apprehended.
not adequate to account for all visual Imere arose among philosophers a dispute,
knowledge, especial1y of three dimensional now centuries old, over whether and how
space. Either, then, some knowledge of WC can !)eIjeve in an external world. If
the world does not come through the objects with solidity and distance were
senses, or the visual sense must be sup- creations or constructions of the mind, then
pleinented in some way by the mind. it could be inferred, for example, that
There must exist a special mental process they were mental objects. Physical ob-
over and above the visual sensations: a jects either did not exist or, if they did,
process which in some way constructs the were unknowable. 1f they were neverthe-
world out of the 'raw data" presented to less known, the explanation must be
the mind. Such a process might be one of supernatural. A vast amount of intellectual
ass ociation and inference the alternative
; effort and ingenuity has been devoted to
would he a kind of intuitive understanding this type of controversy or to some means
of the data of sense which would imply a of escaping from the dilemma on which it
retreat toward the dogma of innate ideas. was founded. And the dilemma itself
This argument was a rationale for the appears to rest, in part at least, on the
theory of perception which still underlies conviction that such properties as dis-
our thinking. The nature of this special tance and solidity cannot be sensed and
mental process has puzzled sorne of the that the apprehension of them poses a
best thinkers and scientists in vcstern unique and special problem. 1f a sensory
civilization for two hundred \ears. basis for such properties could be dis-
The obvious puzzle in giving any exact covered in the retinal image, however, the
account of perception was the visual third dilemma might collapse and the whole in-
dimension. A very knotty question arose: tellectual superstructure would fall with it.
how can we apprehend the 'real world The accepted view of perception is still
as distinct from the world of sense, or, in that the percept is never completely de-
other words, the world which appeared to termined by the physical stimulus. In-
be "eterna1" as distinct from the play of stead, the percept is something essentially
light within the eye ? Various criteria of subjective in that it depends on some
the visual reality of objects were de- contribution made by the observer hirn-
scribed, such as maintaining their posi- self. Perception goes beyond the stimuli
tion despite eye movements and conform- and is superposed on sensations. The
ing with impressions of touch, but the sensations are basic and, being parts of
distance and depth of objects were their our organic equipment, tend to be the same
most obvious featares, and these it seemed for all. Perceptions, however, are sec-
impossible to explain. The 18th century ondary and, depending on the peculiari-
scholars understood that the eye can ob- ties and past experience of the individual,
tain an image of an object but cannot may vary from one observer to another.
Whenthis doctrine of perception is ap British philosophers in the eighteenth

plied to such abilities as the apprehension century and experimental psychologists
of meaningto the understanding of lan- in the nineteenth strove to explain
fluage for instanceit works very well and perception with as little appeal as
accounts for most of the experimental possible to intuition or innate ideas. Such
facts which psychologists have accurnu- theories they considered mystical and not
lated. Meanings do depend upon the past consistent with a scientific psvcho1o.
history ¿f the individu1. 13ut when it is Visual space, they were convinced, must
applied to the apprehension of material be somehow learned. On the other side,
objects and of the spatial eiironment, it many philosophers and sorne experimental
is less satisfactory. For the visual uorlds psychologists could find no satisfactory
of different observers are more alike than way of understanding how this could occur.
they ought to be if the doctrite H'ere the At least some features of visual space,
complete truth. The evidence accumulates they argued, are so immediate, simple, and
that men, and moreover even animals, clear in our consciousness that thc' nust
appear to react to the spatial environment be either intuitions which are fundamental
with an accuracy and precision too great to "mind itself" or else must he innate
for any known theory of space perception features of the sensations themselves.
to be able to explain. The fundamental The speculations and debates of these
modern difficulty is this. If the solid two groups make ur what lioring calls
visual world is a contribution of the mind, the "long and barren controversy " over
if the mind constructs the world for itself. nativisin and empiricism (12).
where do the data for this construction In order to understand what was meant
come from, and why does it agree so well by space in this controversy, it is neces-
with the environment in which we actually sarv to remember the scientific conception
move and get about? If space perception of the world which began to be current at
is a subjective process then why are we the beginning of the eighteenth centur>.
so seldom actually misled by illusory The discovery of gravity by Sii isaac
perceptions? Thy are the optical illusions Newton led him to conceive a physical
of the textbooks actually the exception universe so logical and simple that it be-
rather than the rule? carne the wonder of the age (85). The
Nativism and Empiricism facts of astronomy and physics were united
in it; these and many other facts became
The history of past attempts to account
for the process of space perception is pro-
predictable from a, few simple laws. This
tracted, involved, and difficult. Even at the physical universe consisted of three
risk of oversimplifying, however, its main things only, space, time, and matter, and
issues need to be skethed if we are to
clear the way for any novel approach to 1The writer follows the usage of Boring in
using the term empiricism as the alternative of
the problem. It is the hi'story of a con- both nativism and rationalism. "Empirism"
troversy. On the one side, a group of is the term employed by Gestalt psychologists.
from these realities everything else could similar to what has been called, in this
be deduced, Events were reducible to book, the visual world.
matter varying in space with time, i.e. in 1oth nativists and empiricists agreed
motion, and could therefore be analysed in that the visual sensations were innate.
terms of grams, centimeters, and seconds. Sensations were the data of, or what was
The space of this universe, it may be 'given" to, the mind. They disagreed
noted, was empty Euclidean space, defined over whether perception was a matter of
by the three dimensions of the Cartesian learning or of intuition. But they also
coordinates. Inevitably, then, the problem disagreed from the very beginning over
of how we can observe the world was what was sensed and what was perceived.
formulated as the problem of how we can The simplest and most logical doctrine
apprehend the Newtonian universe, and was to suppose that only color could be
by space perception the eighteenth cene sensed and that all the constituents of
tury philosophers and the nineteenth cen- space were perceived, including extensity.
tury psychologists meant geometrical This implied that a color sensation could
p e rc e p t ion. only be a spot or point of color, and that
s_p._!::!
This presupposition influenced the an area of color was the sum of these
psychologists' analysis of the problem and elementary sensations. As thus conceived,
dictated the terms in which theories could the sensations corresponded with the
be propounded. For both nativists and focused points of light in terms of which
empiricists, perceived space seemed to optics had analysed the retinal image.
divide up naturally into certain geometric This was the theory of Wundt, the most
categories. First there was extensity in consistent sensationalist. Another doc-
two dimensions: the bare characteristic of trine was to suppose that extensity was
space as being spread out. This cor- sensed (or was an "attribute" of sensa-
responded to the plane of_theverttical aidd tion) but that the location of points in the
. i
extended field was not sensed and there-
- -----
horizontal axis in geometry. Then tnere
" -------------.--
the aspect of location in two dimen- fore had to he learned by experience. As
sions, or the localization of points in the a third possibility, not only unshaped
f_
.--
iThual
Iñdycoordinates
--- -
This corresinded to the x
-
Fgeometry.4 Next there

areas but also shaped areas, or forms,
might be considered to be data of sense.
in the William James for example, although he did
visual field. This correspbnded to the not actually assert that a form was a
abstrac formsof Greek geometry. Final'ly sensation, did believe that a visual line
there was the aspect of depth br distance, was a simple datum rather than a row of
the third dimension of space, and this point sensations. As a last possibility, it
corresponded to the third dimension of might have been assumed that all con-
geometry. Extensity, location, shape, and stituents of space were sensed. But
distance these were the primary con- actually no one ever supposed that depth
stituents of visual space. They do not, it and distance were simple sensations, and
may be noted, constitute anything very thevisual third dimension was and remained
16 THE PERCEPTION OF THE ViSUAL WORLD
a phenomenon which only perception could signs" in order to account for a visual
explain. Keeping in mind these variations field. A local sign was the unique ac-
in conceiving the sensory elements with companirnent of every point in the field,
which perception had to work, let us ex determining its position in the up-down and
amine the efforts of the empiricists to right-left dimensions. Since every point
explain the constituents of space without could be separately localized, or pointed
resort to innate ideas. to by the observer, each must have its
own locality-characteristic distinguishing
Extensity and Location.
it from every other point. The ques-
Although it was logically possible to tion which divided the empiricist and
assume that a visual field filled with pure the nativist was whether this differentiating
color, such as the blue sky, was a mosaic characteristic became associated with its
of spaceless points which looked cono appropriate retinal point through exper-
tinuous because they had been associated ¡ence or had been intrinsically connected
together in past experience, this seemed with that point from birth onward. The
highly improbable. A more plausible
variations of opinion on this question need
theory was that a color sensation was ex not be described. A possible explanation
tended by its very nature: that color simply for the learning of these locality-signs, in
came that way. The blue sky, then, was a general terms, was that each point on the
simple sensation and no problem to the retina got associated with the movement of
empiricist. The commoner kind of visual the eye just necessary to bring its stimu-
field filled with patches of different color, lus to the fovea. It was practice in
however, was a different matter. This was fixating points (or locating them with the
something like a space with objects in it eyes) which made their location possible
and to this the special process of percep- when the eyes were motionless.
tion might apply. Such a field possessed
order, arrangement, or pattern as we would Form or Shape in Two Dimensions.
say today. But to the early psychologists ( To the empiricist psychologists the
it seemed that the way to start analysing erception of solid objects required two
it was not in terms of order but in terms tages of explanation: first a theory of
of location. How did the spots of different lane geometrical shapes, and second a
color get their position or place in the ex- heory of their three-dimensional character.
tended field of view? If the position of all Since the retinal image was two-dimen-
points in the field could be perceived, they sional, this seemed the most reasonable
reasoned, everything in the held could be approach, and it was reinforced by the
perceived. psychologists' tendency to see things
The space of the physicist was a space pictorially when they analysed their
of points whose position could be defined own perceptions introspectively. The
by the Cartesian coordinates. To the term shape thus came to mean primarily
psychologists, therefore, it was clearly p-rojecsed shape or, more specifically, the
necessary to develop a theory of "local projected shape as the object is commonly
THEORIES OF PERCEPTION
viewed. Form in this sense of the term But the evidence obtained cannot be inter-
can be experimented with, since it can preted as proof that at the outset they
be conveniently represented on paper. A either do or do not see shapes. The im-
b forexatnple can be drawn as a square. plication of the reactions which babies
Black lines on a white background are not, first make to faces and other visual objects
in actual fact, much like the edges and con- is that they see them as forms, but of a
tours of objects in a visual field, but to sort incomprehensible to any adult: forms
civilized vision they are good equivalents. which can only be called indeterminable
The study of drawn shapes, accordingly, or undiqerentiated from one another. These
has been pursued for centuries and it terms do not mean that vision in the infant
constituted an obvious problem for the is what adults would call blurred, or that
early psychologists. the contours and details of things appear
If sensations were points of color, then as they do in an out-of-focus photograph.
a shape must be a mosaic of such point- They can only suggest, not describe, what
sensations associated with one another the perceptions of the infant are probably
during the course of past experiences with like. There is evidence, for instance,
that particular shape. A line, for instance, that the typical baby at 3 to 5 months can
was a row of contiguous black spots. seehuman faces as Ti1y distinguished
An extended shape was an array of colored from other things but not as distinguished
points. But, as we have already noted, from one another (99). \Phe development
f
only the most radical of empiricists were of perception seems to proceed from the
explicit in believing that the sensations of seeing of gross differences to the seeing of
vision were points. The more common fine differences. Whether this development
opinion was that color possessed ex- is principally a matter of learning or
tensity as an innate attribute. The formed principally the result of the natural growth
or shaped character of a t'piece" of ex- of the optic nervous system is not now
tensity might, however, be learned even if known. In any event the learning process,
the extensity itself were not, and this is if it is that, is not like the learning of geo-
what empirically-mindedpsychologists have metry which proceeds logically from points
tended to believe up to the present day. to lines to planes and thence to solids in a
l3ut no one has yet demonstrated precisely wholly different kind of sequence.
how such learning could occur, or has even Figure 3 shows what a nine-months-old
explained just why, if extensity is ari un- baby is supposed to see when his mother
learned feature of experience, form should plays peekaboo with him. An ingenious
be a learned one. attempt has been made to suggest how the
The experimental evidence on whether visual field becomes progressively less de-
or not we have to learn to perceive forms terminate from the center to the periphery,
has proved, over the years, to be not very when the viewer fixates an obect of ¡n-
conclusive. One can study the behavior of terest, a fact as true for adults as for
infants systematically and make infer- babies. The photograph is increasingly
ences about their first visual perceptions. blurred away from the center. The baby's
ç b
41
')
(ßy Life photographer 1/erben Gehr. Copyright Time, Inc.)
FIGURE 3. An Attempt to Represent the Vision of an Infant
perception, however, may be indefinite Toward the end of the nineteenth cen-
without being optically out of focus; this tury a few psychologists began to em
the photograph fails to convey. The evi- phasize the fact that a form may be transe
dence will be discussed in Chapter 11, posed on the retina, as the observer scri
(p. 207).2 the object he is observing, without its
making any difference in the perception.
2 . . .
For a description of the year-old infant s Although the sensory elements differed, the
visual behavior, see A Gesell, F. L. hg, and
G. Bullis, Vision: Its Development in Infant form did not. Moreover, the form was the
and Child (Harp& and Brothers, 1949). same whether the color it was made of
consisted of white on black or black on line represented on the retina as a single

white. The form, they reasoned, must point. Any external point on the line of
therefore be independent of the anatomical sight would be optically the same as any
retinal points of the image and also in- other point. There was nothing to indicate
dependent of the color stimulation of the whether it was near or far, or even for
points. lt must in fact be a "form-quality" that matter outside the eye. The data for
(Gestaltqualitlit) somewhat analogous to perceiving the distance of a point must
the color qualities of hue and brightness therefore be provided by the use of two
and therefore incapable of being analysed eyes.
into sensations. Form, then, was some- Since both eyes are always aimed at an
thing that fell into n'ither the category of objective fixation-point so that there is a
p!íE!ition nor sensation; itwas irreducible clear image of it on the exact center of
and elementary like a simple sensation each retina, the distance might be known
but, unlike a sensation, it had no corn- by a sort of triangulation. The eyes might
prehensible stimulus-equivalent in the operate as a surveyor does when he, in
retinal image. What could be the stimulus effect, aims two telescopes at a distant
for a visual form? Either it must be a set object from the two ends of a fixed base
of point-stiriuli, and this was not easy to line, or as a gunner does when he operates
understand, or it was the form of these a range-finder. The visual process in the
points, and this was a mere tautology. brain would have to include a kind of
The dilemma was one whiçh, as we shjl automatic reasoning not unlike the corn-
see, the Gestalt theorists attempted to puting mechanisn of a range finder, which
resolve. can solve problems in trigonometry au-
1Ar/
Depth and Distance, Th'Tho'(,çf"Cues"
tomatically. Helmholtz called the process
inference."
on the basis of the sensatÍns óf color The sensory data for this estimation of
conceivedeither as points or in sorne vague distance could only be the eye-muscle
way as formless and sizeless extents, the sensations which accompany the con-
empiricists supposed that human beings verging or diverging of the eyes according
somehow construct a three-dimensional as near or far points are fixated; the muscle
world in perception, or, in the terms of the sensation, then, was a "criterion" or
philosophers, that we have knowledge of a "cue"3 for the estimate. This idea can
three-dimensional world. How could this be credited to Bishop Berkeley
t ' -.
who based
occur? Specifically, what information his "new theory of vision" on it in 1709.
could the eye transmit on which such per-
ception or knowledge could be based? 31n American psychology, as Boring has
Considering the problem as one of Car- pointed out (48), the words 'cue" and "clue"
have both been used to mean a kind of sense-
tesian geometry, it seemed obvious that a fact on which to base perception or behavior.
single eye could not yield any information "Clue" implies reasoning whereas "cue"
implies the touching off of sorne response, but
about the third dimension since the latter their meaning has never been clearly dis-
/
consisted of the line of sight itself, i.e. a tinguished.
20 TIlE PERCEPTION OF THE VISUAL WORLD
He added co it the idea that sensations of

accommodation (the adapting of the lens
which brings the point of fixation to a
focus) might be supplementary cues for
distance. The theory has had a long
scientific history. It is not however, as
later computations proved, adequate to
account for estimates of distance as far
away as we can actually judge distance. The
faruaccommodation of the lens approaches
a maximum limit at around fifteen feet and
the convergence of the eyes approaches a
zero limit at about fifty feet. For lack of a
better theory, however, the cuesof con
verence
_r_,_ø,t.__'o I__ accommodation
and fl "w continued to
be, and still are, given as a partial ex
planation of depth perception in the text-
books.
In 1833 a new correlate of visual depth
was discovered. In contrast with previous
theorizing based only on self-observation,
this was a truly experimental discovery.
With a theory in mind, Wheatstone in-
vented an optical device to test it, whicI
he called a stereoscope. His idea was
that the discrepancy between the two reti-
FIGURE 4. The Disparate Views
na! images of an object on which the two
of an Object by the Two Eyes
eyes converged was not simply the paradox
it had previously been considered (how ptically combined figure, then disparity
can we see two different views as the L a cause ofçlepth-perception. Every-
same thing?) but was a basis for perceiving one who has looked at stereograms knows
the object in depth. how strikingly this theory was verified.
iThe stereoscope produced a synthetic The fact of binocular image-disparity at
of the two images, and this could once became accepted and still remains
(disparity
be modified at will. The experimenter the chief explanation of how we see
could draw pairs of geometric figures, one the third dimension. Whether the dis-
for each eye, differing in various ways and parity should be thought of as a clue for
te instrument would project each utpn an interpretive perception of depth or as
its appropriate retina. If alawful relation- a kind of binocular sensation yielding
ship could be established between the depth immediately was not easy to de-
disparity and the perceived depth of the cide. It was in any event a demonstrable
correlate of the experience of dep exception of the last named, they had
like the hypothetical sensations of con- been employed for centuries by painters
vergence and accommodation, and its in their effort to reproduce a segment of the
discovery made the dogma of an innate in- world on a fiat surface. When the philo-
tuition of space of space as an inner sophers and psychologists began to ex-
-
condition of all experience less likely amine their visual sensations they in-
than ever before. evitably began to view the world pic-
In the outcome, the three classes of tonally, as artists had learned to do, and
data derived from convergence, accommoda- these rules of picturing were recognized
tion, and above all from retinal disparity as being indicators or signs of distance
were taken as the primary criteria for dis- for a retinal picture as well as for a
tance and depth and as the only discover- painted picture. But these clues could
able basis for the perception of abstract not be given the same explanatory value
three-dimensional space. There did exist, that could be ascribed to convergence and
to be sure, in any concrete visual world binocular disparity. They were them-
such as a view of the countryside or a selves perceptions, it appeared, not data
scene which a painter might choose to of sensation; even the most convinced na-
represent, a number of other clues for tivist could not argue that they were pure
detecting the distance of things. Elf one intuitions of space; they must obviously
object seems to "cover" another, it must therefore be learned by experience. They
be nearer. If edges known to be parallel were called secondary cues for depth and
seem to converge, they must really re- distance to distinguish them from the
cede; and if objects known to be of similar primary cues of convergence, accommoda-
size seem progressively smaller, they must tion and retinal disparity. Since they did
really be progressively farther away. If nbt depend on the existence of two eyes,
one thing appears above another it is
probably not suspended in the air but
., ,
they became known also as tJe monocular
CUeS, while convergence and disparity were
merely lying on the ground at a greater binocular in origin. Although they have
distance. If an object seems bluish and been described many times, re-observed
blurred it must be distant like the hills by successive generations of curious men,
on the horizon. If an object is partly in and have passed into common knowledge
light and partly in shadow its surface as facts having to do with the perception
cannot be flat but must really be curved or of space, they have never been systemati-
bent. If a thing seems to move, or be cally controlled, varied, and subjected to
displaced across other things when the experiment. In succeeding chapters we
observer moves his head from side to will have much more to say about them,
side, it must really be nearer than the and their significance may then appear in
other things in proportion to its relative a new light.
inotion.JAU these clues had been known The theory of cues as the explanation
long before the perception of distance ever of our perception of the world has proved,
became a philosophical issue. With the in the eighty years since Helmhol
--w per-
fected it, more convincing than the alter.. it does, but only how we can make judg-
native theories. Complicated as it is, it ments about the world. Both of these
has seemed, to Americans 't least, to be objections were raised some twenty-five
the only scientific exp1antiori, for it years ago by the Gestalt psychologists.
did keep operi the possibility' of investiga» The Gestalt theory started with the
tion whereas any appeal to intuition problem of how we can see visual form.
rendered experimentation impossible. It Instead of simply adding a "form-quality"
assumed that sensing and knowing were to the list of sensations, however, it took
two different things and that all knowledge a new line of thought and asserted that a
came through sense. Many phrases of form not compounded of sensations at
commonsense psychology are reflections all. Experience is not reducible to ele-
of this assumption; the "messages" of ments or additive units, the argument went,
the sense organs, or the "information" or and when it is analysed introspectively
"facts" that they supply to the mind, into sensory components it is falsified.
imply a set of clues and a process of in- But if not constituted of sensations, how
terpretation. The_mind, it was assumed, is a unitary perception of this sort to be
intelligent and acts on the sensa- accounted for? That there had to be a
tons somewhat as a geometer or a logi- special perceptual process of some sort,
cianwould act, combining, computing, and the Gestalt psychologists never doubted.
comprehending the data it gets in much the Observing that under experimental condi-
same way as did the philosophers them- tions visual patterns or dimly seen forms
selves when they invented the theory. tended to be perceived as symmetrical,
connected, coripleted, and meaningful,
Gestalt Theory
even though the drawings presented to the
The theory that sensations were data or observer were not, they concluded that
cues for perception lasted a long time, these tendencies were laws of the percep-
but it had troublesome implications. For tuai process in general and were indicative
one thing, unless perception were purely of its nature. Forms seemed to occur
intuitive, it had to be a kind of compound- spontaneously in perception even when the
ing or putting together of elementary sens a- picture constructed by the experimenter
tionS by means of associative learning. was objectively incoherent and meaning-
But these sensory elements could never less. The theory of perception which oc-
be specified. They could hardly be points curred to them was that the process was
of color corresponding to the single spots one of relatively spontaneous sensory
of excitation on the retina since, after all, organization. The process of organiza-
points are nothing but geometric fictions; tion w assumed to occur in the bran,
at the same time no one could dis'. presumably at the level of the cerebral
cçver how they could plausibly be any- cortex. It was conceived as a process in
'thig else. Furthermóre, the theory a field, analogous to the visual field
àf cues could never really explain how itself, and the parts of the field (the con-
we see the world, or why it looks the way tour of the form and its background) were
united or separated by forces of attraction is a contour? Why does the world look
and repulsion similar to electro-magnetìc upright? how is the phenomenal ego
forces. Aerceived form in this theory, located in it? Why do things appear to
i a brain-form. he retinal image yields have very nearly their true size and color
isated single excitations. Only when despite the variations in their retinal
these are projected on the cortex do the images? These were questions about
field-forces begin to operate among them phenomena of a wholly different kind
and only then do they unite in a Gestalt. from the geometrical points and lines of
The causes of sensory organization are the nineteenth century psychologists, and
to be sought in what is sometimes called these were the questions which the
iield-theory. Gestalt psychologists asked. They were
fhe Gestalt theory is not as explicit questions about the characteristics of the
about the perception of space as it is visual worl(/. The only difficulty is
about the perception of form. But it was whether the hypothetical process of
based on a description of what the world sensors' organization yields the answers
looks like, not what it ought to be like to them.
geometrically, and it therefore asserted
that all visual perception is tridimensional The Fact of Perceptual Constancy.Q,./
from the outset. The theory of perception The trend of thought which the Gestalt
as organization led to the following reason- psychologists represented was respcnsib1e
ing. The brain is a three-dimensional for more than a new theory of perception;
rgan and the neural process of dynamical it resulted in a massive amount of experi-
organization must therefore occur in a mental evidence. An important part of this
three-dimensional field. 'I'he perception concerned the problem of what was called
itself, then, would naturally be three- prceptual "constancy." fly this term was
dimensional if the underlying ph siological meant the fact that perceptions, or pheno-
events were. The reader may or ma not menai objects, kept their identity and their
lind this argument convincing. In an objective size, shape, and color despite
event, this was about as far as theGestalt variations in the retinal images with which
theory could go with space, except for they corresponded. Although the retinal
Koffka's analysis of the h>pothctical image was a poor indicator of objective
field forces which might underlie binocular shape (so it seemed, inasmuch as it
retinal disparity (67). changed from one aspect to another as
Perhaps the greatest contribution of the the observer moved) the perception was
;estalt theorists was that, having taken nevertheless in good agreement with the
an unprejudiced look at the visual world objective shape. In short, it tended to
they were trying to explain, they foriru- refflain COflStUflt.
lated problems for space-perception This kind of fact could be tested
which were genuinely relevant. How by experiment and measured; moreover, it
is a figure separated in perception from was meaningful in terms of human behavior
its background? What is a surface? What and it escaped from the atmosphere of
respectable unreality which clung to the Summary

nineteenth century problems of space If everything we are aware of comes
perception. The interests of twentieth through stimulation of our sense organs,
century psychologists began to shift and if some things nevertheless have no
toward the question of why perception counterparts in stimulation, it is necessary
was objective and away from the purely to assume that the latter are in some way
theoretical aspects of this paradox. synthesized. How this synthesis occurs
If the objectivity of perception could is the problem of perception. Our aware-
be studied in the laboratory it was no ness of the world of objects and space is
longer a speculative question of epistemol- particularly difficult ¡o account for but
ogy but a matter for experimental investi- also particularly important, since it per-
gation. In a size-constancy experiment, meates nearly all kinds of experience.
for example, an observer is required to Theories of the perception of objects and
equate the size of a wooden stick near at space, therefore, have a long history.
hand with that of a stick placed at a Nativisn: assumed that the synthesis was
distance. In general this task can be intuitive or innate. Empiricism explained
performed with some accuracy. The size the synthesis as learned or inferred from
of the retinal image of the far stick, how- past experience. More recently, Gestalt
ever, may be only a quarter or an eighth theory has suggested that it is produced
that of the near stick, the rule of optics by a characteristic achievement of the
being that doubling the distance of the central nervous system which may be
object halves the projected size of the termed sensory organization.
image. Since the impression of size The difficulty in postulating a consistent
obviously does not depend on the image, learning theory is that many kinds of per-
on what can it depend? A reasonable an- ception seem to occur in children and ani-
swer would be that it depends on the whole mais who have had no opportunity to learn.
stimulating situation or, more specifically, Sensory organization, as a descriptive
on the stick-image in relation to its back- term, appears to fit these facts somewhat
ground-image of three-dimensional space. better. If it is necessary to assume some
It is only a step from this kind of kind of synthesis of visual stimuli,
reasoning to the proposal made in the organization" is a better word to use
next chapter: that there eiist, as extremes, than "reasoning" or t'inference." As a
two kinds of seeing, (1) the experi- theory of what might go on in the nervous
ence of a visual world in which objects system, however, 'organization" is less
stay the same size wherever they are and valuable. It is true that physical and bio-
in which parallel edges do not converge, logical processes are often characterized
and (2) the experience of a visual field in by organization (the tendency of electric
which the principles of perspective hold circuits to reach an equilibrium and the
true. Constancy of size would then be a subordination of parts of an organism to
corollary of the visible depth and dis- the whole during the growth of the embryo)
tance of the visual world. hut when this concept is applied to the
physiology of visual perception it has a between retinal stimulation and our aware-
fundamental weakness: it does not in ness of things. Koffka, in his Principles
itself explain why a perception is like its of Gestalt Psychology, spoke of a "more
object. The characteristic of perception is comprehensive correspondence between
that the result is not so much spontaneous the total perceptual field and the total
as it is faithful to the thing perceived. stimulation" (67, p. 96) and implied that
The question is not how a percept gets this correspondence would be clarified
organized but why it is always organized when the laws of sensory organization
like the particular entity toward which were known. What this book attempts is
the eye happens to be pointing. a direct explanation of this comprehensive
The Gestalt psychologists made much of correspondence. If the total stimulation
the spontaneous character of the process of contains all that is needed to account
perception, but they were aware of the for visual perception, the hypothesis of
problem of some kind of correspondence sensory organization is unnecessary.
The Vsua1 Field and the Visual World
The Bounded Visual Field . . . . The Gradient of

Clarity . . . . The Effect of Eye and Head Move.
ments . . s The Location of After.lmages . . .
The Apparent Size and Distance of After.lmages
s s e The Effect of the Posture of the Head and
Body . . s The Apparent Size and Shape of Ob.
jects . . . . The Apparent Convergence of Parallel
Lines . s The "Eclipsing" of Forms . . .
The Visual Field during Movement of the Observer
.s . . The Awareness of Distance . . . . Sum.
mary . . . a The Problem of the Visual World
If we are to understand the problem of and walls, with an array of familiar ob-
why the visual world looks as it does the jects at definite locations and distances.
first thing to do is to look at it. What ac Every part of it is fixed relative to every
tua11 does it look like? This question other part. If you look out the window,
is not as easy as it sounds. lt requires there beyond is an extended environment of
that we carefully examine our experience ground and buildings or, if you are lucky,
and then find the essential terms in which "scenery". This is what we shall call
to describe it. The description needs to the visual world. lt is the familiar,
be carried out without preconceptions and ordinary scene of daily life, in which
without reference to theories as to how solid objects look solid, square object-s
vision might occur. The known facts of look square, horizontal surfaces look
vision may be kept in mind, but the known horizontal, and the book across the room
theories and their implicit terms should be looks as big as the book lying in front of
disregarded. The problem is to state with- you. This is the kind of experience we
out any theoretical prejudgment what we are trying to account for.
see when we say that we perceive the en- Next look at the room not as a room but,
vironment. insofar as you can, as if it consisted of
Try making this observation for yourself. areas or patches of colored surface, di
First look around the room and note that vided up by contours. [To do so, you must
you see a perfectly stable scene of floor fixate your eyes on some prominent point
26
THE VISUAL FIELD AND THE VISUAL WORLD 27
and then pay attention not to that point, your attention on the periphery of the field
as is natural, but to the whole range of (a trick that may require practice) you
what you can see, keeping your eyes still can observe that things are visible only
fixed. The attitude you should take is to a limited angle out to the right and left
that of the perspective draftsman. It and to an even more limited angle up-
may help if yo-u close one eye. If you wards and downwards. fthese boundaries
persist, the scene comes to approximate it is true, are not sharp 'lTkè the margins of
the appearance of a picture. You may a picture and they are hard to notice, since
observe that it has characteristics some- all vision is unclear in such eccentric
what different from the former scene. regions, but they are nevertheless present.
This is what will here be called the The field is roughly oval in shaPd) WTr
visual.- -field. It is less familiar than the measured, it extends about 180 degrees
. - ' '-. -. ..-
visual world andit cannot be observed laterally and 150 degrees up and down.
except with some kind of special effort. If you close one eye you will notice that
The fact that it differs from the familiar about a third of the field on that side
visual world is the source of a great deal disappears and also that the boundary is
of confusion and misunderstanding about now the outline of your nose. N1any an
vision. It is the experience on which the otherwise observant individual' does not
doctrine of visual sensations is based. realize that his nose is represented in his
lt is strictly an introspective or analytic visual field. Even if shadowy, however,
phenomenon. One gets it only by trying to it has always been there and its dis-
see the visual world in perspective and to covery only illustrates the unfamiliarity of
see its coiors as a painter does.i this kind of seeing as compared with the
Both the visual world and the visual familiar reality of ordinary perception.
field are products of the familiar but still What Ernst Mach, analyzing his serisa-
mysterious process known as seeing. Both tioris, called the phenomenal ego is illus-
depend upon light stimulation and upon a trated in Figure 5. It is a literal represen-
properly functioning eye. But the differ- tatjon of his visual field, with his right
ences between them are so treat as to eye closed, as he reclined in a nineteenth
suggest two kinds of seeing. Let us try century chaise longue. His nose delimits
to list and describe these differences. the field on the right and his moustache
Most of them can readily be observed appears below. I-fis body and the room
without special apparatus, and the reader are drawn in detail, although he could
should therefore check them for himself as not see them in detail without moving his
we go along. eye. The margins of the field are shown
as definite and clear whereas of course
: .: their actual appearance was very vague.
The Binded Visual Field The point of fixation cannot be shown in
t, ) In the first place, the visual field has the drawing; actually it is the center of
boundaries, whereas the visual world has the field and this should be the only part
none.' If you keep your eyes. fixed but put shown as wholly clear.
I p
. -=-- I i -
i
I I I I rl II II I I t I 11111 1111111 II Iiflhi I I I I II IS I fl I II U III II l I I l I I
4
4
i,''I
/
JL,:
. 0
.IJjØD
FiGURE 5. The Monocular Visual Field of Ernst Mach
J
The visual world, on the other hand, is space behind us is a strictly visual world
certainly not delimited by an ovalshaped or not is a question of definition rather
boundary. Floors, walls, and terrain are than a matter of ordinary observation. It
visibly continuous. As Koffka has pointed cannot be answered by inspection for the
out (67), one is ordinarily aware of a world reason that in the effort ro examine the
which extends backward behind the head experienced world one finds oneself in-
as well as forward in front of the eyes. specting the visual field instead. The
The world, in other words, surrounds us visual world, as we shall discover, will
for the full 360°, in contrast to the visual not bear up under much introspection and
iä*
;l;::1s confined to about 180°.
Whether the world which includes this
analysis without changing its character.
It is at least clear that the visual world
THE ViSUAL FIELD AND THE VISUAL WORLD 29
does not have boundaries. It has a does not have boundaries. The world is
panoramic character which the field does ordinarily perceived by scanning, that is,
not possess.' by moving the eyes rapidly from point to
point, and the objects and surfaces which
The Gradient of Clarity
compose it are always clear and fully de-
A second characteristic of the visual tailed. If the objection be advanced that
fleldis that it is sharp, clear, and fully they are in fact only clear and detailed
detailed at the center, but progressively u'/ien fixated, the answer ¡s that the ob-
vaguer and less detailed toward its jector gets this fact from an inspection of
boundaries. For instance, the contours his visual field, not his visual world.
and patterns of the array of surfaces in
your field can be observed to become The Effect of Eye and Head Movements
gradually less determinate as you attend The visual field shifts whenever the
to those out toward the periphery. So dif- eyes are moved from one fixation point to
ficult are the latter to see that the impulse another, since the eyes normally play
to turn the eyes and fixate them may seem over the visual environment in much the
almost irresistible.2 If you move your same way that a searchlight moves over
eyes down this page of print, for example, a night sky except that light is being
and fixate at random one letter of a single absorbed by them instead of emitted.
word, you will probably find that you can Scanning movements of this sort are termed
perceive that word and the words adjacent saccadic eye movements, and are rapid
to it on the right and left and above and jerks of very brief duration. If the shifts
below, but no more. '1he visual field, of fixation are wide the head also moves
therefore, possesses a central-to-pen- in the same direction as the eyes and, as
pheal gradient of clarity. The visual a result, the boundaries of the visual
world does not. lt does not even have a field formed by the eyelids and nose sweep
center, which agrees with the fact that it across the array of colored patches. If
1To the reader familiar with Koffka's dis- by many writers, or the kind of field conceived
tinction between the "behavioral world" and by what is called "field-theory." These latter
the "geographical world" (67), it should now usages fail to distinguish between the de-
be clear that his is a quite different distinction limited and the panoramic kind of experience.
from the one now being made. The "behavioral The distinction here being made is, however,
world" for Koffka was the whole field of visual similar to Brunswik's conception of two kinds
experience. The point of this chapter is that of perceptual achievements, the seeing of
visual experience needs to be subdivided into "perspectives" and the seeing of "constants"
a bounded or field-like kind of experience and (15).
an unbounded or world-like kind of experience.
The "geographical world" was Koffka's name
for the physical environment. That there is a 2.i hecenter of clear perception corresponds,
physical environment, neither Koffka nor the of course, to the fovea of the eye - that area
writer nor, presumably, the reader doubts. of the retina best equipped anatomically for
It should also be clear that the visual field discrìmination of fine detail and on which is
as here defined is not the same thing as the projected an image of the object toward which
"phenomenal field" as this term is employed the eye ¡s pointed.
FiGURE 6. The Types of Eye Movement
I
me ..
Il primary position
A saccadic Convergence
of the eyes in the head eye-movement of the eyes
I
I
I
I
I
I I
I I
I i
I
I
Compensatory movement of the eyes ¡n the

A pursuit head (the result of either a voluntary turning
eye.movement of the head ora passive rotation of the head)
you will stand in the middle of the room, world as a whole is not stable but moves
close one eye, and turn around or walk in back and forth. In both these situations
circles, you can observe the way in which the same thing has occurred - a dis-
these boundaries sweep over the walls as placement of the retinal image across the
your head turns. -
retina proper bu t there has been quite a
,øo e of the most obvious characteristics different result in perception. This result
of the visual world is its stabi1ity."Ìhe must be due to the difference between the
r1d does not rotate as you turn around two kinds of eye movement, natural and
(you would become badly disoriented if it artificial.
did) nor does it shoot from side to side or
During the natural eye movements of
up and down as you shift your fixation
scanning, the visual world and even the
from one object to another. This fact is
colored surfaces of the visual field appear
so obvious that most of us take it as a
not to move. But there is another type of
matter of course and do not realize that
natural eye movement, the pursuit move-
there is any need for explanation. And
ment, in which it makes a difference
yet it is really a very astonishing fact.
whether the world or the field is attended
Things possess a direction-from-here
to. Hold a pencil in front of your eyes,
not with respect to the margins of the
fixate it, and move it slowly from right to
visual field but with respect to a fixed
left. Looking at the situation as objec-
visual world - an external frame of refer-
tively as possible, the motion that you see
ence which seems unexplainable on the tends to be concentrated in the pencil
basis of the retinal picture. Try the fol- rather than in the world behind it. But
lowing experiment: with one eye closed, if you now continue to fixate the pencil
select some prominent object and then look but attend to the background, the motion of
alternately toward a point just to the right the latter becomes more obvious. Seen
of it and another point just to the left of ìt. pictorially, or as a field, the illusion of a
The object will not seem to move. Try as moving environment is fairly compelling.
you will to see it as a patch of color which
goes shooting from the right to the left The Location of After.Imoges
side of your visual field, you will probably Another way of demonstrating the direc-
have only indifferent success. You may tional stability of the visual world despite
be able to see it as displaced from one movements of the eyes, and at the same
side to the other of your field, if you con- time showing that there is another direc-
centrate on the boundaries, but you will tional system for vision with respect to
not see motion. Next, fixate the object the eyes themselves, is to observe the
and put your finger at the outer corner of location of after-images. Nearly everyone
your open eye so that you can feel the eye- has seen negative after-images and noted
ball under the lid. Press on the eyeball that they behave like "spots before the
just enough to move it and release it eyes" or other so-called entoptic pheno-
alternately. This time you will see the mena. Such phenomena are forms of
object move unmistakably. The visual localized retinal stimulation but, since
they are not projected by light from out- tance of that surface. In this respect,
side the eye, they are not displaced on therefore, after-images do have a certain
the retina when the eye moves. For the kind of location in the visual world. They
same reason they do not disappear when seem to attach themselves to surfaces if
the eyes are closed. When the eyes are there are any surfaces present.
open, they appear to be superposed on This fact has avery interesting corollary,
the objects of the visual world but flot to whichwill have a special significance when
be objects themselves, and they have a we come to consider the perceived size of
filmy insubstantial look. objects in the visual world. The apparent
The fact is that after-images, unlike size of an after-image becomes greater
objects, do jump about when we scan the when one fixates a more distant surface.
environment Their direction-from-here The seen size is very nearly proportional
is given with reference to the center of to the seen distance, a relationship known
clear vision and the boundaries of the field. as Emmert's Law. It suggests that the im-
You cannot "look away from" an intense pression of size must be closely linked
negative after-image, and it will reappear to the impression of distance for, of
wherever you fix your eyes. course, the size of the after-excitation on
After-images, therefore, are localized the retina of the eye does not change.
with reference to the visual field. Insofar If, on the other hand, one observes an
as they have a visible location, it is in this after-image against one's closed eyes, or
field. Objects, on the other hand, are in absolute darkness, or against the cloud-
located in the visual world, which possess- less sky, it seems to float in what might
es its own independent directional system. be called an indefinite space. It does flot
Hence, if one attends to the visual field in seem to have any precise distance and
f
the intervals between movements of the lik$rise no precise size.
eyes an object(as a patch of color) appears -r'he Effect of the Posture of the Head and
to be displaced, whereas if what you are Body
attending to is the visual world the after-
There is still another effect of the ob-
image appears to be displaced.
server's movement on his visual field
Apparent Size and Distance of After' which does not hold true for his visual
IThe
Images world. If you tilt your head 900, or lie
After-images are localized in the visual down on your side, the patchwork within
field, not in the visual world, with respect your field rotates and you may be able to
to up or down and right or left. How are see the physically vertical lines of the
they localized with respect to distance? room as possessing a kind of horizontal
How far away do they look? As everyone quality. They now extend from right to
Icngws who has observed an after-image left instead of up and down. Considering
with his eyes open, it appears to Ese the room objectively from this position,
superposed on whatever surface one hap- however, it is obvious that the room is
pens to be looking at and to be at the dis- still upright and that the lines where the
walls meet are still aligned with the up ear. ¡t serves to keep the visual world
and down of gravity. upright and aligned with gravity. But
The room as a picture may appear as if there are also other systems or frames of
it had been tilted over on its side, but the reference, linked to the boundaries of the
room as a room is still upright. The pic- field of view, or to the axes of the head
tonal direction "downwards" goes one way or body, or to the lines of the visual field,
but the obective direction goes quite which may be in conflict with the physical
another way, and the former has the axes and which vou1d then give a visual
quality of being illusory. The physical field not aligned with gravity. Usually,
vertical of gravity, we may conclude, is but not in all conditions, such a field has
somehow implicit in all such tilted visual an illusory quality.
fields (when they result from a voluntarily
tilted or reclining posture) and it is never J1 Apparent Size and Shape of Objects
quite lost. Consequently, no matter how now come to the differences between
We
we liç or sit, the visual meaning of how the visual world and the visual field with
to stand 'tup" can always be depended respect to depth. These differences are
on. The direction of "up" in the visual not so easy to observe as some of those
world is aligned with the direction of already described, but they are more im-
gravity (42). portant for our central problem. The field
There are, it is true, a number of situa- has been said to have a tpictorial
tions ìn which this sense of the gravita- quality. picture is something that can
A
tional vertical for the visual world is be defined by mathematics and optics. The
temporarily lost, and there are diseases - essential physical fact about a picture is
usually of the organs of equilibrium in the that it consists of a projection of objects
inner ear - in rhjch it is permanently im- 'ìn three dimensions on a plane of two
paired. in some flying maneuvers, in dimensions. Insofar as the field of view
amusement park devices, in a special type can be seen as a picture, therefore, it will
of vertigo, and in a number of experimental have the characteristics of a projection.
situations (42) the visual world and the Keeping this fact in mind, let us compare
visual field cannot be distinguished from the appearance of the visual field with
one another and some illusory frame of the visual world.
reference - a non'.gravitational vertical - In one sense of the words "to see",
may then dominate perception. The ex- objects are seen to decrease in size as
perience is disconcerting and unpleasant. they become more distant. In another
It is in these situations that one loses sense, however, they remain constant in
e quilibrium. size, whatever their distance. There are
In the activities of ordinary behavior we transitional stages of seeing between
may infer that there is a visual vertical- these two extremes, stages which depend
aQd-horìzontal frame of reference which is on the conditions of observation as well
linked to gravity and is presumably medi- as upon the attitude of the observer, but
ated by the muscle sense and the inner the fact is that constancy of size tends
FIGURE 7. An Object Prolected on o Plane
to be preserved under the natural condi- the shape which an object possesses when
tions of attending to the visual world. projected on a plane. This is its shape as
Under these conditions an object is seen a silhouette, or the shape which is de
at a visibly determinate distance. fined by the outlines or contour. This is
The same thing is true of the shape of its "projected shape." That shape of an
objects. Whatever the orientation of an object which remains constant from what-
object to the line of regard, whether we ever direction it is viewed is its depth
see it from the front, the side, or the top, shape. That shape whìch changes with
if the conditions for observation are ade the angle of view -
the t'aspect" of the
quate, it will have the same shape. Now -
object as we say is its projected shape.
there are two meanings for the word shape. lt is obviously important to specify which
In this context, we mean the shape which of these meanings is being employed when
an object possesses in three dimensions one talks about shape, and a good deal of
*nd which is defined by its surfaces. We confusion has resulted from not doing so.
ihaI1 call this its "depth shape." There The visual world contains depth shapes,
is also a more common meaning of the term, whereas the visual field contains projected
shapes. As you walk about in a room you dimensional shape in the visual world.
can, first of all, observe that objects do
not change shape in the first sense of the The Apparent Convergence of Parallel
term, and secondly, you may be able to Lines
not,,e that the projected shapes do change, All of us have observed the fact of linear
especially if you fixate an object as you perspective at one time or anQther - the
walk. The only kind of an object whose fact that equidistant edges of man.made
projected shape would not change in such structures appear to get closer together,
circumstances would be a perfect sphere. after a fashion, as they recede in the dis-
It so happens that most of the controlled tance. When looking down a highway or
observations of this phenomenon carried railroad tracks the effect is strong; when
out by psychologists have been made with looking at a building or observing the in-
flat objects (whose depth shapes are ap- tenor of a room it is less obvious and may
proximately the same as their projected be difficult to note. Even in the case of
shapes when the latter are seen from the railroad tracks, however, two observ-
directly in front). Such objects, unlike ers may differ in describing what they see.
most, have a unique orientation in which One will report that the parallel lines
they are best viewed. The example fre- definitely converge as they go off toward
quently given is a dinner plate. When the the horizon; another will insist that the
conclusion is reached that the shape" rails do not converge since they are
of such a stimulus object tends to remain visibly equidistant. Each scene is per-
constant no matter what its angle to the fectly clear to each observer, but they are
line of regard, it is not clear whether the contradictory to each other. Now this fact
observer means its depth shape or its does not- in the least prove that each ob-
projected shape viewed head on. Iii the server creates the visual scenein his own
case of the dinner plate the former is a fashion and that we all have private
solid disk, bent into a rim around the worlds. It suggests only that there may
edge, arid it is perceived as such in any be two kinds of seeing. Perhaps both ob-
orientation. The latter is an abstract servers are correct, but are simply using
geometrical circle - a special kind of the verb ettO see" with different meanings.
projected shape. Strictly speaking, it is If you lay a sheet of paper on a table in
the former that remains constant. When front of you and then look at its right
you simply ask an observer what the and left hand edges, you will probably not
"apparent" shape of the dinner plate is, be able to see them as converging. Close
without specifying that you mean its one eye and try it again. Unless you have
depth shape, there is room for argument as been trained to visualize things in per-
to whether the shape is a circle or some spective, the sides of the paper will still
kind of an ellipse. The conventional tend to remain stubbornly parallel. But if
statement that a dinner plate always looks you take a pencil in either hand and, with
circular is inexact. What it always does one eye still closed, hold them perpendi-
look like is a dinner plate - the three cular to your line of sight and then align
J
f
i
--
*k ?4rtr
(By Life photographer herbert Gehr. Copyright Time, ¡nc.
FIGURE 8. Convergence of Parallel Lines to o Vanishing Point
them with the edges of the paper, you will You may now have a clearer conception
be surprised to see how much the edges of the visual field as approximating a plane
do converge when projected on an imaginary projection. On such a projection parallel
plane in front of you. Still holding the lines do meet not at "infinity" but at
pencils in position, try to visualize the eye-level, if they are parallel to the ground.
lines of the pencils projecting upward On the other hand it should be clear that
until they intersect. They meet at a point Euclid was also correct in his postulate
exactly at eye level, that is, on the hori- that parallel lines do not meet. Euclid's
zon. 1f you note where this point ¡s super proposition applied to the visual world.
posed on the nearest wall, you can see The observer who saw the railroad tracks
that it is where the wall would be cut by as continuously equidistant was aware of
the horizon of the terrain outside. the environment as a Euclidean scene, not
36
,.-
.
(j
4'
--
---
- -
- '-,*-
.
,ç "-.-
t-'
(From W. R. Ware, Modern Perspective (IV. Y., Macmillan, 1900)
FIGURE 9. Two Scenes in Perspective
37
as a scene in perspective. He saw it as not attend to the spaces between the

a locomotive engineer would, not as a objects the gaps or background and
painter. And if the rails appear even we are almost unaware of their existence.
slightly convergent to an engineer, it is But a little attention to the visual field
time to apply the emergency brakes. shows that these interspaces are just as
truly parts of it as the areas representing
\/)flie
"Eclipsing" of Forms objects. In the field as a projection, the
Whenwe are attending to the visual background is not different from the objects
world and our eyes move over the environ in the compelling way it is when you ob-
ment, the points of fixation are the obects serve the world. The interspaces, like
in it. These are the elements which arouse the objects, are areas of color, and the
our interest and affect our behavior. We do field therefore approximates the appear-
V/'FIGURE lo. Inattention to Interspaces in Ordinary Perception

Con you see two pencils in the photograph? The hidden pencil has on uninterrupted con-
tour, and you might suppose, therefore, that ¡t would be easier to see than the visible
pencil. Its contour however is nearly all "used up" by the two pamphlets, and the pencil
becomes port of the background, merely an "interspace." (The photograph was devised b
Ii,. Metzger and reproduced in Gesetze des Sehens. Frankfurt am Main: Kramer, 1936)
t s'
i ',
'
''"
''
'
,',,
5
'
' '
' ' "i'
,"c'X- ' ' '
" '
' "
; -
j2 , ,
:
s'
'X " s
'
'''a ' ,
ance of an irregular patchwork. Con sider visual field and to the visual world. In
able study has been devoted to the conse the field, the area corresponding to one
quences of this phenomenon by Gestalt object may be diminished by an area
psychologists, beginning with the work of corresponding to another object which
Rubin on what he called figure and ground lies in front of it. Seen as a field, with
(91). the head and eyes fixed, one area can be
Natural visual scenes, however, do not described as eclipsing the other, to use an
divide up neatly into figures and back- astronomical term. Seen as a world, how-
ground. In most of them it is a relative ever, one object lies in front of another.
matter whether a given area be regarded In Figure 11, for instance, some areas
as a figure or as a background. One ob- appear to be in front of others; some do
ject may be the background for another not appear superposed at all; and in some
nearer object, and another larger object a slight change in the common contour
may be the background for the first. The reverses the suggestion of depth. A pos-
largest of all "objects" - the object sible explanation of this will be given at
which is literally fundamental to the the end of Chapter 7.
perception of space and the most corn- Presumably there are transitional stages
prehensive of backgrounds, as we shall between the extreme cases of adjacent a-
try to show later on - is the terrain. reas and superimposed areas, and a number
Consider now how this phenomenon of of factors play a part in determining how
relative backgrounds is related to the these will be seen. It is clear that even
FIGURE 1. One Object in Front of Another
projected shapes like geometrical forms of locomotion he is at least moving his

tend to be seen in superposition under the head from time to time as he changes his
influence of these factors. In a visual posture. To remain motionless for any
world o the sort provided by rooms, length of time is a difficult and unnatural
streets, and countryside, the actual fact is achievement. How does this influence
that we see one object behind another. visual perception?
Koffka once argued convincingly that there
was truth in the statement that he could Every movement of the head produces a
see his desk-top extending uninterrupted deformation of the visual field. This
effect is not a sweeping shift such as oc-
beneath the book which lay upon it (67),
although this statement seemed to violate curs when the eyes alone move, but is
accepted principles of vision. If the state- rather a change in the pattern of projected
ment seems dubious, the reader may try it shapes, somewhat analogous to the shifts
for himself. and distortions of one's image in amuse-
An illustration of how both kinds of ment-park mirrors. If you fixate a nearby
seeing may be obtained from the same pbject with one eye and move your head
stimulus situation is provided in Figure from side to side you can observe the
12. You see in the picture a crowd of way in which the edges in your field move
individuals, each anatomically complete. across the surfaces behind them. The
These objects, however, as projected superposition of one object on another is
shapes are rather thoroughly eclipsed, unmistakable. If you stand up and walk
from side to side, the projected shapes of
how much so you may judge by turning the
picture upside down and fixating it. What objects are transformed, as we noted
you now see is a fairly good example of a earlier. These are actually only incom-
Visual field. plete descriptions of a much more general
phenomenon which we will discuss later
Figure 13 exemplifies the same thing. (Chapter 7). But they serve to suggest
The surfaces in the perception appear to that the visual field is ordinarily alive
slant, recede, and lie behind one another with motion. This motion is not the
in a space of three dimensions, although absolute displacement which goes with
the patchwork of light and dark areas is eye movements (that is generally invisible
wo-dimensionaI. in any case), but the kind of relative dis-
placement which goes with head move-
The VjuaI Field during Movement of the meats.
Observer. It is hardly necessary to point out that
It has been emphasized that in the or- the visual world is not distorted in any
dinary vision of everyday life any long such fashion as this when we move about
continued fixation of the eyes is a rarity. in the environment. We have already noted
It is equally rare to perceive the en- that objects remain constant despite
vironment with the head motionless. If changes in the observer's viewing posi
the observer is not involved in some kind tion. It now becomes evident that visual
4W
FIGURE U. Objects Seen Behind Others
çç
1
ç
',
h
g
l j ,
iV
tk' i. i
,4
-
'
j
* ,
:
A
j
.
ç 44' A A
- A
%
.
.- A 9 A
4
A
+ , A
A * ç
A
, . A
' pi%
4 -vj \ A
fr, ,
A
ç'A
FIGURE 13. Surfaces in Three Dimensions
space in general remains equally constant tonal seeing, then, differs astonishingly
when we move about. If it did not the from ordinary objective seeing. The field
driver of an automobile would face a very is bounded whereas the world is not.
strange situation. The field can change in its direction-from-
here but the world does not. The field is
oriented with reference to its margins, the
The Awareness of Distance
world with reference to gravity. The field
One more characteristic of the visual is a scene in perspective while the world
field should be noted. lt is never flat, like is Euclidean. Objects in the world have
a surface on which a picture is painted or depth-shape and are seen behind one
projected; that is, it is never wholly another while the forms in the field ap-
depthless. Nor is it lacking in the charac- proximate being depthless. In the field,
ter of being outside of us, in externality. these shapes are deformed during locomo-
Nevertheless, it bas less of these quali- tion, as is the whole field itself, whereas
ties than the visual world. The depth of in the world everything remains constant
the visual world is ordinarily just as and it is the observer who moves.
visible as its breadth and its height. But It has the ring of familiarity to say that
you can reduce the depth somewhat by the field is sensed whereas the world is
closing one eye. You can reduce it still perceived. These terms, however, imply
more by fixating a point and maintaining the traditional theory examined in the last
prolonged fixation. It is lessened further chapter. It is also plausible to say that
if you then attend not to that point but to although the visual field is seen the visual
the hazy margins of the field and the pat- world is only known. But this also in-
tern of shapes there. It is also reduced by volves a doctrine of perception which is
tricks such as looking at the environment debatable. The aim of this chapter is to
under your arms or between your legs, so describe the facts, not to explain them.
as to invert the field. The impression of Descriptively, the visual field always
distance never quite vanishes, but the seems a little illusory. There is always
facts suggest that you might be able to the sense that one can bring back the world
see a depthless field if you had enough whenever one wishes. There can surely
practice. Clear and indubitable distance be agreement that the visual world is mar-
is a characteristic only of the visual velously well adapted to be the conscious
world. accompaniment of behavior, while the field
is not. If we adjusted our actions to some
Summary
of the peculiarities of the visual field, we
What we have called the pictorial quality should go badly astray; thuswhen, because
of the visual field has now been described of fog or darkness, the environment is not
in a number of ways. The field differs from seen as a visual world but only as some
a literal picture in some very important kind of a vague visual field, we proceed
respects, of course, but the term will cautiously.
serve for purposes of description. Pic- The reader who is acquainted with
THE VISUAL FtELD AND THE VISUAL WORLD 43
psychological theory will realize that traditional lines. The theories of vision,
the distinction between the visual field generally speaking, have been theories of
and the visual world is a substitute for the visual field, but this type of explana-
the traditional distinction between visual tion is insufficient. What is required is a
sensation and visual perception. Is it not theory of objective seeing.
possible to relinquish the latter distinction The conception of a clear and accurate
with all its theoretical implications in visual world as the end-product of percep-
favor of a description of our experience- tion is unorthodox. The science of
when-we-introspect, that is, the field, and vision, almost from its beginning, has em-
our experience-when-we-do-not, that is, phasized the errors and inadequacies of
the world? vision whereas this conception of the
visual world has emphasized just the op-
The Problem of the Visual World posite. It may strike the reader as naive
The task of stating generally what we to assume that visual perception cor-
see when we say that we perceive the en- responds to its object when everybody
vironment has turned out to be neither knows how misleading perception can
short nor simple. But this lengthy exer- sometimes be. We may not legitimately
cise in introspection has served a pur- assume the correspondence of percep-
pose. It leads to a better understanding tions to physical objects: that would in-
of the problem with which we started. The deed be naive. But on the other hand, we
problem can now be put this way: How may and should consider what correspond-
can we account for the perception of the ence there is, for this is what needs ex-
visual world? For no theory of anything planation. The discrepancies between
less than the visual world will be complete. perceptS and objects are not difficult to
The visual field, as the next chapter will understand; what we need to understand
show, is a reasonably close correlate of is why there are so feu' discrepancies.
the retinal image. Therefore, the explana- That is the real mystery and the really
tion of pictorial seeing is possible on important problem.
The Formation of Retinal Images
The Sequence of Events in Vision . . . . The

Stimulus Variables for Vision . . . . Copies and
Correlates on the Retina . . . . The Retinal Image
and the Excitation of the Retinal Mosaic . . .
The Retinal Excitation as an Anatomical Pattern and
as an Ordinal Pattern . . . . Visual Experience
and the Retinal Pattern of Excitation
If we are to understand visual percep- nature of the optical process or at least

tion we must begin where perception a comfortable feeling that it is known by
begins: with physical objects, light, and experts.
the eye. fI'here is no doubt about the fact The popular idea of the optical process
that all vision, both the pictorial kind and is that a picture is formed on the retina of
the objective kind, is dependent on light each eye. Everybody knows what a pic-
rays and on the formation of images within ture is; hardly anything could he more
the eyes. The discovery of how light be- familiar. It is therefore easy to rest con-
haves and how images are produced - the tent with no more of an explanation than
-
laws of optics is one of the most brilliant that, or simply to assume that the retinal
chapters in the history of science.\ A picture is transmitted to the mind. The
culminating accomplishment of these dis- fallacy of this explanation for perception,
coveries was the publication of a famous if not already evident, will become clear
treatise on physiological optics in 1866 by later on. Rather than examine it now, it
Hermann Ludwig Ferdinand von Helmholtz. would be more useful first to examine the
The intricate and precise series of events way in which the retinal picture is pro-
in the physical world and in the human eye duced.
which make seeing possible were so ac-
curately described by Helmholtz that there The Sequence of Events in Vision
have been few men who could make any im- How is the material environment pro-
portant additions to oui knowledge about jected as an image, and how can this
them for 80 years. As a consequence, image enable us to see? The mechanisms
nearly everybody has me idea of the by which animals obtain an image are ex-
44
THE FORMATION OF RETINAL IMAGES 45
tremely useful since ¡mages enable them to chemically. Depending on how the object
discriminate among objects at a distance is put together (of cells, crystals, and
instead of only those objects in contact so on) and what it is made of (its
with the body. By means of images things chemical substance) it will reflect more
can be approached or avoided and the or less of the light falling on it and it
organism can get about in the environment will also reflect relatively more of one
without collisions. The basic facts of wave-length or more of another. This
physiological optics are fairly complex. differential reflection is the physica(7t
An effort will be made to simplify the referred to when we speak of surfaces as
technical explanations, however, and to having brightness and color. In addition
take a fresh approach to the facts. Only to these simple differences in reflectivity
those facts will be considered which con- there are a great many other complex
tribute to an understanding of perception. differences produced by the structure of a
Treating the problem in this way, the surface. We have names for these denoting
sequence of events can be divided into a the sensory quality but not the physical
number of stages, and we shall take these character of the surface - such names
up one by one. as. shiny, rough, textured, and pebbled.
TIte Array ofPhysical Surfaces. The One thing is certain - that a 12icular
material world, as we all know, is made up kind of surface reflects light in a particular
of solids, liquids, and gases. In actual kind of way.
fact, a great part of it consists of earth, Transmission of Light to the Eye.
water, and air. The first two of these - The1iht reflected from the surfaces of
but not the last - possess surfaces. The the world radiates freely through air but
most common surfaces are those between t1ot through other surfaces, most of which
solids and air. Of secondary importance are, as we say, opaque. The light can be
are those between water and air, and considered analytically to consist of rays
these surfaces, incidentally, are almost which travel in straight lines. Any given
invariably horizontal. Surfaces are ex- point in the open air, therefore, will be
tremely important for our perception of the the juncture of rays from every surface of
world because obviously they determine the material world which is not eclipsed
what we know as objects or things. by another surface at that point. If an eye
The surfaces of objects reflect light, if is stationed at such a point, light from a
they are illuminated, and this fact is the wide array of objects and surfaces will
original basis for visual perception. fall on the cornea and pass through the
Generally speaking, airdoes not reflect pupil, although this light is only minute
light but transmits it; most objects reflect portion of the sum of all the light being
light but do not transmit it. radiated from the surfaces of the world.
-Diflerential Reflection of Light from (Only the rays intercepted by the eye are
Surfaces. The surfaces of the material relevant to vìsion For a pair of human
world differ with respect to their structure eyes taken together, the array of sur-
and composition, both physically and faces represented in the incoming cone
of rays extends about 180° horizontally image on its rearward surface, the retina.
and 150°vertically. This is what we mean Behind this statement lies a long and
by" the field of view. Actually, however, complex story. The exact nature of an
th(s state of affairs is only momeitary. optical image and the way in which it
Any one such cone of rays gives place occurs depends upon the properties of
to another overlapping cone as the eyes light rays and their refraction, or the
move from one point of fixation to another. bending of their paths, in transparent sub'
The comparison already suggested is that stances. 1J'he forepart of the eye is an ex
the eyes play oyçr the. çnvirQneflt like a ceptional kind of solid substance which
searchlight, with the difference that they transmits the light which falls on it
absorb light from the constant flux of rays instead of reflecting it. The behavior of
in the air about them instead of emitting it. light passing through such substances,
4. The Projection of the World as an when they are of certain regular shapes
Image. The cone of light rays which pass called lenses, is to produce a conver-
through the pupil of the eye forms an gence, instead of a continued radiation,
N N
N N
N 'N N
SINGLE N 'N SINGLE

N
N N
FOCUS POINT
-
" -",'- ' "' 'N
N REFLECTING
" "N POINT
: NN
--
---
-
'' /
EYE ------- z' ,/
-'--
- -
'- - --- -.---
-
V'
//
--- -- -y- PHYSICAL
,-
-
, V SURFACE
V,
FIGURE 14. The Theoretical Reflecting

Point and its Corresponding Focus Point
THE FORMATION OF RETINAL iMAGES 47
of the light. Everypoint from which rays of that surface. The total of all these '
originate " is then represented by
øM*. dt ' a ,
a cora
. V
fcus points is the image. The cornea and
responding point behind the lens called lens of the eye have been shown to pro'
the point of focus. As a result, there duce a avery satisfactory image as thus
exists a correspondence between reflecting defined.
points and focus points, each to each, The proof that light can be considered,
such that the character of the light r& for imageforming purposes, to consist
flected at each external point is duplicated of rays is given by the fundamental ex
at the corresponding focus point, as periment of the pinhole camera. If light is
Figure 14 illustrates. The rays of light made to pass through a very small hole it
which pass into the eye from a single behaves like straight lines intersecting at
point constitute what is known as a a point. This is what makes light pro
focused pencil of light. In theory there jective, in the geometrical sense of the
are an infinite number of reflecting points j term, and defines a projective corres-
on a given surface, and the same is true of pondence. The essential fact about the
the corresponding focus points in the image optical image, however, is that it is a
V--
V-
FIGURE 15. The Optical Projection of a Room
geometrical projection of the array of can actually see the inverted image on
surfaces whose reflected light reaches the the curved rearward surface, looking
pupil. Ari illustration of this projection something like a miniature photographic
is given in Figure 15. The observer is transparency. It is this demonstration
looking at the wall of the room in front of which has led to the theory that the retinal
him, and his visual field includes part of image is a "picture."
his body at one extreme and part of the 5. The Mosaic of Retinal Elements. The
ceiling at the other. The surfaces from surface of the retina on which the image
which light reaches the eye are drawn in is projected is composed principally of
solid lines; all others are given in dotted extremely minute cells which contain
lines. The environment outside the visual photosensitive substances. Like the
field is indicated by shading. The straight substances used in photographic emul-
lines from the surfaces to the retina re- sions, these are capable of reacting dif.
present focused pencils of light which ferentially to the energy and wave length
form the image on the retina. The only of light. They are superior to any photo-
rays of light actually shown in the drawing graphic emulsion, however, since they are
axe those from the edges of physical sur- self-renewing and capable therefore of
faces which, it will be argued, have a registering the image continuously. A1
special significance for perception. The though the television camera can register
areas of physical surfaces correspond to an image continuously and in this respect
areas of the image. Not all the surfaces is more like the retina, its mechanism is
of the world at different distances will be quite different. The cells of the retina are
in perfect focus at the same time, it is of two types, rods and cones. The cells
true, but with a normal eye there is very are distributed much more thickly in the
little blurring. Although the retinal center of the retina than in the periphery,
image is inverted and the order within the grading off from a density of 160,000 per
image therefore reversed, it nevertheless square millimeter at the fovea to a much
corresponds to the physical world as a sparser distribution outward from the
projection. The assumption here (and fovea.1 Both rods and cones are wholly ab-
throughout this book) is that for certain sent in a small peripheral area where the
purposes we may treat the retinal image as nerve bundle has its exit from the eye.
if it were a two-dimensional pinhole image.
It is important to note that this is not the
kind of image defined by physical optics 1The distribution of rods is quite different
and used in the design of optical instru- from the distribution of cones. The fovea con-
tains only cones whereas in the peripheral reti-
ments, for this latter is three-dimensional. na rods predominate over cones. The decrease in
The formation of an image on the retina visual acuity toward the periphery is probably
related to the decrease in density of receptor
can be observed directly. If the excised cells and to the fact that the foveal cones have
eye of an albino rabbii is fixed into a hole individual neurons while the peripheral rods
are grouped with single neuron. Even at the
in a card and pointed toward a scene, by outer edge of the retina there appear to be more
holding itin front of one's own eye, one than a thousand cells per square millimeter.
FIGURE 16. Scheme of the Retina
Note the different types of cells (neurons) and the variety of
relationships and interconnect,ons of their tips (synapses).
The lower edge of the figure represents the layers lying fo-
ward the interior of the eyebaH; the top edge shows the
long, narrow rods and the shorter, thicker Cniies, which
point outward away from the interior of the eyeball. Light
admitted into the eye through pupil, lens, etc,, has to pass
through layers 10, 9, 8 etc., before falling upon the sensi-
tive tips of rods and cones in sayers 2cx and i. (Prom S. L.
Polyak, The Retina, L niversity of Chicago /rp,cs, 1Q41.
By permission of the publishers.)
The nature of their differential reactions to ter of the light focused on them, and
light has been studied for many years therefore indirectly to the character of
without, until recently, any close approach the surfaces from which the light was re-
to an understanding of it. The type of flected. Another fact is equally certain,
photo-chemical reaction which corresponds that they are connected with the individual
to wave length is particularly puzzling, as fibers of the optic nerve, although not in a
the various theories of color vision bear perfectly one-to-one fashion.
Witness. One fact is certain, however, 6. The Anatomy of the Optic Nerve.
that the elements making up the retinal The rodlike and conelike cells of the
mosaic do react specifically to the charac. retina, when they are stimulated by light,
initiate nerve impulses in the neurons \ has been assumed that the connection of
which make up the sheaf of libers, a \the retina with this visual area was an
quarter of a million or so in number, which xact point-to-point relationship and it
we call the optic nerve. So far as the evi was possible to infer that therefore the
dence goes, these nerve impulses are etina1 image was projected on the brain
excited independently of one another and ¡in the same way that the physical world
travel their paths separately. Very little projected on the retina. But the ana-
more than this is known about them. The facts provide only a puzzling and
itomical
anatomical connections of the nerve very incomplete support for this assump-
fibers can, it is true, be traced. Some of tion (6). Lateral connections exist among
them connect with centers in the brain adjacent fibers both in the retina itself and
governing the movements of the eyes, the at later stages in the tract between retina
regulation of the size of the pupil, and the and brain. The amount of overlap is such
accommodation of the lens.) By far the that the "image" on the brain (if there
largest part, however, connect with an were such a thing) would be very blurred.
area on the surface of the occipital lobes In all probability, it should not be thought
of the brain. The excitation of this corti- of as an image, and even less, as a
cal area is probTy essential to all vision literal picture. It is an event composed
in man, for destruction of it produces not of light, but of nerve-cell discharges,
blindness just as much as would injury to and if a surgeon exposed the brain to
the eye or severing of the optic nerve. It view, there would be nothing to see.
POSTERiOR CHAMBER
(viTREous HUMOUR)
BONE OF ORBIT
/MUSCLE
OPTIC NERVE
EYE BROW
EYE LID
SUSPE: NSOR Y
LIGAMENT OF LENS-.
co R N EA
ANTERIOR CHAMBER '(
-
(AQuEous HUMOUR)
LENS-
IRIS
-\ RETINA
MUSCLE
CHOROI D
CHEEK bONE SCLERA
FIGURE 17. A Diagram of the Human Eye
(After H. W Haggard, Man and His Body (Harper & Brothers, 1938)
F
FIGURE 18. The Sequence of Transforma-

tions in the Process of Visual Perception
A. The physical environment: a wedge-shaped physical oblect, reflecting hght. B. A
"picture" of the physical environment: a plane projection of the light reflected from the
physical object. C. The reflnal ¡mage (the proximal stimulus for vision): a curved pro-
ection of the light reflected from the physical object. D. The pattern of excitation: a
mosaic of photosensitive receptors. E. The brain process: a bifurcated and oddly-shaped
projection of excitations on the rear surfaces of the hemispheres. F. The visual world,
or phenomenal experience: the experience of a wedge-shaped object. G. The visual field,
or the color-sensations obtained by introspection: the impression of two fjot patches of
color òdjacent to one another.
Sensitive electrodes placed upon the cor- world. This experience is both elaborate
tex might pick up regions of high and low and exact. So this discussion has come
activity with gradients or contours between around again to the same problem with
them, but this observation has not been which it started.
conclusively made. The fact is that no
The Stimulus Variables for Vision
one yet has an adequate conception of it.
7. The Unknown Activity Producing Fortunately, it is not necessary to
Vision. All that is certain about the last understand the events within the nervous
stage of visual perception can be put into system (stages 6 and 7) in order to be
a few words. There are unquestionably able to make a scientific attack on the
neural processes at the occipital surface problem of perceptions One can by-pass
of the brain. These processes arouse still the nervous system and jump from the
others. They are almost unknown.2 Never- retinal image directly to the perceptual
theless, these unknown events are the experience. One can, in other words,
sole basis of our visual experience of the seek to establish an empirical corres-
2There are, to be sure, a number of 4es- level without knowing the principles at the
tablished facts about the processes within psychophysiological level has already been
the brain which correlate with visual per- referred to in Chapter I.
ception, and more are continually being es- The processes referred to above may not
tablished. The facts are, however, as yet even be wholly cortical. Certain mammals are
PUZzling and incoherent, or at least they seem known to have some residual sub-cortical
so to the writer. The study of visual brain- vision persisting after the extirpation of the
processes is being pursued by Kohier (69) and whole occipital cortex. Whether the same
others (66, 50). The possibility of reaching condition is true in any degree for man is not
principles of explanation at the psychophysical yet certain.
pondence between the stimulus and its accounted for. If our environment con-
conscious resultant. This is what psycho' sisted of nothing more than a homogeneous
logists have been doing with color stimuli, sea of light, without surfaces or objects,
sound stimuli, and others for more than a then all our visual experiences could be
hundjed years, and the results of this en- specified in terms of these variables with
deavor have produced the most securely nothing left over. Each retinal point
established body of facts known to psycho- would be stimulated in the same way as
logy. lt is traditionally known as psycho- every other retinal point. But obviously
physics. A retinal spot of light having a our visual world consists of more than
wave length of 760 millionths of a muli- this. The stimulus situation for a typical
meter yields a visual spot of the quality environment is diagrammed in Figure 15.
red; an airwave of 256 double vibrations Not merely colors but surfaces and edges
per second in the ear produces the quality are projected in the retinal image. There
of middle C. Experiences generally have must exist, therefore, a second type of
a specific relationship to the stimuli stimulus variable in the image. The locus
which arouse them, as indeed they must if of the classical stimulus variables is the
the experiencer is to adapt his behavior single spot of light, since each focused
to his environment. This is the principle pencil of light may possess its own
of sychphy%içal cQrrespondence. Little unique combination of wave lengths and
as we may comprehend its physiological intensity. But surfaces and edges are
and nervous basis, the rule is that varia- not related to this kind of variation within
tions in stimuli are co-ordered with varia- the spots of the image; they are related
tions in the character of the perceptions. instead to variations among the different
Musical tones, for example, are related to spots of the image.
the frequency of air vibrations in much the The facts of the situation are represent-
same way that the letters of the alphabet ed in Figure 15 . The array of physical
are related to the number series from i to surfaces whose reflecting points are
26. This rule has never failed of verifica- duplicated in the image is shown in solid
tion for stimuli which can be ordered in lines. As projected, these surfaces border
physical terms, or in other words, for on one another. In other words the edges
stimulus variables. Let us now analyse of these surfaces correspond to abrupt
the retinal image to see what kinds of changes in the energy and wave length
stimulus variables are included in it. character of the light spots composing
The classical stimulus variables for the image. For example, if the floor is
vision are, of course, the physical varia- dark brown and the table is light gray, the
.
tions of light itself. Th ar i wave color stimulation in the image will shift
length or frequency
._--_--. .-.. -
and its energy or in- accordingly along the margin between the
tsityl By combinations of these, and
-
two parts of the image. The image, there-

by mixtures of wave lengths, all the ex- fore, is made up of areas of different light-
perienced qualities of pure color and character, and it is the transitions between
brightness of color as such can be these areas which give rise to visual lines
or contours. In the illustration, the pen Terms like distribution, pattern, and ar-
cils of light from an edge to its corres- rangement are not very exact, it must be
ponding border in the image are represent- acknowledged, and an effort will be made
ed by straight lines. in Chapter 5 to be more specific about
Not only the edges, however, but also this variable. It is evident, however, that
the areas of the physical surfaces have the kind of arrangement we are talking
specific representation in the image. The about is simply that of adjacent order on
physical irregularities of the surface, the retinal mosaic. A transitional arrangea
both its gross composition and also its ment of color-points yields a line or con-
minute structure (if the lens of the eye is tour. An alternating or scattered ar-
accommodated for that surface), are pro- rangement of color-points, so far as we
jected as correlated irregularities in the know, yields a surface. An array of
image. A tiny depression of the surface, homogeneous color-points, all identical
for example, is focused as a dark spot, a (which is not an arrangement at all),
slight protuberance as a minute high light, yields pure insubstantial color (61).
and an array of such surface-elements as
an array of dark and light spots. This
type of stimulation gives rise to what we Copies and Correlates on the Retina
will call the quality of visual texture. It The foregoing short survey of optics
seems probable from the evidence of points to a conclusion. The image is an
Metzger (81) and others that texture is what arrangement of focused light on a physical
makes a surface perceptible as a surface surface of two dimensions which is specif-
instead of as mere insubstantial areal Ic to an array of reflected light from
color. It may be noted that physical things piysical objects and surfaces in three
like clear sky, dense fog, and regions of dimensions. Since the reflected light is
complete darkness such as the mouth of a specific to the objects and surfaces thern
cave do not reflect light as a surface selves, the image is also specific to them.
does, do not possess texture, and are not Geometrically, we say that the image is a
seen as surfaces. projection of the world. The conclusion
A typical retinal image, then, contains is that the image is not a replica of the
two fundamental types of stimulus varia- world. If taken seriously this conclusion
tion, one in the character of the focused has far-reaching implications.
light at any point and another in the tela- Unfortunately, the word "image" has
tion of these light-points to one another. more than one meaning. It may refer to an
The first is the classical variation in effigy or copy - the "graven image" of
stimulus quality and intensity; the second the Bible - or it may refer to the projected
is variation in what has loosely been arrangement of light as just described -
termed stimulus "distribution" or ccpat the image of physiological optics. The
tern." An image is an arrangement of two meanings of the term are easily con-
Color-points, and it may vary either in the fused and there may be intermediate
color of the points or in their arrangement. meanings between them. But the retinal
image is unquestionably a projection rather dimensional surface.

than a facsimile. Everybody knows that The correspondence between the world
objects themselves do not get into the and the optical image need not be that
eye. Neither do small replicas of things between a thing and its copy; it need
get into the eye, although this was a only be that between a material quality
reasonable theory of perception before and its correlate. There is no counterpart
the discovery of the nature of light. in the image of that physico.chemical
Nothing gets into the eye but radiant character which gives a surface its parti-
energy. Only because it is focused is it cujar hue, but there is a correlate, wave
speciñcally related to the object. The length. There is not a counterpart in the
object therefore does not have a copy image of the physical microstructure
in the image but a correlate. The fact is which gives a surface its texture but it
that the optical image does not have to be does have a stimulus-correlate, as will
like its object to make vision possible. be evident. Above all, there is no copy
The chief source of misunderstanding in the image for the shape of an object in
here is the assumption that the retinal three dimensions, or what has been called
image is a picture. It might be argued its depth-shape, and this is something
that even if the image is not a replica which all genuine objects possess.
of the environment it is at least a re- Similarly there is no copy in the image of
presentation of it. The apparent simpli- the solidity and distance of the environ-
city of this pictorial analogy for vision ment in general, but there must be some
makes us reluctant to give it up, scientists correlates for these variables, or we could
as much as anybody else. But a picture not see them. Finally, the size-relations
as a representation of something is nothing of the objects in the environment and the
if it is not presented to an eye. An unseen interspaces between them, following as
picture is only an arrangement of pigment they do the laws of Euclid and not the
spots, if it is a painting, or ari arrangement laws of perspective, are not copied in the
of metallic grains of silver, if it is a photo- image. But for these and the other features
graph. It is simply a part of the material of the world there must be some basis in
world which has to be seen, like anything stimulation, however complex. This
else. If the retinal image were really a basis remains to be discovered.
picture there would have to be another eye There could, theoretically, exist a
behind the eye with which to see it. The material environment for which the retinal
notion that we see our retinal images is image would be almost a duplicate. It
based on some such idea as a little seer would consist of a large picture at right
sitting in the brain and looking at them. angles to the line of sight and filling the
The question which then arises is how he entire field of view. The hypothetical
can see. The retinal ¡mage should not be 1tpicture plane" which is posited by the
thought of as a picture or a representation perspective draftsman at a fixed position
but as a physical arrangement on a two in front of the eye would define such a
picture. It has been illustrated in Figure The light composing the irnae of
7. The lines, areas, shapes, and sizes f9rm environment is egually dense over
in the picture would then be duplicated s whole area, being evenly distributed,
in the image, the principal difference the pattern of excitation is most
being that the image would be upside down dense - in the center of the retina where
relative to the picture. Such an environ- the cones are concentrated and least
ment, of course, never existed. But most -' the peripherywhere the cells are
of the experimental research on visual thinly distributed. Above all, since the
' -,-- ' .
perception in psychological laboratories imae is an event in the light-flux of the

has been performed with stimulus objects physical world, it has reference -.- -to the
of just this kind, drawn or exposed on a Lct.. a!? is tixed in relation to ir. It
plane surface. And the assumption that keeps a constant alignment with gravity,
the fundamental kind of seeing is "pic- for instance, when the head is tilted and
tonal seeing," or the perception of a the retina rotated. The rI retinal excitation
»it..,, ,- ,w
depthless world, ìs also consistent with on the other hand, having its reference
the duplication theory of the retinal image. to the retina itself, is.ø a pattern composed .
Perhaps these facts provide the explana- of retinal elements which remain- N,,_
--'-- --. . . ,
the same i
tion of its persistent hold on our thinking onL1c longas theeye does not move.
about the perceptual process. When the eye moves, the image is trans-
The Retinal Image and the Excitation of posed on the surface of the retina and con-
the Retinal Mosaic sequently there ¡s a shift in the pattern of
excited elements in the anatomical mosaic
ÇThe step between the formation of an
image on the retina and the excitation of of cells.
the mosaic of rods and cones, stages 4
and 5, is one which must be kept in mind The Retinal Excitation as an Anatomical
if our reasoning is not to go astray' lt is Pattern and as an Ordirtol Pattern
easy to assume that the retinal image .4
Another troublesome question now arises:
aid the retinal excitation are the sanie how can the retinal image be transposed
thing. But the former, clearly, is a matter and still retain its equivalence as a
fr
of physics while the latter is a matter of
physio1ogy. jhe image is an arrangement
-
of light-points while the excitation is an
stimulus for vision? The pattern of excil
tation, it must be
-
remred, set
is a- -4-w
of units of finite size, correlative with
-
arrangement of discharging nervous ele- I?aLby no means exactly duplicating the
ments. These ----.,
-- .. .- ,.-..,----.---
individual points
-- -- -,
of the
_4 . fl-,-
This spot-pattern is about as close
may be noted, together with the to the immediate basis of visual percep-
rays of light which explain the correspond- tion as our present knowledge will take
.
i
ence to the world, are pure geometrical I'*"*"

-L____Jl-----*_J
us, so it needs to be examined with care.
fictionntroduced for purposes of analy- rom one point of view itiscomposed of
sis, whèrèas'tFie Individual spots of 'the nerve cells in an anatomical relation to
excitation-pattern are anatomical facts. one another. But from another point of
view it is composed of nervous excita- It seems possible that the organism can
tions in a purely ordinal relation to one react to an ordinal pattern as well as to
anoth erJ an anatomical pattern. The television
CAn analogy may
be useful here. Every- camera can register a purely ordinal and
one has seen electric signs, consisting of transposable pattern - why not the eye of
a mosaic of light bulbs wired to a corres- a living animal? An attempt to deal with
ponding set of contacts, of such a sort the ordinal type of stimulation more exactly
that any number of patterns can be lighted will be made in Chapter 5.
up on the same sign. The retina is more For the present, it is enough to empha-
like this than like a photographic film, size the fact that the identity of a given
which is only good for one exposure. In point-stimulus in the eye depends not at
some of these signs the patterns of lighted all on the anatoiìica1 point of the retina
bulbs can be transposed on the mosaic so stimulated but entirely on the position of
that words or figures will move across it. that point relative to other points of stirn-
Now a stationary pattern can be defined ulation. I iZ'('fl sf)()t of light in (i iVPfl
by specifying the contacts of the bulbs re t in (2 i im a e i s ti, e s a ni e s ot (i t di fie re n t
which light up. This would be an anatomi- i n s t an ts (i f i in e i /1 e n i t s p o s i t i o n re la t i ve
C
cal pattern. But a moving pattern can to the order of spots is determined, not
best be defined by specifying not contacts ,,/!en its f)OSitOfl relative to the retina is
but the adjacent order of contacts. This (Le terrflifle(J.
would be what was called above an This fundamental fact, that a spot of
ordinal pattern. In other words the pat- light is a stimulus for perception by virtue
tern can legitimately be thought of as of its ordinal location and not by virtue of
either an arrangement of bulbs or as an its anatomical location, can be illustrated
arrangement of lights without regard to by experiment. Even in the case of two
bulbs. Likewise the pattern of excitation spots in an otherwise blank field of view,
on the retina might be defined either in the principle holds. Let a, b, and c re-
terms of the units of the uosaic or the present three separated retinal points in a
units of excitation as such. The former line. If points a and b are stimulated
pattern is embodied in units which have briefly and a moment later points b and c
an anatomical meaning; the latter pat- are stimulated in the same way, an ap
tern is nor embodied in. this way but parent movement will occur. What the ob.
nevertheless it is definable in terms server sees is a pair of spots which move
which have mathematical meaning. The toward the right. The identity of the ana-
units of excitation maintain a constant tomical point b with itself is not sensed.
ordered relation to one another when the The spot at a moves to b and the spot
retinal image is transposed even though at b moves to c.
the anatomical units do Rot. The ordinal a1 /)1 C2
pattern, therefote, is preserved when the
eye moves although the anatomical pat- Points a'
and b1 flash on first. Points b2
tern undergoes a complete rearrangement. and c flash on a moment later.
The relation of one spot on the right of clear in all its parts. Only the suggestion
another is what is sensed; the relation of of an explanation of these facts can be
either spot to the retina is not apprehended given here and the answer is therefore in
at all. This becomes wholly compelling complete and tentative. We can be fairly
when the number of spots is increased certain, however, that the visual world is
and when the "form" of the spots is dependent on eye movements and is not
easily seen. The experiment was carried seen as the result of a single fixation or a
out and elaborated by Ternus early in the momentary visual field. It must correspond,
program of the Gestalt theorists (104, therefore, to successive patterns of ex
translated in 32) and it illustrates their citations on the retina, united perhaps by a
main contribution to the science of vision kind of immediate memory. These patterns
- the doctrine of the "transposable" will overlap one another anatomically as
Gestalt and the abstract conception that the eye moves, and the basis for the
the "parts" of a perception exist only in visual world, therefore, must be what has
relation to a "whole." been called the ordinal pattern of excita-
tion rather than the anatomical pattern. If
Visual Experience and the Retinal Pattern it be assumed that there is an ordinal pat-
of Excitation térn which keeps its integrity during eye
\r movements, then it is possible for any
can now consider once more the two
kinds of seeing described in the previous part of it to be brought to the center of the
chapter, the visual field and the visual anatomical mosaic and registered in fine
world, and we can assess their corres- detail. A complex of this sort, over time,
pondence or non-correspondence with the would be both uniformly differentiated and
retinal stimulatton.L n several obvious unbounded, and might therefore provide a
ways the visual field corresponds with the basis for the perception of the visual
anatomical pattern of excitation. It is world (Chapter 8).
finely differentiated at the center of clear There is also the question why the
vision and becomes progressively less de- visual field shifts during eye movements
terminate away from the center; this re- but the vìsual world does noti, It was made
ulects the dense distribution of cells at clear in Chapter 3 that the visual fields be.
the retinal center, the fovea, and their fore and after a change of fixation are
thinning Out toward the periphery. It has different. The first array of color patches
boundaries which can be plotted; so also is by no means the same as the second.
does the mosaic of rods and cones, and This fact is parallel to the rearrangement
these boundaries agree An object which of the anatomical pattern, and implies
passes out of the visual field corresponds that the field corresponds specifically to
to an object which ceases to project rays that pattern. The visual world, however,
on sensitive elements. The field can is not rearranged after a change of fixa-e
easily be accounted for in these respects. tion; the objects which are seen appear
But how aboux. the visual k to be the same before and after the eye
*
; have boundaries and it is more nearly movement. The stimulus complex to
which it corresponds must therefore be visual field and the visual world, the
based on the ordinal pattern of excitation anatomical pattern and the ordinal pattern
which is transposable over the retinal provide correlates. But there remains the
mos ak. most important of all the differences be-
Will these hypotheses also explain the tween the visual field and the visual
appearance of after-images? if a negative world, namely the three dimensional
after»image is a patch of excitation in the character of the world. The psychological
anatomical pattern, that is, one which is fact that the visual field tends to be
fixed on the mosaic and not displaced made !IP of projected shapes rather than
when the eye moves, it should be seen in depth-shapes, and that it tends to have an
a fixed position with reference to the appearance consistent with the laws of
visual field. It should, in other words, ap perspecti ve rather than the laws of Eu-
pear wherever we turn our eyes, and this is
what it does. If, moreover, the visual
world is a correlate of the ordizial pat
tern, then the after.image should appear
to vary its location with reference to the
pro; i
did, is in agreement with the physical
fact that the retinal stimulation is a
But how are the depth shapes
themselves to be explained, and why does
the horizon really look like the world at a
world. It should, in other words, be very great distance and not like a line at
superposed on different object surfaces which sizes vanish and where parallels
as the eye moves, and so it is. The meet? The eclipsing of projected shapes
patch of excitation corresponding to an in the visual field is consistent with the
afterimage and the patch of excitation projected character of the retitial stimula-
corresponding to a transposable object tion, but what about the "seeing behind"
are both results of stimulation, but the impression which we get with objects in
difference between them is highly signi- the world? How about the obvious but
ficant for the theory of vision. The for- puzzling fact that the world, and the field
mer reminds us that stimulation of a too, are external? The stubborn fact is
mosaic must always necessarily have that we see and get around in a world
reference to the mosaic. The latter which stretches from here to there and
reminds us that stimulation of a mosaic this fact remains in need of explanation.
must with equal necessity consist of a The effort to find such an explanation will
mathematical order. be the principal concern of the next few
For all these characteristics of the chapters.
A Psychophysical Theory of Perception
Abstract Space and the World of the Flier ....

Stimulus Correlates . The Hypothesis of
Ordinal Stimulation ....
Retinal Gradients as
Variables of StimulationTexture The Cues
for Distance as Stimulation Gradients .... The
Concept of Gradient .... The Concept of
Psychophysical Correspondence .. Summary
At the beginning of World War II, the plication to the problem of flying. The
theoretical problem of space perception theory of the binocular and the monocular
became a practical problem almost over- cues for depth, perfected eighty years
night. The skills of aviation began to be before by Helmholtz, could explain how a
of vital interest to millions of individuals. pilot might see one point as nearer than
The abstract question of how one can see another point. But the pilot was not
a third dimension based only on a pair looking at points of color in a visual field;
of retinal images extended in two dimen- he was typically looking at the ground,
sions became very concrete and im- the horizon, the landing field, the direc-
portant to the man who was required to tion of his glide, not to mention several
get about in the third dimension. If the instruments, and visualizing a space of
visual world of the airplane pilot were air and terrain in which he himself was
not in fairly close correspondence with moving very fast and possibly in a cold
the material world on which he had to sweat.
land his airplane such as a carrier-deck,
the practical consequences could be dis- Abstract Space and the World of the Flier
astrous. The theories of space perception, The space in which the pilot flies is
therefore, became of more than academic not the abstract space of theories, nor
interest in the rapidly developing field of ,the lines and figures of the stereoscope,
aviation psychology. nor the space of the usual laboratory
But the fact was that all the evidence apparatus for studying depth perception.
from the laboratories and all the theories It does not consist of objects at varying
of ingenious men had little practical ap- empty distances. It consists chiefly of
59
one basic object, a continuous surface of for the problem of visual perception.
fundamental importance the ground. A Space, time, points, and instants are use-
pilot who cannot see the ground or sea is ful terms, but not the terms with which to
vt to lose touch with reality in his flying. start the analysis of how we see, for no
A'bisual field of blue sky, or fog, or total one has ever seen them.1
darkness yields an indeterminate space The world with a ground under it the
which is the nearest thing to no space at visual world of surfaces and edges is
all. Only a substitute for the ground and not only the kind of world in which the
its horizon in the form of instruments will pilot flies; it is the prototype of the world
permit him to maintain the level flight of the in which we all live. In it, one can stand
airplane under such conditions and to pro- and move about. It conditions and pro-
ceed from one place to another. The vides support for motor activity. A ground
spatial situation which needs to be analy- is necessary for bodily equilibrium and
sed, therefore, must involve the ground posture, for kinesthesis and locomotion,
and everything that it implies. Instead of and indirectly for all behavior which de-
calling it a space it would be better to pends on these adjustments.
call it a world. An out-of-doors world is one in which
The conception of an empty space of the lower portion of the visual field
three dimensions was a conception of (corresponding to the upper portion of
philosophers and physicists. It was each retinal image) is invariably filled by
appropriate for the analysis of the ab- a projection of the terrain. The upper
stract world of events defined by Newton. portion of the visual field is usually filled
It was and still is of enormous value for with a projection of the sky. Between the
analysis in the physical sciences. But upper and lower portions is the skyline,
the fact that it simplifies such problems high or low as the observer looks down or
does not make it the best starting point up, but always cutting the normal visual
field in a horizontal section. This is the
kind of world in which our primitive
ancestors lived. It was also the environ-
'The theories of space-perception which

flourished in the 19th century were all theories
of abstract, empty space. The experiments
concerned lines and points in an indeterminate
visual field, as seen in a stereoscope or a
depth-perception apparatus. The theories
and the experiments alike may be characterized
as geometrical. They were great intellectual
achievements (the theories of the "horopter"
are an example), but they will not be con sider-
ed at present. The theory of disparate retinal
images as it applies to surfaces rather than
abstract points will be restated in the next
FIGURE 19. The Typical chapter, and the conception of geometrical
Vi scat of _JJne Eye space will be treated in Chapter 10.
A PSYCHOPHYSI CAL THEORY OF PERCEPTION 61
ment in which took place the evolution of this explanation was impossible so long
visual perception in their ancestors. Dur- as the continuum of distance was con-
ing the millions of years in which some ceived as the third dimension. The solu-
unknown animal species evolved into our tion of the difficulty is to recognize that
human species, land and sky were the the continuum of distance depends on a
constant visual stimuli to which the eyes determinate surface which extends away
and brain responded. In the typical in- from the observer in the third dimension.
doors world of civilized man, a ceiling Such a surface is projected as an image
and walls take the place of the horizon which is spread out on the retina, not
and sky, but the floor is still an equi- confined to a point.
valent to the ground. This basic surface Figure 20 illustrates the two formula-
is the background for the objects to which tions of the problem. The points A,R,C,
we normally give attention and, as we and D are not discriminable on the retina.
learned in Chapter 3, its horizontal axis Distance along this line may be a fact of
is implicit in every visual field whatever geometry but it is not one of vision. The
the posture of the body may be. It is points II ,.k, Y, and Z at corresponding
little noticed, but on the average and distances are discriminable on the retina.
over the ages it must have determined the They represent the image of an extended
fundamental pattern of retinal images for surface, the points being, for example,
all or most terrestrial animals.2 highlights on the surface. It may be
The classical theories of space per- noted that the retinal spots become pro-
ception conceived the third dimension to gressively closer together as the distance
be a line extending outward from the eye. increases. What kind of a theory, we may
Space was therefore empty between the now ask, is implied by this latter formu-
eye and the object fixated. The per- lation of the problem? How is a surface
ceived distance of this object seemed to seen?
be what heeded explanation, and the ex- Stimulus Correlates
planation was sought in the consequences
The first place to look for an explanation
of the possession of two eyes. It would
have been better to seek an explanation of is obviously the retinal image. If, con-
the sensory continuum of distance as such trary topast teaching, there are exact
which, once visible, determines how dis- concomitant variations in the image for
tant all the objects within it are. But the important features of the visual world
a psychophysical theory will be possible.
The image, according to the evidence in
2So universally is the ground taken to be
the background of objects that the mere loca- Chapter 4, is a good correlate (but not a
tion of one patch of color above another in the ",copy) of the physical environment. It
visual field tends to make it appear more dis- may also prove to be a good correlate of
tant. Height in the visual field can be a
genuine clue to distance only if upright pos- perception, despite an entrenched opinion
ture, a level ground, and a tendency for ob- to the contrary. The retinal image, it is
jects to be on the ground are assumed. This
point will come up in later chapters. true, is not much to look at when one
XW
w X Y
FIGURE 20. Two Formulations of the Problem of Distance Perception
compares it with the elaborate reality of perceptions, it can and should be per-
the visual world, but the fact is that it formed (40).3
is not something to be looked at; it is a It is, of course, a departure from tradi-
stimulus. The question is not how much tion to conceive that a surface, or an out-
it resembles the visual world but whether line, or the depth of a surface should have
it contains enough variations to account a specific stimulus. The stimulus for
for all the features of the visual world. vision, we are accustomed to think, is
If we can analyse the retinal image for simply light energy. But such a definition
its stimulus variations, we shall open up reflects the preoccupation of nearly a
the possibility of experimental control of hundred years of research on color vision
these variations. Given a means of and light-dark discrimination, the outcome
producing them, an experimenter, and an of which still leaves us ignorant of the
observer, it can be determined whether vision employed in everyday living. The
the -variations are or are not in psycho- higher animals do not simply react to the
physical correspondence with the ob-
server's perceptions. This is the method 3Students
of psychology will recall, in this
by which the sensory capacities, so connection, that the Gestalt theory denied
called, of men and animals have been de- any one-to-one correspondence between the
stimulation of receptors and the experience
termined. The test is simple: does a which resulted. The assumption of such a
specific variation in the observer's ex fixed correspondence was called the "con-
perience (or behavior) correspond to a
stancy hypothesis." It seemed to be un-
tenable since everyday visual experience was
variation of the physical stimulus? Al- so demonstrably unlike its retinal image. The
though this experiment has seldom been aim of this chapter is to reassert the constancy
hypothesis on the basis of a broader concep-
appliecP4to what are traditionally called tion of stimulation.
A PSYCHOPHYSICAL THEORY OF PERCEPTION 63
direction of light as a plant does; they object" will never be used, since it can
have a specialized neural surface, the serve as a cloak for ignorance. The dis-
retina, on which their environment is pro- tinction of Heider (51) and KofIka (67) be-
jected by means of focused light. As a tween the "distant" stimulus (the object)
result they can react indirectly to the en- and the "proximal" stimulus (the image)
vironment itself, and the point would be is illuminating just because it implies,
missed by insisting that they are actually and just so long as it implies, that the
reacting only to light. In what respects latter stimulates the organism. The term
is this projected light a stimulus? "stimulus situation" likewise will never
be used since the situation does not
The Hypothesis of Ordinal Stimulation exist in the retina any more than the
Before attempting to answer this ques- object does, and the question is how both
tion it would be useful to agree on just are seen.
what the term "stimulus" is to mean, for How can we specify the order of visual
it is a much misused word. Let us assume stimulation? The retinal image as a
that a stimulus is a type of variable physical event may be treated as an in-
physical energy falling within certain finite series of geometrical points or as a
ranges of variation (the limits being called finite number of minute areas of arbitrary
absolute thresholds) which excites a re- size. The latter is the more useful as-
ceptor or a set of receptors differentially. sumption for our purpose. In either case
If it does not release physiological ac- the image can vary in two fundamental
tivity in a receptor-mechanism it is not a ways: first in the character or "color" of
stimulus. As the energy varies succes- the focused light at a given spot, and
sively, the excitation varies concomitantly second in the distribution of these spots,
in some specific way. This is a strict or their geometrical relations to one
definition of a stimulus. For our present another. This second variable is the one
purposes, as applied to the retina, we that makes all the difficulty. It is not
wish to extend the term to mean also a easy to deal with the complexities of dis-
simultaneous variation over the set of tribution or arrangement in a mathema-
receptors, or a differential excitation of tically precise way. Nevertheless this
different receptors, and the order of such a variable is the one on which everyday
variation. For the extended meaning the perception depends. Let us assume, as a
term ordinal stimulation will be used. start, that organisms can react specifically
"Ordinal" simply refers to order or suc- to the order of the light-spots as well as
cession. This is what has usually, but to the character of the light in each spot.
inaccurately, been called pattern stimula- How an organism can do so, we do not
tion. know that is another question. But if
In this book, the term "stimulus" will it can react differently to a spot-sequence
always refer to the light energy on the such as "black-gray-white" from the way
retina, never to the object from which it does to the sequence "white-gray-
light is reflected. The term "stimulus- black," then the order is the effective fact
and it would be legitimate to invent for (112, p. 85). Each hypothetical light ray
it the term ordinal stimulation.4 was supposed to be an individual stimulus.
The spots or elements of the image, in It has been argued, however, that light
this analysis, are to be understood as rays are analytic fictions. Furthermore,
arbitraty units of area. Like the points the homogeneous total field experiment de-
of geometry and optics they have only a monstrates that when every ray has the
logical existence. They are necessary same wave length and intensity there is no
to a mathematical treatment of retinal perception, and this experiment implies
stimulation, but they are not to be con- that the organism cannot respond to ray-
sidered the elements of visual experi- directions as such. What the retina does
ence, or the sensory units of perception. respond to is a differential intensity in
They are analytical fictions, and they adjacent order over the retina. The neces-
do not add up to a visual field any more sary condition for pattern vision is an in-
than the geometrical points employed in homogeneity of the set of hypothetical
the operations of a surveyor add up to the rays, not the rays themselves. The ray-
surface of the earth which he surveys. It direction theory of the stimulus, the point-
should be emphasized that the fundamental theory of objective space, and the local-
variations in light energy and in order or sign theory of subjective space all collapse
arrangement which constitute the retinal together if this implication is correct and
image are both abstractions. require a thorough reformulation.
The prevailing assumption about pattern Considering the retinal image as an ar-
vision has always been that the ocular ray of small adjacent areas of, different ra-
mechanism enables the organism to re- diant energy, let us try to state the kinds
spond to a specific set of ray-directions of order into which the elements might fall.
For the sake of simplicity we may con-
4As will appear in a later chapter, there is sider a hypothetical case in which there
evidence that organisms can react specifically
to a successive order of stimulation of the are only two levels of light intensity in
same spot as well as to an adjacent order of the image and no differences of wave-
stimulation of different spots. Both kinds of
order are present in retinal stimulation. The length. An element may be relatively
successive order "black-gray-white" yields "light" or "dark," but nothing more. If
a lightening effect and the reverse a darkening
effect; the adjacent order yields a patterning the former it may be indicated by the
effect. A co-variance of successive and ad- letter 1 and if the latter by d. The sim-
jacent order seems to be the essential condi-
tion for visual motion. For the present we are plest of all orders would then be 11111111 or
concerned only with adjacent order. dddddddd. All the elements of the order
The term order is often used by philosophers
and artists in a very inclusive sense. It may are the same. This is what Koffka has
mean form, pattern, arrangement, position, called homogeneous retinal stimulation
direction, and even magnitude or distance. (67, p. 110). In a two dimensional array
The term is here used, however, in an exact
and literal sense to refer to that character- it is the stimulus correlate of visual
istic which numbers have of making a sequence fields like the sky, absolute darkness, or
which is not the same in one direction as in
the other. the "film-colors." The experience is one
of pure areal color, seen at an indeter- visual quality of texture, and that this is .
minate distance. the stimulus correlate of a visual surface.

A second type of order would be one con- The varieties of texture in experience are
taining a single step or jump, such as innumerable, of course, but the varieties
1111ddddd or dddd11111. This order may of a cyclical order of elements could be
occur along one or both dimensions of an equally enormous in number. With only
array of elements and, when it does, what the two elements / and d, there are many
we call lines or discontinuous areas will repetitive sequences possible; when all
appear in the visual field. Presumably the levels of intensity and wave length
this type of order is a stimulus correlate are taken into account the variety of
1111 dddr1 dddd1111 11111111 dddddddd cycles become incalculable. The as-
1111 dddd ddddl 111 11111111 dddddddd sumption is that the microstructure or
1111dddd dddd1111 dddddddd 11111111 texture of a visual surface is the pheno-
1111 dddd ddddl 111 dddddddd 11111111 menal correlate of some repetitive type of
for the margins or outlines which are the retinal stimulation. If physical surfaces
necessary conditions for seeing figures have regular structures peculiar to them,
and shapes.5 as wood, cloth, or earth have, the re-
A third type of order would be one gularity will be projected in a focused
similar to llddllddl, which contains a image, and this repetitive character of
cyclical or alternating change. It is a the stimulation, in turn, may well be the
reasonable hypothesis that when such an basis for the perception of a surface.6
order is found in both dimensions of an The three kinds of order just defined are
array of elements there will occur the hypothetical stimuli for pure visual ex-
tent, for outlines, and for surfaces in the
5
must be remembered that we are de-
It abstract. But we need to account for
scribing what first happens on the retina, not surfaces as they are seen in three dimen-
what might happen at later stages in the
physiological process of seeing. The oc-
currence of a margin or outline in perception is 6A
determined primarily by the step from light to striking illustration of this point has
dark but also of course by the subsequent been suggested by Dr. Leonard Carmichael.
events in the optic nervous system. The Many of the great classical painters, es-
latter may be guessed at from such phenomena pecially those Dutch painters who worked
as brightness contrast at a border and the in- with magnifying glasses and the finest of
hibition of one border in perception by an brushes, could simulate velvet, satin, the
adjacent border or a succeeding one, all of texture of flower-petals, and even the peculiar
which suggest some kind of a process of sheen of a drop of water on the flower by the
tt contour building." The significant experi- precise arrangement of spots of pigment. The
ments on this problem are described by microstructure of the paint was quite differ-
Bartley (6) in his chapter on visual contour. ent from the microstructure of the real fabric,
The stages intermediate between a true con- the real petal, or the real water-drop. What the
tour and a shadow penumbra have been painter could reproduce was the microstructure
studied by MacLeod (78) together with the ac- of the light reflected from these surfaces.
companying effects (contrast or constancy) on Qualities of lustre, softness, hardness, wet-
the areas separated by the contour or penum- ness, and the like are very clear in these
bra. Here also there is presumably some kind paintings. The analysis of visual texture will
of interaction between adjacent areas. be carried further in the next chapter.
sions, and for the background surface exemplified by the ground; the latter are
best exemplified by the terrain. For that characteristic of objects. The surfaces
it is necessary to consider a serial or of the physical environment and its parts
progressive order of elements in the retinal are either longitudinal, frontal, or some-
image, or gradients as forms of stimulation. where between these two extremes.
In Figure 21 the material surface AB is
a longitudinal surface, and the surface BC
Retinal Gradients as Variables of Stimula- is a frontal surface. In the retinal image
tion .Texture ab, there exists a gradient of texture
Consider for a moment the physical en- from coars: to fine, whereas in the retinal
vironment from which light is reflected and image be no such gradient occurs, and
which is projected on the retina. The the texture is uniform throughout. The
problem of distance perception has been diagram may be conceived either as a
reduced to the question of how we can cross-sectional view from the side (A B is
see surfaces parallel to the line of sight a floor or the ground), or from above (AB
(Figure 20). These will be called longi- is a wall to the right of the observer).
tudinal surfaces to distinguish them from The slant of a surface is something that
frontal surfaces, which are perpendicular we can see, and the surfaces of the
to the line of sight. The former are best visual world are in fairly good agree-
A
FIGURE 21. The Optical Projection of a Longitudinal and a Frontal Surface
ment with the surfaces of the physical must, in other words, be a stimulation
environment with respect to slant. More- gradient.
over, as everybody knows, a photograph We can now define a fourth type of
or a painting can serve as a good sub- order among the elements of a retinal
stitute for a physical environment in image. It would be a serial change in the
yielding a picture-world with surfaces length of the cycles of a repetitive order.
which seem to slant or confront us just An example might be ddddlllldddlllddlldl.
about as they did in the original. The If a repetitive order is the stimulus for
picture surface is flat, but we have all visual texture, this would constitute a
learned to neglect that impression and to gradient of the density of texture.
see an array of longitudinal, transverse We know from ordinary experience that
and slanting surfaces which make up the the texture of different surfaces may vary
"space" of the picture. from coarse to fine. The various grades
The retinal stimulus-variable which of sandpaper used by carpenters differ in
makes possible the perception of a just this respect. Figure 22 shows the
longitudinal surface must be a continuous same texture in various grades of density.
change of some sort in the image of that When the image of a single surface varies
surface. To the distance of the physical progressively in this way, it may be that
surface at successive points there must the gradient of density is an adequate
correspond a variation in the image at the stimulus for the impression of continuous
projected points. Then, as the image distance.
.
-°: ". _0%
differs progressively from point to point, order to verify this hypothesis a
a
In
a
the perceived surface can differ corres- program of experiments would be neces-
pondingly in its distance or depth. There sary, and a beginning on such a program
III
_,,
---'
.. .5
-' O
%.%%"%%%%%% a
I
0.0.0.*.O.M.0.$
O..
... .........amosalos.
....... " ." " "' ...
Seeas....%%%e%%%%*:*%54.
.
11.0.0%.11.0.0.0.0e0.0%.896.0.0%.0.0.0.12 01048.0
0
W. 011/
es
1001,61.
%%.%%* aa ...................
:..%: 9111
.
.......... ......................................e.
1..
* 4 OO, 0.151,0.0
. 0.0.0.O.M.0.0_4,
....0.6.4.10.6.00. moo,
:"6.0.4,:u:s.d.V18:6, or.oe:s:m.e...11.41.e.o.e.s.e..
.41sese.o.I.0.1114Vesssees*eseees.e.41%.*.e....e:e.0
4.4.0.11...0.11..0.8.0. ellessessesssisese
%Os IV 11. (81
11.1111111
s Ss
1106e
. a Is eak
......:.::...:......,........o..
..
0.0.11.15.11.0.0.1b.O.O.O.0...0.4.11.m.
............................................... ., %60104%. slime

se , a . . a
a f.'s"
....................................666
..................................................... to
ea
.....e........:% isees e .. Is
%.1Vesseesse%%%.%/41...
...6.8....
. .11 I
.S......e.S668a908°..........., .S.55595.
.r.s.s.........o.s.e6e.e.....e.s.e.
. ea ....
IN. AI
a 111Papep°585111:: 111aaaanasaIllaaalga
a
*%%...%%%!VesuWe
S
. ...ease....Ves
.............v.
es*:
limo. ovese.e...e.
...0.0.0.0.8.090.9.16.11.090.0.1.11.0%
a s
11 11.ppme,*
0 eow.m.,,,, /
es
."5595.%%54.545aVe%*,. ID 4111 "" "" ''' "" ''' '*- -"' "" "'" -
.. ... ................................oh
.
...
..*.-.*. es's°.
',..
411 e
S __11_0_0_0_0_8_0_00sessesssessise
:.0.0.0:0:0:0:0.5665.11.505.00.0:11::::*".:
-.
.. .
.
:.%5:5,5556%YelleSes.590,6_111
OS 110.6.O.Opy.ssessseeses
..S.S.S.U.e.".'.:'.6..*. %iv...*
_.__.. _. 0 .5.5.5.0*..........__
6 aO.0.0.0.15.860 eYeseese.... * e' N.
a , 5555s555 aaaaaaaaaaaa
FIGURE 22. Grades of Texture

k
4
,r
710
w.
FIGURE 23. Gradients of Natural Texture and

the Resulting Impression of Continuous Distance
FIGURE 24. Texture-Perspective and the Impression of Receding Surfaces
Courtesy of Professor R. B. MacLeod
.41
I1 4A
If 4'14 ,I*4141` IPOwler.14".%; ?
P.,641,141`
.4 A, "A144 .""r\.;:r.
rd
kl,WYk,
;;,
o430 Ato "I.Mb

at'
will be described in subsequent chapters.

The hypothesis can be illustrated in a
preliminary way by examining pictures --
with respect to the impression of depth and
distance which they yield. Figure 23
shows two examples of textured sur-
faces found in a natural environment in
which a gradient from coarse to fine runs
from the bottom to the top of the picture.
Although the elements of the texture in
the two cases are of different shape and
mean size, the gradients in both pictures
are similar. In both pictures there ap-
pears a continuous increase in the,visible
distance of the surface. The impression
of a ground extending away from the ob-
server is fairly compelling. In Figure 24,
many different gradients of texture-density
are combined to yield a complex scene.
Half a dozen different kinds of texture
FIGURE 25. A Gradient of
are visible in the photograph. Artificial Texture and the Im-
These photographs represent surfaces pression of Continuous Distance
which are familiar in everyday vision.
Although the gradient of texture is the in the others. This result suggests that
only noticeable variation to be discovered the gradation of texture elements, not the
in them, they are interpreted by most ob- familiarity of elements, is the principal
servers from cues present in the picture cause of a depth-impression. The last
and are given a meaning. The meanings picture may also be interpreted as a level
usually assigned to the upper pictures terrain extending off to the horizon, but
are a ploughed field and a field of growing there are no actual cues for such a mean-
alfalfa, which are correct. It is pos- ing, and we may conclude that the im-
sible to suppose that the interpreta- pression of distance is an immediate pro-
tion is the cause of the depth-impression. cess, while the interpretation follows
Such would be the explanation given by upon it. "Immediate process" does not
an empirical theory of space-perception. imply an innate intuition of distance; it
Figure 25, however, was constructed arti- only implies that the impression of dis-
ficially out of line-segments, with a tance may have a definable stimulus just
gradient of lines and gaps decreasing as the so-called "sensations" have.
toward the top of the picture. The im- The line segments of Figure 25 were
pression of a surface extended in dis- not drawn so as to fall one above the
tance is clear in this picture as well as other in straight lines converging to the
horizon, but were instead offset. Aligning similarly, the gross size of the texture
them would have induced the familiar ap- elements on the retina will vary depending
pearance of linear perspective. The on whether they are predominantly sand,
gradient of texture is not the same thing as grass, bushes, or trees, and also on
ordinary perspective, although the two are whether the observer is flying an air-
united by underlying principles as will be plane, perched on a telephone pole,
shown. The projected size of things in standing, or sitting on the ground. What-
the environment does decrease as their ever their size may be, however, they
distance increases from the observer and diminish to zero in a gradient up the visual
as their size approaches zero or "van- field.
ishes" at the horizon. In this respect the The hypothesis implies that a gradient
texture of a longitudinal surface and the of texture in the visual field corresponds
perspective of objects are alike. But the to distance in the material environment
former leads us to a general phenomenon, on the one hand, and to distance in the
of which the latter is only a special case. visual world on the other. If true, this
The artificial texture of Figure 25 might principle should apply not only to dis-
have been drawn with the line segments tance-perception on the ground, in aerial
at the bottom of the picture twice as long and out-of-doors space, but also to dis-
as they are and the line segments above tance-perception in the civilized spaces
also twice as long all the way up to the of rooms and other manrmade surfaces. In
level where they diminish to zero. In order to apply the principle,. we need to
other words the horizontal dimensions, remember the types of surfaces already
but not the vertical ones, could have been distinguished: longitudinal, frontal, and
proportionally increased. The resulting slanting, with respect to the line of sight.
impression of distance on a surface, how- Gradients of texture on man-made surfaces
ever, would have remained as strong as may decrease, but the texture does not
before. The only change would have been diminish to a zero limit as that of the
a faster rate of decrease of the line seg- terrain does. On such bounded surfaces,
ments from the bottom to the top of the the rate of the gradient is a function of
picture, or a larger angle of convergence the slant of the surface. A gradient of
of the theoretical lines connecting their texture may decrease rapidly, slowly, or
ends (linear perspective). So long as the not at all, and these are the three respec-
elements approach a vanishing level at tive conditions for a longitudinal, a slant-
the top of such a picture, the impression ing, or a frontal surface. The texture of a
of a sort of disembodied terrain is the surface faced directly does not change
result. An increase in the gross size of from coarse to fine, and correspondingly
the lines suggests an impression either an unchanging texture gives the impression
of larger texture elements or of viewing of a frontal surface. When there is any
the terrain from a lower position, down gradient of texture, it may decrease up-
near the ground (Figure 61, Chapter 7). wards, from left to right, right to left, or
In perceiving distance on a real terrain, downwards, and these are the four respec-
1.
FIGURE 26. A Gradient of Density in Four Differ-

ent Directions, as Compared with an Even Density
tive conditions for a floor, a left-hand The Cues for Distance as Stimulation Gra-
wall, a right-hand wall, and a ceiling. ents.
These rules are illustrated in Figure 26,
where an artificial gradient has been con- The historical origins of the traditional
structed in each of these four directions. cues for distance have already been dis-
It can be compared with the similar figure cussed in Chapter 2. The accepted list of
beside it which lacks any gradient and these criteria or signs usually includes
where the surface, insofar as a surface is the following:
represented, appears to lie in the plane of 1. Linear perspective.
the picture. 2. The apparent size of objects whose
If the slant of any plane surface, such real size is known.
as a floor or wall, has a unique gradient
of texture, then the changing slant of a 3. The relative apparent motion of ob-
curved surface or one with edges, such as jects as the observer moves his head.
an object possesses, should have a unique This is often called motion parallax.
change of the gradient of texture. It 4. The covering of a far object by a
therefore seems possible that a change of near one, or the superposition of one con-
gradient may be a stimulus for the im- tour on another produced when one object
pression of depth and relief in an object. "eclipses" another.
5. The change in color of distant ob- problem of distance and suggested a theory
jects, to which is sometimes appended the of texture gradients, these factors in
loss of sharp outline and detail. This is depth perception must be re-examined.
called aerial perspective. For when they are considered as variables
6. The degree of upward angular loca- of perception rather than as facts of knowl-
tion of the object in the visual field, the edge once we understand that they ap-
ground and skyline being necessarily imply to an array of objects in the visual
plied as the background. field rather than to a single object they
may all prove to be gradients of stimula-
7. The relative brightness of the ob-
tion, or related to such gradients.
ject. This has been conceived by some
writers to be an inverse indicator of its Linear perspective, for example, might
distance; optically, however, this is be a special result of the decrease in size
based on a misconception. It is sometimes of figures in the visual field from the lower
mistakenly assumed that the more distant margin to the horizon.? Motion paral-
an object is in the ordinary environment lax, as seen from a train window,
the lower will be the intensity of its reti- might be a special result of the gradients
nal image, but this principle applies only of deformation which fill the visual field
to point-sources, not to reflecting surfaces. when the observer moves. Superposition
of one shape on another is best under-
8. The relation of the lighted to the stood by analyzing the outline between
shadowed areas of an object, or shading. them, and this outline may prove to in-
This is an indicator or sign, not of dis- volve a step separating two continuous
tance but of the depth or relief of a single gradients. Aerial perspective is, for the
object. most part, a simple gradient of hue in the
The "secondary" signs listed above visual field, a gradient running toward
have traditionally been considered less the violet end of the spectrum. The
important than the "primary" signs of shading on a curved surface is obviously
distance and depth listed below: a gradient, as every artist knows. It
9. The disparity of the binocular would be surprising but significant if
images of the object as a cue to its depth, retinal disparity, like the other signs of
and the relative disparities (crossed and depth, could be defined as a gradient of
uncrossed) of different objects as cues to stimulation not of the single retina, it
their relative depth. is true, but a gradient of the theoretically
10. The degree of convergence of the combined images of the two retinas. A
eyes on a fixated object, the convergence visual field obtained with both eyes open,
being inversely related to its distance. as we shall see, always contains a gradi-
11. The degree of accommodation of
the lens for a fixated object necessary to Linear perspective is also a geometrical
maximize the definition of the image. technique for drawing the edges of straight-
sided objects, but the two meanings should not
6i we have now reformulated the
4 001,.. 4 In
be confused.
ent of "double images." Finally, the cue the slope of a road should end at a cliff
of upward height in the visual field be- it is properly thought of not as a gradient
comes intelligible in the light of a "ground- but as a step or discontinuity. These con-
theory" instead of an "air-theory" of cepts appear to be admirably adapted for
visual space, inasmuch as the ground em- describing the retinal image, since both
bodies and is a precondition for the gradients and steps of stimulation can be
gradients mentioned. found within it.8
These possibilities for a general theory According to the evidence of C.M.
of visual gradients are, at this stage, Child and his students (22), all living
merely claims. They may appear plausible tissue is characterized by physiological
after they are developed. If they are to gradients. Along the axes of an organism,
be convincing they will have to be de- from head to tail, from front to back, or
monstrated, and the attempt to do so will from the apex to the base of a limb, there
be made in the next two chapters. exist gradients of metabolism, excitability,
9/land growth. Now these gradients of acti-
The Concept of Gradient vity are not merely spontaneous self-
The word gradient means nthing more '41' generated phenomena but are also reactions
complex than an incteasesrder.r.oe-st.--Di of the living cells to their environment.
something along a given axis or dimension. Although conditioned in part by the genes
As such it is related to the plots or within each cell, these reactions are
curves of analytical geometry. The gra- primarily determined by differentials of
dient of a railroad or highway, for example, temperature, light, chemical concentration,
is its change of altitude with distance. or electrical activity that is to say, by
This change may be positive or negative gradients of these kinds of energy. The
or zero (the last being a level gradient), proposal that the light-sensitive cells of
and it may also be rapid or slow, cor- the retinal mosaic and the neural tissue
responding to a steep or a moderate in the brain connected with them can react
gradient. The gradient may itself change, to gradients of stimulation, therefore, is
as the slope of a road does in hilly coun- not without analogy in other kinds of
try. When the change is very abrupt if organic tissue. The special application
81n
Bartley's work on vision (6), he has gradients of luminous intensity. The cues for
used the term gradient to refer to a change in the depth or slant of a surface, on the other
the luminous intensity of stimulation at a hand, seem to involve macrogradients over a
border within the retinal image. He is thinking considerable dimension of the retina. A
of a microscopic shadow-edge as it falls on gradient of the density of texture would be one
the mosaic of retinal cells which can be con- case. A gradient of shading in the hollows of
sidered a gradient since the change must be a surface or the shading toward the unlighted
distributed over the width of a number of cells. side of a curved object would be another.
This is what was called a step above. It The penumbra of a shadow is such a gradient
might be termed a microgradient as distin- according to MacLeod (78), and he has demon-
guished from a macrogradient. Visual con- strated with "artificial penumbrae" the different
tours, visual acuity, and the elements of effects of a steep gradient as compared with a
visual texture all seem to involve micro- gentle gradient.
texture all seem to involve micro- gentle gradient.
visual
visual acuity, and the elements of a

effects of a steep gradient as compared with
tours,
from a macrogradient.
guished Visual con- strated with "artificial penumbrae" the different
might be termed a microgradient as distin- according to MacLeod (78), and he has demon-
is what was called a step above. It The penumbra of a shadow is such a gradient
This
over the width of a number of cells.

distributed side of a curved object would be another.
sidered
a gradient since the change must be a surface or the shading toward the unlighted
the mosaic of retinal cells which can be con- case. A gradient of shading in the hollows of
microscopic shadow-edge as it falls gradient of the density of texture would be one

of a on
border within the retinal image. He is thinking considerable of the retina. A

dimension
intensity of stimulation
the luminous at a hand, seem to involve macrogradients over a
term gradient to refer to a change in the depth or slant of

used the a surface, on the other
Bartley's work on vision (6), he has gradients of luminous intensity. The cues for
81n
try. When the change is very abrupt organic tissue.

if The special application
as the slope of a road does in hilly coun- not without analogy in other kinds of
gradient. The gradient may itself change, to gradients of stimulation, therefore, is
responding to a steep or a moderate in the brain connected with them

can react
and it may also be rapid or slow, cor- the retinal mosaic and the neural tissue
or zero (the last being a level gradient),

proposal that the light-sensitive cells of
gradients of these kinds of energy. The
of a gradient theory to a sensory surface forces, in analogy with a gravitational or

such as the retina or the skin has not electrical field, has found application not
yet been attempted, however, and the ways only in physiology but also to problems
in which neural tissue may differ from of visual perception and of goal-directed
other kinds of tissue in this respect is behavior (68, 74). Child argues, however,
not known. Is it suggestive for the that in physiology the field-concept as
psychologist's problem of the physiological such is vague, and that a field theory is
correlate of visual form that biologists useless without analysis. Such analysis
have found the concept of gradients useful can only be carried out in terms of the de-
in understanding the development of finable and measurable gradients which
organic form? constitute the field. The writer agrees
Child has pointed out (22, p. 275) that with Child in this criticism. Field theory
physiological gradients may overlap and in psychology, as practiced by Gestalt
combine geometrically within the organism psychologists, is not always rigorous or
to yield what could be termed a physio- precise. Assuming that a field is deter-
logical "field." The concept of a field of mined by its gradients, an analysis of
400 500 600 700 800
WAVE LENGTH OF LIGHT
(IN MILLIMICRONS) I I
I I
I I I
I I
Y VI
UE Of SENSORY IMPRESSION I
VIOLET BLUE GREEN YELLOW ORANGE RED
0 25 50 100 200 500 1000 2000 4000

REQUENCY OF AIR WAVES
(IN CYCLES PER SECOND) 1
1
1
1
I
1
1 1 1
1
TONE OF SOUND 1
, 1
V
1
Y
1
V
1
Gi G gi cil CHI
(IN PIANO SCALE) 9
...-----,----------,
NORMAL TEMPERATURE OF SKIN
12° 33° 45° 52°

0° 20° 40° 60° 70° 80° 90°
EMPERATURE OF SUBSTANCE
1
a I I I I I
I 1
I I 1 I I I
IN CONTACT WITH SKIN li A A A A
(IN DEGREES CENTIGRADE)

1 1
I I
I 1 I
1 I I I
i 1 1
Y Y V V
QUALITY OF IMPRESSION
PAINFUL " --COLD AND COOL ---1.- -4-- WARM --0- -4- HOT -a-4 PAINFUL
..--......,
NEUTRAL
WEIGHT OF OBJECT 0 I Gr 10 Gr. 100 Gr I Kg. 10 Kg.

1
PRESSING ON SKIN 1 1
(IN GRAMS) I 1
I
I
1 I
INTENSITY OF IMPRESSION
A CONTINUOUS INCREASE IN "HEAVINESS" OR "PRESSURE" WITHOUT QUALITATIVE CHANGES OR NAMED POINTS OF REFERENCE
FIGURE 27. Examples of Psychophysical Correspondence
the stimulation gradients involved in per- to be found in Stevens (101), and the
ceptual (and possibly in behavioral) background of this problem is given by
fields would probably be more profitable Boring (10).9
than further attempts to discover the laws A few pairs of corresponding points are
of field-phenomena as such. indicated by dotted lines. It is note-
The Concept of Psychophysical Corres-
worthy that, for some variables like tem-
perature and others like weight, the
pondence.
correspondence of sensory qualities to
The correspondence of the variables of their physical variables may be shifted by
perception to the variables of stimulation adaptation. For example, after holding the
is exemplified in Figure 27. Four pairs of hand in warm water, a stimulus which
such variables are given. The lower formerly felt warm now feels neutral and a
line of each pair represents a variable of stimulus which formerly felt neutral now
experience. Each line is to be regarded feels cold. The correspondence has been
as continuous. Points on the upper line displaced, but it is still a specific and
represent possibilities or instances of regular correspondence (38). It is reason-
stimulation, and points on the lower line able to suppose that the spatial qualities
represent descriptions or judgments of the of the world, as well as the "sensory"
ensuing sensory impression, but these qualities illustrated above, may undergo a
"points" are not isolable. They cannot shift in their correspondence to stimula-
be thought of as stimuli and sensations tion without a destruction of the corres-
respectively; the points are simply pondence. Something of this sort probably
numbers in a serial order. The variable occurs in the process of getting adapted
of physics and the variable of experience to eyeglasses.
in each case are in a one-to-one corres-
pondence. In terms of the geometrical
Summary
model, for every point on the upper line
there is one and only one point on the A theory of visual space perception has
lower line. The lines (or numbers) need now been outlined. Its strength or weak-
not be conceived as scales possessing ness can be estimated better if its postu-
units of length. For present purposes lates are made clear. It may be useful,
they are continuous series merely. An therefore, to summarize the theory in a
introduction to the problem of scaling the series of propositions.
variables of physics and of experience 1. It was assumed that the fundamental
the "dimensions of consciousness" is condition for seeing a visual world is an
9
Boring has also discussed the seeming sions of the stimulus as those variables of
paradox (exemplified by auditory "volume") stimulation, however mathematically complex,
that there may exist more dimensions of with which the variations in discriminative
sensory experience than there are simple response prove to be correlated as the outcome
dimensions of the physical stimulus (11). The of a psychophysical experiment.
difficulty is resolved if one defines the dimen-
array of physical surfaces reflecting light 6. The perception of an object in depth

and projected on the retina. This is in is reducible to the problem of the changing
contrast with the usual assumption that slant of a curved surface or the differing
the problem of perception should start slants of a bent surface. In either case
from the geometrical characteristics of the problem is similar to that of how we
abstract "space." see a longitudinal surface.
2. In any environment, these surfaces 7. The general condition for the per-
are of two extreme types, frontal and ception of a longitudinal or slanted sur-
longitudinal. A frontal surface is one face is a kind of ordinal stimulation called
transverse to the line of sight, and a a gradient. The gradient of texture has
longitudinal surface is one parallel with been described, and it has been suggested
the line of sight. that gradients dependent on outlines, a
3. The perception of depth, distance, gradient of retinal disparity, a gradient of
or the so-called third dimension, is re- shading, a deformation gradient when the
ducible to the problem of the perception observer moves, and possibly others, all
of longitudinal surfaces. When no sur- have the function of stimulus-correlates for
face is present in perception because of the impression of distance on a surface.
homogeneous retinal stimulation, distance
is indeterminate. Although the ground is Conclusion
the main longitudinal surface, the walls
The correspondence of the visual field
and ceilings of man-made environments
to the total retinal image is an anatomical
constitute three other geometrical types. point-for-point correspondence which is
4. The general condition for the per- not hard to understand. The 'correspond-
ception of a surface is the type of ordinal ence of the visual world to the total retinal
stimulation which yields texture. image is an ordinal correspondence which
5. The general condition for the per- is more difficult to analyse and specify.
ception of an edge, and hence for the per- But the latter correspondence is no less
ception of a bounded surface in the visual literal and exact, we may believe, than the
field, is the type of ordinal stimulation former, and it is clear that the way to de-
consisting of an abrupt transition. The termine it is to find the obscure variations
simplest and best understood kind of of the projected image which yield co-
retinal transition is one of brightness. ordinate variations in perception.
The Stimulus Variables for Visual Depth
and Distance -- Momentary Stimulation
The Stimulus Gradients of the Density of Texture

and the size of Objects . .. . The Depth-Shape
of ObjectsGradients of Texture and Grades of
Illumination . . . . The Stimulus Gradient of
Binocular Disparity Between Images . Defini-
tion or Resolution in the Retinal Image . . . . The
Gradients of Aerial Perspective . . . . Summary
Consider once more the way in which the Let us make two assumptions about the
physical world is projected on the retina typical physical world of ground and ob-
of the eye, remembering that a typical jects and assert that, first, objects tend
scene consists of the ground and of ob- to be in contact with the ground instead of
jects (and remembering also that the image up in the air and that, second, they tend to
is inverted). Near objects will be imaged be distributed over the ground with an even
large and high up on the retina. Far ob- scatter. The first assertion will probably
jects will be imaged smaller and lower meet with no objections. As for the
down. Very far objects will be imaged so second, it can be proved that the physical
small that their size approaches a vanish- spacing between many kinds of things
ing point. At the line where the earth tends to be regular. The principle holds for
ceases and the sky begins, the separate grass in a meadow, for trees in a forest,
images of objects become indiscriminable. for the boards in a floor, and the patterns
This is the fundamental world for vision. of a carpet. Above all, it holds excellently
The rule is that those parts of the world for a special type of object possessing
just under a man's nose are projected the greatest importance for vision the
large and those parts at a distance are single element of the texture of a sur-
projected small. In the visual field, the face. The grain or structure of a sur-
patches and spots of color are gross and face is made up of units of one kind or
far apart at the lower boundary of the field another which are repeated over the en-
and become progressively smaller and tire surface. These units are character-
denser upwaids towards the horizon. istic of the physical substance in ques-
77
tion. They determine an array of light of motility in the image as the corres-
reflections from the surface which can be ponding objects and elements recede. If
focused as an image. The number of these both eyes are functioning there will be a
units which can be counted in any single gradual change in the disparity of the
square inch (or yard or mile) tends to be elements of one image relative to the other
the same as the number in any other square as the corresponding objects recede.
unit of measure, and this is what is meant There may be still other changes in the
by an even spacing of the units. retinal image corresponding to the physi-
It is of course true that the spacing of cal recession of the environment, but
either particles on a surface or things on they all presuppose a textured image.
the ground is seldom perfectly regular The gradient of size and density, there-
except in the case of mechanically pro- fore, is a necessary correlate of reces-
cessed surfaces or man-made layouts of sion. This condensation of the image
things. Tiled pavements, planted fields, cannot be eliminated by holding the head
telegraph poles, and railway tracks are motionless or by closing one eye. It has
examples of exact spacing. In this a special status, and consequently it is
special case, as we shall emphasize, the this gradient which should receive first
retinal image and the resulting perception consideration.
can be fitted to a set of special rules, but The purpose of this chapter and the
this fact must not distract us from con- next will be to consider, one by one, these
sidering the importance of natural distri- various so-called cues for distance per-
butions. 1 ception when they are reformulated as
Our hypothesis is that the basis of the gradients of the retinal image. In this
so-called perception of space is the pro- chapter, for the sake of simplicity, the
jection of its objects and elements as an retinal image we shall refer to is the
image, and the consequent gradual change image obtained by a motionless ob-
of size and density in the image as the server with his eyes fixed straight ahead.
objects and elements recede from the ob- In the following chapter we shall go on to
server. Whenever the observer moves hi:3 consider the image obtained by a moving
head there will also occur a gradual change observer, and only then attempt to under-
stand the image of an observer who scans
'The structure of substances at microscopic his environment the image which samples
and sub-microscopic levels of size is studied
by crystallography and physical chemistry. a different cone of light-rays from one
The structure of the universe on the scale of moment to the next and which according-
miles and light-years is studied by astronomy. /
But the structure of the world on the scale of ly registers in succession different sectors
millimeters and meters the textures of sur- of Iihe physical world.
faces and the distributions of objects is so
familiar that it has been very little studied. T e Stimulus Gradients of the Density of
The eye and the retina ate adapted to register
the structure of the world only at this range of xture and the Size of Objects
sizes. Finer and grosser structure can be
tt seen?! only by the use of special devices Illustrations of a gradient of texture
such as microscopes and telescopes. have already been given in Figures 23 and
STIMULUS VARIABLES MOMENTARY STIMULATION 79
25 in the preceding chapter. What is now One might suppose, therefore, that the
required is a closer study of this little existence of a gradient in the retinal
understood form of visual stimulation. A image could be verified only by getting at
method of isolating and varying texture the retinal image itself in some way and
Is needed. Its implied relation to linear measuring it.
perspective should also be explored, to- Actually, the measurement of a retinal
gether with its relation to line or contour. gradient is not necessary, and the measure-
But first we need to consider how a gradi- ment of its corresponding plane projece
ent of texture on the retina can be arti- tion may be substituted for it without
ficially produced and hence experimentally error. As Figure 28 illustrates, they are
controlled. in a perfect point-to-point correspondence
The Relation of a Picture-Plane to its with one another, and the one arrangement
Retinal Projection. The obvious method of may be mathematically transformed into
constructing a gradient of texture would the other at any time if the dimensions of
be to draw it on paper and present it to the the retina and the picture are known. It
observer's eye in the position of the pic- is obviously much more convenient to
ture-plane, so that it will produce a retinal specify the retinal distribution of light
gradient of texture. This is the method em- on a picture-plane than it would be to
ployed in the exploratory experiments specify it on the curved surface of the
described later in this chapter. Is it a retina itself. Moreover, the plane projec-
legitimate method? The retinal image of a tion is more readily compared with the
given object is not a plane projection of experienced visual field and hence is
that object, such as a drawing or a photo- easier to conceptualize. We will therefore
graph would be, inasmuch as the retina is speak of the plane gradient as a visual
a curved surface. The retinal image, stimulus, recognizing that the picture
moreover, is inverted relative to the plane gives no more than a convenient substi-
projection; it is not actually a picture. tute for the retinal gradient.
HORIZON LINE
... 1.
.., ..., ...,
%., ''... -'s .......
N.
.....
\ .....
s.,
...,
,...
...,
....,
,. \
,..
., .,
A
\\
``..
N.
\ .....
\\
\,
N
N
N.
\ ..,
'...
..,
-...
.....
-,
N .,
\A' IMMO UMW
N
--,
soma ammo
-....
Ilho Immo .inwo memo low

-... -,
mai 11111.
EI
FIGURE 28. A Picture-Gradient, its Retinal Gradient,

and the Longitudinal Surface Corresponding to Both
The method of drawing the elements of magnitude and shape of the typical crest
a texture gradient on the picture plane is and trough. If the surface is smooth but
undoubtedly a crude one. Techniques of not chemically homogeneous if it is com-
photography, photoengraving, and of con- posed of different substances the re-
trolled optical distortion offer other flectivities of the different particles are
methods of producirg and varying texture likely to differ. An example would be
which ought to be exploited. polished granite, or any conglomerate
The Elements and Gaps of Visual Tex- material. In either event, whether the
ture. There are, as theatrical magicians reflecting particles are structural or
know, some physical surfaces which are chemical or both, they will reflect light
invisible. The test for whether a physical differentially and the image of the surface
surface does or does not possess visual will consist in an array of cyclical changes
texture is whether the surface can or can- in light energy which we experience as
not be brought into focus by a lens, that variations in brightness or hue. The
is, whether an image of the surface can be optical image, it must be remembered,
produced. It makes no difference, theoreti- implies a correspondence between two
cally, whether one uses an eye or a camera sets of abstractions, reflecting-points and
for this test. Perfectly flat transparent focus-points, such that the character of
surfaces or perfect reflecting surfaces, the light at the former is duplicated at
such as glass, cannot be focused on by a the latter, point for point. The structural
camera or accommodated for by an eye in and chemical cycles of the surface, there-
the absence of any highlights or luster. fore, are projected on the retina as cycles
Neither can any surface when its illumina- of color in corresponding order.'
tion is sufficiently low. It is no more
These cycles, we suppose, constitute
possible to get an optical image of a
the stimulus for visual texture. Both the
sheet of plate glass or a large mirror (if cycles and the resulting texture can be of
highlights are absent and the edges of the
many different types, such as the rippled
surface are not in the field) than it is to
surface of water, the complex roughness
get an optical image of the cloudless sky
of a plaster surface, or the regular units of
or the interior of a completely blacked-out
a grating or a tiled pavement. The type
room. Surfaces of this type are fortunately
of texture to be tried out in experiments
not the surfaces on which we walk and sit
should be as simple as possible and at
and which characterize the objects of our
visual world. 2The
Ordinary surfaces are rarely both assumption is that a texture can be
analysed by plotting it in two dimensions,
physically smooth and chemically homo- that is, by specifying the alternations or
geneous, like plate-glass. If the surface repetitions of light stimulation along two axes.
This is what is meant by cycles. Admittedly
is rough, it has crests and troughs. A this assumption needs mathematical study. A
piece of cloth, a ploughed field, or a hilly texture cannot be analysed conveniently in
terrain seen from the air are all alike in terms of lines, nets, grids, or other patterns
with which the writer is familiar because
this respect except for the difference in these are themselves special cases of texture.
Mr, 4r Mr).
" '71441/4: 7A4-4's
r,54,,,,
:,!,,19, `, ' ': 4.::',Z44.0in:% '"I4'..10Ak ','

,,j
4, ,
A, ,,',,t. 0$ ,i
,',L,W
,,..41,0,,,,, a. ,.,,..,.',,c<
4.
- ' ii.,,,
4 'r'''
4, A6,,4 ' , ' .,,,,,, r''', .",4 r ,,,' 'am , ir,,,,," 'Y''
, * ',,,,r, , $: : ^ ,,,` 1, ?T'
,
1:111P4 1;
,1 9 n S10
46"
, ,trt"P
rr'n. 414 kr '41
41
64,""tAT
,1446., 1
4, -
417,
4;A
,A, 54
A ''',...
' 4,-
.. '
r *
1'444:
54 tn.
'i
fr-b,''''',', 1
-srm,
40,44,04,00,
. .- ,
..t, -,r:1.,-'
'1
-f jj KR , 0
.
i. ,... . -- 4
4 UR .4 ---4
..c.i .,,,k,,,
,
........ .
. 4114.-
r.
^-4 -4"4
i 9,.
,0,
,.
-trlf
...,
rt
1 1
ennntr:k
1611r,,
' 4
0
,
ntA
IS
tve
,1 644'41141411/1
111.11N0.44...
44.
FIGURE 29. Some Examples of Visual Text
the same time unfamiliar or abstract. The geometry. The principles governing such
kind of surface which can vaguely be a projection are illustrated in Figure 30,
termed "spotted" might be the best to which is itself a perspective drawing.
begin with. Its texture could be defined The surface projected in Figure 30 is a
more accurately as consisting of elements series of adjacent squares, like a side-
and gaps. The elements correspond to the walk, extending from the eye of the observer
spots and the gaps to the areas between to the horizon. We may think of the
them. If the elements are black and the squares as unit areas of surface. The pro-
gaps white, the texture can be artificially jection is a series of adjacent trapezoids
constructed with a pen and ink. The diminishing gradually to a point, so that
cyclical character of texture can be pre- the whole constitutes a triangle. Consider-
served by making the size of the elements ing the width of each projected square one
a constant ratio of the size of the gaps. notes that it decreases upward on the pic-
Several types of abstract texture can be ture evenly on the scale of the picture.
produced by arranging the elements and This linear decrease in the width of
gaps in any desired gradient on the pic- pictorial things is what makes possible
ture-plane with respect to their size. We the techniques of drawing guide-lines
can then determine whether the gradient which converge to a vanishing point.
produces a corresponding impression of a What is popularly known as "linear per-
longitudinal surface in experience. spective" is comprised by this fact.
The Method of Drawing Gradients. The Considering the height of each trapezoid,
projection of a longitudinal surface on a however, one notes that it decreases up-
picture-plane is obtained by perspective ward on the picture with a gradient which
FIGURE 30. The Principles of Perspective for Square Units of the Ground
STIMULUS VARI ABLES MOMENTARY STIMULATION 83
is not linear but negatively accelerated. of diagonals or a variant of it (116), which

The projected height of flat areas, some- requires the use of more than one vanish-
times called foreshortening, can be drawn ing point and is therefore not illustrated
only by a more complex technique than in Figure 30.
that of lines to a single vanishing point, The novel feature of this use of the
a technique less familiar but equally im- perspective techniques is that they are
portant in the practice of perspective. here applied to the texture of a surface
These two techniques enable the experi- instead of to the edges of a surface, as is
menter to construct on paper a gradient usual. Conventional perspective is the
of "flat" elements and gaps of texture perspective of the edges and boundaries
which will bear a specific relationship to of things and when a surface is to be
a material surface in the third dimension. represented only the outlines of the sur-
The decreasing projected size of a unit
object on the ground with increasing dis- FIGURE 31. The Perspective of
tance from the observer is subject to two Obiects is the Same in Principle
rules, one applying to the frontal dimen- as the Perspective of Texture
sion of the unit object and the other to its
Gendreau
longitudinal dimension. The first rule is gar....0011ggr
simply'the Law of the Visual Angle, which

states that a frontal dimension is pro-
jected as a size (S) which is the reci-
procal of the distance (D). Algebraically
this means that S is proportional to
Geometrically these frontal
D
dimensions
project so as to give a perfectly linear de-
crease up the plane of the projection, and
the ends may therefore be joined by straight
lines to a vanishing point, as illustrated. ,
The second rule states that a longitudinal H In

` 4.4
y uod,
dimension of the unit object is projected ,
as an altitude (A) which is a negatively

accelerated function of the distance (D).
Algebraically it comes out that A is pro-
-.
portional to L The longitudinal dimen-
D2
sion is the one which is said to be
foreshortened in the projection, that is,
it is the dimension which is compressed
relative to the frontal dimension. The
successive altitudes of unit objects can
be constructed geometrically by the method
gO. r o
r op as r
a! or or 0, or or or or ro
AID dm _ ... e .. ..., .... ft.

flb IP No WV ID MI lo MP a a
tib
40 lib 41,
fa 01) II
41,
416 fib
ID lb lb
0 0
FIGURE 32. Spot-Distributions which Yield Im-
pressions of a Longitudinal and a Frontal Surface
FIGURE 33, Spot Distributions Based on

Perfectly Regular Spacing of the Elements
IV
MD NM. or
NO.
.r.
m.
ww
.... or
I=.
op
vo.
oo
. .m.
or
.N. . .. ro
AMP
r.
or
N.
or
M
or
I OP
or
MO
or
.1.
maa Aft
I I
IM MD OIM OM. ea. =1 NW .11. MEW a. 411
ID Mb. nElb M MO. MIS MI OM
OD OM MP NW oll AN. a or aia 0 I 4

S OW S S S S
WO. +IA 5
I 4
S S alb 4110,
S I 4
alb S
I 4
410
411, 4111 I
r 4
-
NO alb S
4
D 4
AM.
STIMULUS VARIABLES - MOMENTARY STIMULATION 85
face are drawn. We require a perspective tance looked at from below a ceiling
of visual areas. scene. The retinal gradient is reversed
in the latter situation. There will be
Texture Gradients and the Impression of other instances later in which the re-
Distance. Figure 32 consists of two versal of a gradient yields the impression
different arrays of spots and gaps. In the of the inversion of a surface.
left-hand array, the width of the gaps is According to the theory formulated, the
governed by a linear decrease upwards distribution on the left should produce a
and the height of the gaps by a negatively longitudinal surface and that on the right
accelerated decrease. In the right hand a frontal surface. With respect to being a
array there is no decrease; the gradient is surface this prediction is not wholly con-
zero. The dimensions of the spots are de- firmed, since for many observers the
termined by the same rules as for the gradient of spots suggests an array of
gaps but, in the drawing, these dimensions objects or an invisible ground. Inter-
are only approximate. For both pictures pretations such as "lily-pads on the sur-
the presumption of a perfectly equal face of a pond" or "the heads of people
spacing of the spots on the corresponding in a crowd" or "disembodied cabbage-
objective surface would produce vertical patch" are frequent. To common sense,
and diagonal alignments in the pictures, an array of objects seems to have nothing
similar to the alignments observed in in common with a surface, and a stimulus
planted fields or wall-paper patterns. for the first could hardly also be a stimu-
Such alignments would produce the familiar lus for the second. Nevertheless a transi-
appearance of perspective (Figure 33). tion must be possible between an array of
In order to demonstrate that this special objects and an array of texture elements.
type of perspective is not necessary in a It is not always clear whether a given
texture gradient, a slight irregularity of scene is one or the other. A forest, for
spacing has been presumed such as to example, appears to consist of trees to
destroy vertical and diagonal alignments. the observer nearby but a surface to the
This was achieved in the drawing by off- observer at a distance or the flier at an
setting each successive row of spots to altitude. The significant fact is that the
the right or left in a random fashion. The same laws apply to both.
horizontal alignment of spots was left Figure 32 was composed of flat texture
undisturbed by this expedient. elements of the sort found on a physically
The left-hand array of spots gives an smooth surface. They might have been
impression of continuous receding space, spots on a floor, for instance. Many
while the right-hand array does not. The physical surfaces, however, are composed
space is, moreover, one on which the ob- of elements which do not lie on the sur-
server looks down from above. It is what face but project upward from it. A stubble
might be called a ground scene. If the field would be one example. Figure 34
pictures are inverted, the sensory impres- was constructed in the effort to produce a
sion of distance remains but it is a dis- synthetic gradient of texture of this sort.
The artist who knows how to see the

various textures of the material world
and knows how to reproduce some of them
will be able to think of a hundred inter-
relations and subtleties not even hinted
at. The drawings do suggest that the
theoretical-experimental approach of the
psychologist to such problems may
be illuminating, and they tend to verify
the hypothesis that a gradient of texture
is, in isolation, a stimulus for the im-
pression of continuous distance on a sur-
face.
Gradients of the Spacing between Edges

Straight Lines and Linear Perspective.
1 1 The impression of distance on the left-
hand section of Figure 36 is not sur-
FIGURE 34. A Gradient of Tex- prising since it is common in pictures and
tural Elements which Pro-
ject Upward from the Surface photographs stimuli to which we have
all been exposed since infancy. Floors
The elements, like those of Figure 25 in and pavements generally have rectilinear
the last chapter, are line segments but, joints between the parts which compose
unlike them, they ate vertical on the pic- them (boards or tiles, for example) and
ture-plane. The height of the elements these joints reflect light in much the
decreases up the picture in a linear same way as do the lines of a drawing.
gradient, like the width of the lines in One might think of them as in lines as
Figure 25. The width of the gaps be- distinguished from outlines and they may
tween the line segments of each row have the effects of producing a surface in
also decreases in a linear gradient. The perception just as texture produces a sur-
only non-linear gradient in. the drawing is face. It may be noted in passing that
in the height of the gaps, or the mean verti- both the inlines and the outlines of a longi-
cal distance between the bottom of the tudinal surface are projected on a picture-
lines. In this drawing, as in the last, all plane by the rules of geometry illustrated
observers see a plane of distance and earlier, We are, however, deferring all
also see the lines as perpendicular to consideration of outlines until we come to
this plane. the problem of objects.,
These drawings, including Figure 25 Surfaces composed wholly of such in-
in the last chapter, scarcely make a be- lines are represented in these three draw-
ginning at exploring the complex rela- ings. When their density increases to-
tions between visual texture and space. ward the top of the picture, the impression
or' r A r, aY,
1"1 "
,
4 '1 ' ' '
.4 or ,o ,4 rop
" rr, ' ,;11 Ir . r 'al ? , fl'ri
,
ur 4 II, 1 I, 1-4 : ;., 4`

I r1 l / ,
S: 4 1
rIJ S y 1,11,
'or
4, 1 r
r r7111,
41,
jrroj
,
A
,,441/441 '
'01r4i44e
14,414 01;,,;0t,,,s,lutt ' 44, '`.4141. 4,
"'";
44 4, ',''.1)N1')1
1'
ffa,shtl'1,,0
A
, kV.
rP S'
,
qA
''',A14040','
r40
4 '41ir '04 41,
11,4
4
,,,,ow a!, .4
FIGURE 35. Two Pictures with Varieties of Texture within the Same Scene
Courtesy of Ilarris-Seybold Company

A
11 3
-
Ft ,0:1,,,, ,';,':::,l'i2 :' :,
'11
4 , `,", .4,4,410*.. : , ' ly4 ' ., .,.
'' P'''. ''' ''

1 S,'. ilnr.!'In ,K.'r ;,..
/ ,
CiN,NITAP* <e, ' '' 1 .. .411.10111N:90?},k001,044410

s'
4,
41,1,1 1.1Ak14'4,,0004/ ,
4,04,:try,,0 ' ,1,;.,,,,L'g.:;:c4,"'Vtt,co,,;45io,,,,i,,,,",:;,,',s;:,,
4111,iSliWC44 (444.4,4' ,,,A ,t,ict,1,0 ;

3, ,' '? 1,', ,, '
.,',t',,*(ii11',14P 01 f 14 '0;4; V , q'''''tr 46" )

4:00.4,:,:i'
' '44,',.4(:1;110
' , ' OA; o,,i, A ' ';' 4d
, , Ir . ,, .. f.t ,
v 1 .., /
41'
4'11;04,2P.,
.11041 V'" ,°4');:..`,'
FIGURE 36. The Effect of a Width-Gradient, or Linear Perspective
is that of a longitudinal surface; when longitudinal surface like the ground.

the gradient is zero the impression is that That in the center yields the impression
of a frontal surface. The meaning most of a frontal surface. That on the right
frequently suggested by the drawing is yields an impression of increasing dis-
that of a floor made of boards, or (since tance at the top, but the drawing does
the texture of the boards is missing) an not have the kind of flat distance which
open grille of some sort. The direction the first has. The first drawing is deter-
in which the lines run within the visual mined by the specific gradient derived
field makes no difference for the impression from the perspective of a floor. The second
of distance; it is the gradient of density has a zero gradient of density, or the
which yields the sense of a continuous perspective of a wall. The third has
third dimension. In the right-hand draw- an increase of density up the picture-
ing, for instance, although the lines run plane, such as to give the perspective of a
off toward the right, their density in- curved surface which slants upward from
creases up the picture, and the surface we the observer and then away from him. Some
see is the same as in the left-hand draw- observers can see just such a curved sur-
ing. face in the drawing.
When the inlines of a surface run hori-
zontally across the visual field, as in The fact that a mathematically simple
viewing a board floor crosswise of the gradient corresponds to a geometrically
boards, the space between lines decreases complex surface in this case should not be
on the scale of the picture in the specific considered puzzling. All gradients be-
gradient longitudinalappropriate to come mathematically very complex when
dimensions on a plane surface. Figure 37 projected on the retina. The implication
consists of three drawings of this sort. to be noted is that a curved physical sur-
That "o the left yields the impression of a face may have just as specific a correlate
N11'
.ti WNlu s
('Ni
Jet<
.ti
88
,...1..1.010
FIGURE 37. Gradients of Vertical Spacing Corresponding to Three Types of Surface
when it is projected as a fiat physical classical rules of linear perspective. All

surface has. lines either converge to some vanishing
A flat ground surface composed of point on the horizon or are horizontal and
geometrically simple forms such as tiles parallel. Figure 38 is an example of such
or paving stones can be projected on a a surface. The units are the triangles
plane with none other than straight lines. formed by the diagonals of a square.
It is an ideal exercise for the use of the There are four sets of parallel lines on
FIGURE 38. The Perspective of a Pavement
89
the pavement, and they are represented imaged size of a given object not as
by three sets of lines to three vanishing merely something related to the other
points, plus one set of horizontal lines. sizes in a graded array but as something
The drawing illustrates the width and to be sensed in isolation and then inter-
height gradients which we have isolated preted. A given retinal image as such
in previous drawings, and gives some could give no evidence of the distance of
indication of the method of constructing its object, for the object might be either
them. The combined gradients yield a something small and near or something
vivid and compelling perception of con- large and far off. Such a confusion would
tinuous distance. actually occur in the empty space under
The physical composition of a surface consideration. Only if the real size of
like this differs of course from the physi- the object were known because the ob-
cal texture of most natural surfaces. In- ject was familiar could a decision be
stead of what we have called elements made between these possibilities. In
and gaps it is composed of flat geometrical that event the sensed size could be
forms which fit into one another. It never- compared with a remembered size and the
theless yields the same type of visual distance of the object could be obtained
perceptions as a surface of natural tex- by a kind of unconscious computation. A
ture and it may be considered as simply human figure could be perceived as 100
a special case of a much more general yards distant rather than 50 yards be-
phenomenon. Our conclusion is that a cause it is seen as a 6 foot man and not
gradient of inlines is a stimulus for con- a 3 foot boy. If the man were a midget,
tinuous distance on a surface as well as however, the perception might be erroneous.
a gradient of texture composed of ele- According to this theory the perspective
ments and gaps. size of an object is only a cue to its dis-
The Apparent Size of Familiar Objects tance in the sense that it provides a fact
as a Cue to their Distance. The classical upon which the mind can work. The theory
theory that distance is perceived by a mind of cues is obviously a very roundabout
seated in the brain and making use of the way of explaining distance perception as
cues presented to it by a retinal picture is compared with hypothesis, which
our
very different from the theory we are now suggests that gradients of size and
engaged with. While ours conceives an density, being in correspondence with
array of texture elements or objects on a physical distance, are stimulus correlates
substratum, the former conceives an for a continuum of seen distance. The
isolated object or two in empty space. gradient theory accounts for the distance
Instead of asking "How do we see con- of all objects in the array, rather than the
tinuous distance from here in all direc- single object on which attention is fixed.
tions?" the classical theory asks "How It would be too much to say that the
do we judge the distance of that object, inferring of the distance of a familiar ob-
or the relative distance of those two ject on the basis of its perspective size
objects?" Accordingly, it treats the never occurs in human perception, for
7........4000.....7 100 foot span exactly as large as a plane with a wing-

\ i span of 50 feet at 5,000 feet. If the
i silhouettes of the two were the same, no
\ / observer could tell one from the other by
I
vision. But if the shapes differ, and one
can recognized as a large heavy
be
bomber while the other is a smaller attack
plane, both being familiar from repeated
10,000 feet experience, the former is known even
seen to be high, while the latter is
50 foot span perceived to be low. This type of dis-
\ i tance estimation improves with training,
\ / and thousands of observers have had ex-
/
perience in it during the last war. The
/
\ point to be noted is that it is not the kind
5,000 feet\
of distance perception which occurs in
the everyday visual world.
\
I
The Depth-Shape of Objects Gradients
of Texture and Grades of Illumination
The problem of the perception of dis-
tance is not really separate from the prob-
lem of the perception of depth or solidity.
Granting that abstract empty space is
irrelevant to either problem, the impression
of a surface is the basic factor underlying
FIGURE 39. Two Airplanes
with the Same Retinal Image both the experience of space and the ex-
perience of objects. Space is the visual
this rational kind of judgment is the background of objects and, when it is de-
only one possible in certain circum- terminate, space reduces to the surface or
stances. Airplane spotters, for instance, surfaces which comprise the background.
can be trained to estimate the altitude of Objects are also defined by their surfaces.
the craft they must identify and report, But the visual surface of an object has
although their error is at best consider- some characteristic features which we
able. An airplane seen against the clear have not yet considered. It is necessarily
sky is not set into a background of con- either curved or bent in some way (this
tinuous distance; it appears in the closest is the depth-shape of the object) and it is
possible approximation to empty space. always delimited by a contour (the pro-
A plane with a wingspan of 100 feet at jected shape of the object). What are the
an altitude of 10,000 feet will be imaged stimulus gradients which might give rise
to a depth-shape, and what is the nature planes of objects we shall frive gone a
of a contour? long way toward accounting for their depth-
The Density of Texture and the Depth shape. A corner is geometrically the
of Objects. The first problem is to ac- intersection of two planes with differing
count for the perception of what has been slants. An abrupt variation in slant cor-
variously called relief or modelling of ob- responds to an abrupt change in the
jects in the third dimension. The tradi- gradient of texture-density. Figure 40
tional explanation has been the stereo- represents such a change, employing a
scopic effect of binocular vision, sup- width gradient for simplicity in drawing.
plemented by the mental interpretation of The change is from one rate of convergence
the shadows formed on the side of the to a slower rate, and the impression is
object away from the light. Little or no that of a corner concave toward the ob-
attention has been paid to the fact that server. If the change had been to a faster
the texture of the object varies in density rate of convergence the impression would
according to the laws of frontal and have been one convex toward the observer.
longitudinal surfaces (see page 66 ). The locus of the discontinuity also gives
Imaged or projected texture varies with the impression of a visual line even though
the slant of the surface, and with the no line is drawn. It may be thought of as
facing of the slant floor or ceiling, a kind of stimulus for a corner.3
right wall or left wall. The analysis of The contour line, at which one surface
texture-perspectivesuggests a general eclipses another, is often accompanied by
formula for bounded surfaces in the third the impression of a jump in depth whereas
dimension, whatever their inclination to the inline of a joint or corner is not. The
the line of sight. It is that the gradient abrupt variation in brightness or color,
of density in a projection of a physical which accompanies and ordinarily accounts
surface bears a fixed relation to the slant for a contour according to the suggestion
and facing of the physical surface pro- in the last chapter, does not account for
jected. If this principle holds for a sur- this stepwise depth effect. Perhaps an
face it holds for portions of that surface, abrupt variation in texture, however, can
and if the slant of a surface varies con- account for it at least as a contributing
tinuously or suddenly from point to point factor.
the projected gradient likewise varies Figure 41 is like Figure 40 in showing
continuously or suddenly. Here is a a width gradient. Instead of a change in
possible stimulus basis for the percep-
tion of a curved or bent surface.
An illustration has already been given of ;There are evidently different ways in which
a gradient in the density of texture which a line may be evoked in vision. One should
yields a curved surface, Figure 37. What not be too preoccupied with the lines produced
by pens and pencils. Besides the outline or
kind of a drawing would be expected to contour of an object on its background there
yield a bend or corner of a surface? If we are the inlines produced by joints within a
surface (Figure 38) and also, as we have just
can account for the curves, corners, and seen, by the corners of a surface.
curves, corners, and seen, by the corners of a surface.
can account for the
surface (Figure 38) and also,

as we have just
we
a surface? If
yield a bend or corner of
a
are the inlines produced by joints within
a
drawing would be expected to contour of an object on its background there
kind of
by pens and pencils. Besides the outline or
37. What
yields a curved surface, Figure
not be too preoccupied with the lines produced
which
gradient in the density of texture
a a line may be evoked in vision. One should
evidently different ways in which

;There are
been given of
An illustration has already
tion of a curved or bent surface.
possible stimulus basis for the percep-
continuously or suddenly. Here is a

a width gradient. Instead of a change in
the projected gradient likewise varies Figure 41 is like Figure 40 in showing
tinuously or suddenly from point to point

factor.
and if the slant of a surface varies con- account for it at least as a contributing
face it holds for portions of that surface,

can
abrupt variation in texture, however,
jected. If this principle holds for a sur-

this stepwise depth effect. Perhaps an
and facing of the physical surface pro-

in the last chapter, does not account for
surface bears a fixed relation to the slant

for a contour according to the suggestion
of density in a projection of a physical

which accompanies and ordinarily accounts
the line of sight. It is that the gradient

abrupt variation in brightness or color,
FIGURE 40. The Change of Gra- FIGURE 41. The Jump Between Two
dient Corresponding to a Corner Gradients Corresponding to an Edge
the gradient of density, however, there is parently produce a visual line, but of dif-
a change in the amount of density with the ferent types. A corner and a contour, al-
gradient remaining constant on either side thOugh both are lines in the visual field,
of the change. This discontinuity also have separate contributions to make to
gives the impression of a visual line even the perception of depth in the visual world,
though no line is drawn. It differs from the one helping make an object look solid and
previous drawing, however, in that the the other making it stand out from the back-
lower surface now appears in front of the ground. The perception of depth at a con-
upper surface with a jump in depth between tour (particularly when it is enhanced by
the two. The picture looks like a floor binocular vision) probably has a great deal
ending at a step, or a plateau with the to do with the impression that we can see
edge of a cliff and the country beyond. empty space.
It seems likely that here is a contributing This evidence indicates that variations
stimulus for the experience of a contour or and changes in the density of texture are
outline with depth. Both a sudden change specific stimulus correlates of the planes
in density and a sudden change in the and curves of an object, with their various
rate at which density changes will ap- slopes, and of the corners and contours
FINE
-II
TEXTURE
COARSE
TEXTURE
I.
I I I
TOP OF CORNER BOTTOM OF TOP OF EDGE BOTTOM 01
RETINA RETINA RETINA RETINA
FIGURE 42. Plotted Variations in the Stimulus-Gradients at

the Corner of an Object and at the Contour of an Object
of an object which separate it from the tion. The principal factor determining the
background as well. The retinal image degree to which a physical surface is
is evidently richer in opportunities for lighted or shaded is whether it is directly
depth-perception than the classical or only indirectly illuminated by the light-
theorists realized. source the sun, let us suppose. A sec-
Grades of Illumination and the Model- tion of surface facing the sun is brighter
ling of an Object or Surface. Painters than a section of surface facing away
have known for centuries that an object is from the sun. The latter possesses what
modelled in perception by the light and is called an attached shadow. There are
shade over its surface. In present ter- also, of course, cast shadows which are
minology, this means that the shading of projected on a facing surface by an object
a visual object can give it a depth-shape. which intercepts the light, but these are
Evidently it can do so independently of less important for our problem.
the density of the object's texture and The illumination of a given section of
the perspective of its edges, as Figure 43 surface, then, is a function of the orienta-
illustrates. What is the logic of this tion of the surface toward or away from
effect? Much has been written about the source of light. A fact to be especially
shading, for it is easy to produce with a noted is that illumination is not a function
pencil or charcoal, but little or nothing of the distance of the surface from the
with a view to establishing the psycho- observer. The physical world gets visually
physical correspondence, if there is any, denser as it recedes, but it does not get
between depth-shape and grades of illumi- either darker or brighter, and a distant area
nation. of uniform terrain is not different from a
Excepting objects which are themselves nearby area in this respect.4 Grades of
sources of light, no surface can be seen
unless it is illuminated in some degree. 4In the writer's opinion, some authorities
Considering the array-of surfaces which fill have misapprehended the relation between
the ordinary visual field with patches of brightness and distance, probably because
they considered isolated points as the sources
color, some of these will usually have a of light instead of surfaces. The question will
high illumination and some a low illumina- be considered at the end of the next chapter.
95
!rk.',0
m'mqr,lir
OCON40 r 4 3,,
/***6121Peit' orP'
4044',1 q'S'
VA1V Oi403,4, r
Oro c
' i014,titte0((,1A,
fit. {07,1,07-{1
R4f, A
40r0Fret,,,ii ir4:11r 4,4 12;
)10 1'r
%kr ,41)
Jo)
0,4,k6
Nigf.44/N ,
/Ivo 9.
toeAv
4
r
164
FIGURE 43. Gradients of Shading which Yield Depth
illumination, therefore, are not like the less opposite faces. We may call the
gradients of density, disparity, and first a protuberance and the second an
motility with which this chapter is princi- indentation. The three-dimensional shapes
pally concerned. They are a correlate of things, abstractly considered, are made
not of continuous distance or "space" up of protuberances and indentations.
but only of the depth-shape of objects This statement is perfectly consistent
and surfaces. with the analysis of three-dimensional
The depth-shape of an object seems to surfaces in terms of slope; it only im-
be that quality which, apart from its con- plies that an abrupt or a gradual varia-
tour or silhouette, makes it object-like. tion in slope produces a sharp or a
This kind of shape must have as a mini- rounded junction of surfaces. The junction
mum either a pair of surfaces making a may be either a bend or a curve.
corner or a curved surface. It therefore If one face of a protuberance is lighted
necessarily involves the opposite facing the other is necessarily shadowed, and
of adjoining surface areas. The elementary the same is true for an indentation. But
depth-shape (a pure abstraction like all the two shapes are distinguished by the
elementary variables) would be a con- order of stimulation. For, obviously, if
vexity or a concavity, both of which pos- the sun is in the south, then the southern
ti rV
0- 1
-0 I ,r
74,A
(;t144''
v.
1
A
f`
FIGURE 44. Shading which Yields Edged and Curved Surfaces

There is nothing more in either of these drawings than shades of black, gray, and white.
Nevertheless we see in each a complex array of physical surfaces in three dimensions.
But note that the transitions between shades are of different kinds: some are abrupt or
"sharp ", others are gradual or "rounded". Accordingly, we see edged surfaces in some
places and curved surfaces in others. That the curved protuberances and identations of
the instrument-case are produced by gradual transitions between light and dark one may
verify by inverting the picture; transitions which formerly went "in" now come "out," or
tend to do so whenever the factor of superposition does not inhibit this reversal. The
relation of the drawing-instruments to their pockets is no longer precise but ambiguous.
This rendering is to be contrasted with the abstract drawing on the right where the transi-
tions have intentionally been arranged so as to make the depth-relations equivocal and
nonrepresentative, and therefore to make the space of the picture fluctuate in an interest-
ing manner. (Left: Rendering by Paul Madden. Right: From a painting entitled "Composi-
tion", by Van Doesburg. Courtesy of Mrs. Peggy Guggenheim.)
face of a protuberance but the northern impression of a depth-shape. If the transi-

face of an indentation will be the one tion from light to shade is gradual the
illuminated. A pair of adjoining lighted shape is a curve; if the transition is
and shaded regions on a picture (and sudden the shape will be a corner. When
presumably on the retina) can yield the convexities and concavities of a picture
96
gyp; v'Gt
114 4,Ftl,' 401
40,0'
t 0\
'14 0 r
4P
' "
'
e44
ph, h i,NjO,
,44,4
#A41,0K.
, 4
,
);' 41 4
4
r. *10
' r
"r 114'
411
FIGURE 45. Shapes Which Are Reversed when the Picture is Inverted
(a) Hills turn into valleys. (b) Quonset Huts turn into towers. (c) Craters turn into mounds.
97
FIGURE 46. Reversal of Shape Due to Reversed Gradients of Density
are produced by light and shade alone, with- will be in the south, if he is in the north-
out any significant contribution by changes ern hemisphere and if the hour is not too
in the gradient of texture-density, it might near sunrise or sunset. Then the order
be predicted that inverting the picture, shadowed-lighted in his visual field (from
which reverses the order of light and shade, left to right) will correspond to a pro-
would turn each protuberance into an in- tuberance and the order lighted-shadowed
dentation and vice versa. This effect can to an indentation. But if he faces about
be observed in Figure 45. The hypothesis and looks west, the order shadowed-lighted
which suggests itself is that for any given will be an indentation and lighted-shadowed
visual field the order of projected light will bea protuberance. The order of
and shade is a contributing stimulus for shading is a stimulus for depth, therefore,
an elementary shape in depth. If the only in relation to the orientation of the
order "lighted-shadowed" yields a pro- observer to his total visual world or, more
tuberance in perception, then the order specifically, his orientation to the direc-
"shadowed-lighted" anywhere in the tion of the illumination. Is it conceivable
same field will give an indentation. that the way things face and the way the
Assume that an observer in an average observer faces are reciprocally inter-
outdoor environment faces east. The sun related even in stimulation? The fact is
4
STIMULUS VARIABLES MOMENTARY STI MULATI ON 99
that when the observer turns around the tions in the intensity of the image, or
convexities of an object are not converted both together. Insofar as the drawings
into concavities. If, however, the observer have served as psychophysical experi-
maintains his orientation and the light on ments to the reader, it may also seem fair
a surface in relief is experimentally re- to conclude that the retinal stimulus
versed in direction without the observer's variations yield impressions of depth, of
knowledge the protuberances tend to be- slope, and of surface-shape which cor-
come indentations just as they do in an respond to some of these abstract features
inverted picture. These inversions of of an object.
depth occur, of course, only when gradi- The gradient of texture, we may repeat,
ents of texture are absent or ineffective. is a function of the slant of a physical
The latter gradients are never equivocal surface away from the observer and the den-
with respect to depth (Figure 46). sity of the texture varies with physical dis-
The reversal of relief in relation to tance. Variation in shading, on the other
illumination has puzzled and fascinated hand, is a function of the physical orienta-
men for centuries. Convex and concave tion of the surface to the light source. It
relief correspond to the coin, and the varies not with distance but with the
stamp that made it, or to the wax impres- curving or bending of the surface relative
sion and the seal that produced it, both of to the direction of illumination. Even
which were familiar to the ancients. A the slightest curve or bend, insufficient to
modern instance of the phenomenon is en- make much difference in a texture gradient,
countered in the interpretation of aerial can produce a variation in shading if the
photographs. Boring's Sensation and Per- direction of the light is favorable. Hence
ception in the History of Experimental arises the capacity of light and shade to
Psychology traces the investigation of the give what artists call relief to a surface,
problem back to 1786 (12, pp. 266 and and to supplement the modelling of the
304). Experimental investigation of the surface in three dimensions.
phenomenon is still needed. In conclusion, the stimulus variations
The Perception of Objects as Such. A we have described need mathematical
visual object in depth may be analysed in analysis, and the impressions of depth
terms of several abstract variables which require verification. The mathematical
are interrelated with one another: the analysis is possible, however, and pre-
slope of its surfaces to the observer's line dictions can be drawn up which should
of vision, the orientation of its surfaces make experiments decisive. The theory
to the source of illumination, the corners suggested is far from being a complete
or curves of its surfaces (either convex or explanation of the perception of material
concave), and above all, its outline or objects. Although retinal correlates have
contour separating it from the background. been proposed for shape in depth, for
To all these features of an object there a contour, and for depth at a contour
correspond either variations in the density (with additional correlates still to be
of texture of the retinal image, or varia- described), no theory has been ventured
00 90 eo
110
Its
'1
1
4' 1
4ttatikikk
row
09 06 001
FIGURE 47. The Visual Fields of Each Eye and

the Binocular Field, as Measured with a Perimeter
to account for shape without depth, or binocular field (Figure 4 7 ) as distinguished

what is ordinarily called form. This is the from the monocular fields.5 The double
problem with which the Gestalt psycholo- images are not easy to observe, since they
gists began. Paradoxical as it may seem,
the perception of shape without depth is 5The exact locus of all those points in the
visual field which are not doubled will depend
more difficult to understand than the per- on the particular arrangement of. objects or
ception of shape in depth. We shall return surfaces in the environment projected. The
locus of all points in empty space which
to this problem in Chapter 10. theoretically should not be doubled is quite a
different matter. This latter is called the
The Stimulus Gradient of Binocular Dis- horopter, and a great deal of effort has been
parity Between Images expended this geometrical abstraction.
on
EMpirical determinations of the horopter do not
In addition to those differences between agree well with theoretical constructions,
the visual field and the visual world which however, and it can be argued that the horopter
and the theory of corresponding points which
were described in Chapter 2, there is it expresses do not clarify the problem of why
another difference which can be observed we see objects in depth. In a surface theory
of space perception the horopter is irrelevant.
only with both eyes open. The visual field Carr suggests that the horopter is merely a
of both eyes is usually filled with double geometrical curiosity (19). The clearest ex-
planation of it in English is given by Troland
images. They are a characteristic of the (111, p. 342).
are mostly peripheral and always out of point on which the two eyes converge. It
focus. If, however, one brings a finger is known that the stimulus for this visible
into the central part of the field, con- disparity is an optical disparity of the two
tinuing to fixate the wall of the room, the retinal images, each being a projection
two fingers become obvious, one on the of the environment from a different posi-
right and the other on the left. With a tion.
little practice in observing other things Retinal Disparity as a Function of Dis-
in the field which one is not focusing on, tance and as a Stimulus for Perception.
their doubled character usually becomes Figure 48 illustrates this disparity of the
visible. In contrast to this appearance two retinal images schematically. The
of the field one can see that the visual scene is a road down which the observer
world as one looks at it uncritically is looks toward the horizon. The image of
not in the least doubled, but is thoroughly the left eye is drawn in dashed lines and
single. that of the right eye in solid lines. F in-
The doubled character of things in the dicates the point of fixation. The drawings
field is a product of binocular vision, as are upright, as if the images had been
the reader can verify by holding up his projected on a picture-plane, and the dis-
finger, closing and then reopening one eye. placement in one image relative to the
The monocular field of view appears other has been exaggerated. Assuming for
single, clearly outlined, and "photo- the moment that the eyes are fixated on the
graphic" as compared with the binocular horizon, which implies that their axes
field. If the monocular field appears some- are parallel and without any convergence,
what less deep than the binocular field, it can be noted that the image of the left
or to lack a special quality of depth, the èye is relatively displaced to the right
implication is that the double images are (or that of the right eye to the left) except
symptomatic of that particular increment at the horizon, and that the amount of dis-
or quality of depth. parity is inversely proportional to distance
If you fixate the right index finger at from the observer. This disparity is cross-
arm's length and then move the left in- ed, the image of the left eye being on the
dex finger toward your eyes, you can see right in the visual field.
the disparity of the double finger increase. When the two images are combined by
You can also do the converse, fixating the superposing one on the other, the result
near finger and moving the far one. The represents something like the total field
fact is that in the visual field as a whole, of view of the two eyes. The binocular
objects, edges, or surfaces appear doubled field, strictly speaking, is only the central
in proportion as they are physically nearer portion of the total field, where the two
or farther than the point of fixation. In monocular fields overlap, and this has been
the central portion of the binocular field represented in the drawing. Only in this
the disparity of visible objects and ele- middle region are there double images.
ments is a direct function of their dis- The disparity of the edges of the road in-
tance along the line of sight from the creases as the distance of the road from
.
1/0.... MEMO. MOO. IMIIINIIIMIS
.1111.....
.00"
....o'
....--- ......
... ......
""s.
/
/
.111
F
/
11111111. NIEMEN 41111 emwomr
I\
Image of Left Eye Image of Right Eye
Combined Images when Fixating the Horizon
Combined Images when Fixating a Near Point

FIGURE 48. Schematic Projections of the Retinal Images of
the Two Eyes and their Combination, showing Disparity
The images ar.: represented as projections on a picture-plane in front of each eye, not as
if seen from behind each eye. Hence they bear some resemblance to the visual field.
This expedient makes the disparity easy to visualize, but the relationship of inversion
is left out of account. It must be remembered that retinal images as such are never seen
kw +km einirrwil theut krie thorn_
the observer decreases. This increasing before, but now there is also an opposite
displacement of the image toward the gradient of disparity above the level of
bottom of the visual field is, of course, a fixation up to the horizon. In the latter
displacement relative to the other image, gradient the image of the left eye is dis-
for there is no other standard with respect placed to the left instead of to the right.
to which it could be displaced. It is in This kind of disparity has been called un-
fact a gradient of disparity between the crossed to distinguish it from the crossed
binocular images. disparity already described. In ordinary
The diagrams represent only a single vision, with the eyes moving rapidly from
pair of edges in the environment since one object of the environment to another,
more could not be drawn without con- the disparities will shift with each new
fusion. Actually the two images are fixation. In this situation, we may inquire,
usually composed of an array of texture- is there any longer a clear relationship
elements and contours, like a patchwork, between degree of retinal disparity and
each image being somewhat skewed or distance of the environment?
stretched horizontally relative to the The correspondence between disparity
other, as if it were printed on a rubber and distance is seen to be preserved if
sheet which was then pulled in such a one takes account of the order of the
way as to transform a square into a disparity in the combined images, and
parallelogram. These disparate images considers the stimulus not as a geometri-
are united in the process of vision to cal picture but as an algebraic variable.
form a single visual field even though they Beyond the fixation point the projected
are not geometrically congruent. A disparity is a left-right one whereas in
fusion occurs between the neurological front of the fixation point it is a right-left
processes originating in each eye sepa- one. The terms right and left are merely
rately, and the visual world which results conventional; the essential fact is that
from this process is seen as a complete disparity may vary in opposite directions.
unity of objects in three dimensions. Physical distance from the observer to the
If we no longer assume that the eyes horizon can be put into correspondence
are fixated on the horizon but are con- with a scale of minus to plus disparity as
verging in some degree toward a spot or well as it can with a scale of minus to
object on the ground, the combined images zero disparity.
are represented by the lowest drawing in When double images in the visual field
Figure 48. The center of each visual are thought of as cues for distance per-
field has now moved downward so that ception and retinal disparity is conceived
the horizon is higher in the combined field as falling into two types, a difficulty
than before. At the level of the point of arises. How can the mind distinguish be-
fixation there is no disparity in the com- tween a crossed pair of images and an un-
bined image and no double imaging in the crossed one, considering that the one
binocular field. There is a gradient of looks like the other so far as any experi-
disparity below the level of fixation, as menter has been able to observe? In the
face of this difficulty, the fact remains increases in depth to the right. This
that the impression of distance is not yields the left hand wall of a corridor, or
changed by a change in fixation, such as is the left side of an indentation. The pair
represented by the two lower pictures of of pictures at the top have this gradient
Figure 48. The difficulty is resolved if one of disparity, and yield a surface of this
treats binocular disparity as a stimulus sort when they are viewed in a stereoscope.
gradient. A rate of decreasing crossed dis- If uncrossed disparity increases to the
parity can be the equivalent of a rate of in- left, the wall increases in depth to the
creasing uncrossed disparity in the same left, as does the right hand wall of a cor-
way that the numbers -3, -2, -1, 0 are ridor or the left side of a protuberance.
equivalent in order and rate to the numbers The same pair of pictures, when inter-
3,4,5,6. It is a reasonable hypothesis that changed right for left, have a reversed
some graded process in the brain of the ob- gradient of disparity and will yield the
server reacts to the disparity of his bino- latter type of surface in a stereoscope.
cular images in just this fashion. The Agradient of disparity which runs
rule would be that an impression of in- vertically up or down the combined images
creasing depth on a surface is in psycho- corresponds to a surface which slants as
physical correspondence with any gradient a floor or a ceiling does. If the uncrossed
of disparity running in the direction toward disparity of the two views increases up-
uncrossed, and that increasing nearness ward, as it does in the bottom pair of
is in correspondence with any gradient stereoscopic views, the surface slants
running toward crossed disparity. The into the distance as it goes up (like a
rule assumes the conventions of Figure floor); if it increases downward, as it
48 and refers, for convenience, to the would if the bottom pair of stereoscopic
picture-projection of each retinal image, views were interchanged, the surface
not the retinal image itself. It can be slants into the distance as it goes down
generalized to include the inclined sur- (like a ceiling). These are the slants
face which we have called a ceiling scene which actually appear when the pair of
as well as that of a ground scene, and a views is presented in a stereoscope. It
wall scene looked at from the left as well may be noted in a stereoscope that, since
as one looked at from the right. These the lines do not have a gradient of mono-
types of slanting surface, it will be re- cular perspective, the floor (or ceiling)
called, are as fundamental to the percep- gives the appearance of boards which get
tion of object-surfaces as they are to the wider as the distance increases. The
perception of surfaces of the environment. same types of surface may equally well be
Figure 49 illustrates these stimuli. A obtained with a pair of views composed of
gradient of binocular disparity which dots or short vertical line-segments.
runs horizontally across the combined The aniseikonic glasses devised at the
images corresponds to a "wall scene." Dartmouth Eye Institute for experiments
If the uncrossed disparity of the two on stereoscopic distortions of space and
views increases to the right, the wall recently described by Ames (1) are re-
105
LEFT EYE
RIGHT EYE
A HORIZONTAL GRADIENT OF BINOCULAR DISPARITY
A VERTICAL GRADIENT OF BINOCULAR DISPARITY
FIGURE 49. Gradients of Binocular Disparity in

Relation to the Fundamental Types of Slanting Surface
lated to the present theory. They have the lenses produce when an observer
the effect of magnifying or stretching the wears these glasses ought to be predictable
image of one eye along one dimension from the above rules. The validity of the
only or, in other words, of increasing the rules may be tested by their success in
relative size of one retinal image on a making the predictions.
single meridian. The lens employed and In general a vertical gradient of dis-
the disparity produced have been reported parity can be described as a horizontal
by Ogle (86). These glasses should have skew of one image relative to the other.
the effect of increasing the gradient of A horizontal gradient of disparity can be
disparity over certain physical surfaces described as a horizontal stretching of
and thereby increasing the apparent slant one image relative to the other. The dis-
of those surfaces. Allowing for the in- parity is always horizontal whereas the
fluence of conflicting gradients of texture gradient may be either vertical or horizon-
and perspective, the changes in the ap- tal. These relative distortions of the
parent slant of a physical surface which image are not to be confused with a rota-
tion of one image relative to the other, and for at least part of our impression that
such as would occur in cyclotorsion of we look out upon the world. It is impor-
the eyes. An image which has been tant to account for these impressions since,
skewed or stretched has undergone a non- unexplained, they lend plausibility to the
rigid transformation, mathematically speak- prevalent but sterile conceptions that
ing, whereas a rotation, like a simple the perception of depth reduces to the
transposition of an image, is a rigid type perception of an abstract third dimension
of change. These two types of distortion and that somehow visual sensations are
are also not to be confused with an en- projected outward from the eye in percep-
largement or magnification of one image tion.
relative to the other, which is the original Binocular Disparity and the Depth of
meaning assigned to the term aniseikonia at Objects. Figure 50 illustrates the way in
the Dartmouth Eye Institute. A relative which two projected views of a solid ob-
enlargement makes the two images dis-
parate in quite a different way from the one
defined above: the disparity is not horizon-
tal, it does not fall into a one-dimen-
sional gradient, and it would not occur in
natural vision. The effect of relative en-
largement, indeed, appears to be equivocal
and not completely understood.
The image of the nose, that unnoticed
but important feature of every binocular
visual field, is actually a crossed double-
image. It is, in fact, the ultimate limit
of crossed double imagery the end of
the gradient of disparity in the crossed
direction. Between the disparity of the
nose and the disparity of the next color-
patch above it in the visual field the
.
ground or floor nearby there is always a
very considerable step or jump. A dis-
continuous step in an otherwise continuous
gradient, we have suggested, is the
stimulus for the impression of depth at a
contour. Between the nose and the next
visually adjacent surface, therefore, a
considerable appearance of depth should
occur. This depth is probably the basis
for our impression that we see empty space FIGURE 50. The Disparate Views
between ourselves and the nearest object, of an Object by the Two Eyes
ject are disparate for the right and the left stimulus basis for the immediate experience
eye. One projected shape is skewed of the scene in depth.
horizontally relative to the other. When The most obvious kind of depth in
the two projected shapes are fused in stereoscopic photographs, especially of
binocular vision a depth-shape results, the the parlor variety, is the depth which ap-
depth being in correspondence with the pears at the contours of objectsthe depth
amount and direction of skewness. The which makes objects stand out from the
figure illustrates at the same time the background and which gives one the im-
principle of the stereoscope, the inven- pression that there is empty space between
tion of which provided Wheatstone with a the object and whatever is behind it. The
dramatic proof that binocular disparity stimulus for this impression can be
was a cause of depth perception. The ascribed to an abrupt change id the dis-
stereoscope is simply a device which per- parity of fusing elements, or a change in
mits the two views of Figure 50 to be pro- the amount of disparity without any grada-
jected independently on the two retinas. tion. It is to be distinguished from the
The drawing represents the edges and gradual change in disparity which is
corners of the solid object, as most draw- characteristic of a sloping surface, and
ings do, but not its textured surfaces. from the abrupt (or gradual) change in the
The gradients of disparity with respect gradient of disparity which is character-
to texture are not evident in 'the illustra- istic of a bend (or curve) in a surface.
tion. In the binocular vision of a real The Importance of Binocular Vision for
object, or when stereoscopic photographs Depth Perception. It has been commonly
of a natural scene are used, the gradients believed for many years that the only im-
of disparity on its various surfaces will portant basis for depth perception in the
produce the kinds of slant we have already visual world is the stereoscopic effect of
described and will have the effect of giving binocular vision. This is a widely ac-
the object a shape in depth. The gradient cepted opinion in the medical and physio-
of disparity on each surface area is, of logical study of vision, opthalmology,. It
course, concurrent with a gradient of the is the belief of photographers, artists,
density of its texture and a gradient of motion picture researchers, and visual
perspective (if the edges are parallel). educators who assume that a scene can
The change in the gradient of disparity be presented in true depth only with the
at an edge or a corner, moreover, is con- aid of stereoscopic techniques, and of
sistent with the change in the gradient of writers and authorities on aviation who
texture and with the change in the grade of assume that the only kind of test for depth
shading from one plane to another. In perception which a flier needs to pass is
ordinary vision the gradients supplement a test of his stereoscopic acuity. This
one another, it must be assumed, so that belief is based on the theory of the in-
even if the texture is indefinite, the object trinsic cues for depth, which is rooted in
irregular, the illumination flat, or the ob- the assumption that there exists a class
server blind in one eye, there will be some of experiences called innate sensations.
With the increasing tendency to question and that all of them are effective in every-
this assumption in modern psychology, day vision. The effect of any one may be
the belief is left without much foundation. experimentally isolated from that of the
Depth, we have argued, is not built up out others, as we have tried to illustrate, but
of sensations but is simply one of the di- ordinarily they function as concurrent
mensions of visual experience. The ac- stimuli. Whether one or another of these
cepted belief is contradicted by the fact gradients is more important for the re-
that persons having vision in only one sulting perception is something that can be
eye, or temporarily limited to the use of determined only by experiment, and some-
one eye, see the visual world in depth thing that will probably vary with the
just about as the rest of us do and get conditions of stimulation. If the gradient
about in it without conspicuous loss in theory is correct, binocular vision simply
efficiency. There have even been one-eyed takes its place as a determinant, but only
fliers, of whom Wiley Post was the most fone determinant, of visual space.
notable. Itis also contradicted by the j Definition or Resolution in the Retinal
fact that many animals who do not pos- .) Imoge
sess overlapping binocular fields and who
therefore lack disparate images rabbits A disparity of the image in one eye rela-
and rats, for instance tive to the other presupposes that both
seem to dis-
criminate depth and distance accurately images are textured. Only with respect to
since their behavior is nicely gauged with a pattern of spots or lines could a dis-
reference to it.6 The implication is that parity exist; there must be something to
the emphasis on binocular disparity as a be incongruent. The homogeneous blue
cue for depth has been exaggerated. sky would not yield binocular disparity
The theory of retinal gradients as because it would not produce a patterned
image. Binocular stereoscopic vision,
stimuli for visual depth implies that the
gradient of disparity is only one of several, therefore, depends on textured vision.
The very mechanism by which the single
6 Binocular
eye gets a focused image, the reflex
stereoscopic vision has probably response
evolved as a sort of alternative to panoramic of accommodation, must have
vision in which each eye has its own field of something to do with texture since focusing
view with little or no overlapping. The pri- can be thought of as a maximizing or
mates, apes and men, have forward pointing
eyes with coordinated eye movements, with sharpening of texture. What we call dis-
convergence of both eyes on the same object,
and with a total field of view approximating tinct or clear vision must be related to
1800; but many other mammals, specially those the texture of our impressions. Are all
preyed on by carnivores, have eyes which these facts interrelated? If so, what
point each to its own side, each moving in-
dependently, with little or no convergence and does optics contribute toward under-
with a total field of view approximating 360°. standing them?
The suggestion is that, in the course of
evolution, the primates sacrificed the ability The Problem of Clear Vision and Visual
to see all the way around at once for an en- Acuity. It is a fact, so far neglected in
hancement of the ability to discriminate depth
(23). this discussion, that our vision of the
world must be clear if it is to be useful. 1. Determine the smallest familiar form

Anyone who suffers from nearsightedness that can be recognized, starting with large
or an alcoholically induced imbalance of sizes and proceeding to smaller ones.
the eye-muscles will agree. So also would Block letters are usually employed. This
anyone who has to drive a car in a dense is the familiar eye-chart devised by Suel-
fog. Clear vision is easier to describe len, and modified by others.
than to account for, depending as it does 2. Determine the smallest form among
on three separate sets of conditions, a set of identical forms rotated to differ-
physical, optical, and neurological (see ent positions which can be seen in correct
Chapter 4). Descriptively or psychologi- orientation. For example, what is the
cally, a scene is clear when it is sharp, smallest broken circle, or Landolt ring,
or detailed, or definite. It is not clear for which the location of the gap can be
when the is fuzzy, foggy,
impression reported? This test eliminates the factor
vague, dim, indeterminate, or blurred.? of meaning or familiarity.
The ability to see very distant objects 3. Determine the smallest noticeable
or very small objects is called acute interspace between two dark objects on a
vision. Such ability is important in a light background, just before they merge
number of human occupations, and is also into one. Two points cannot be used un-
theoretically interesting because there is less they are big enough to be each visible,
a limiting size and a limiting distance, so the commonest practice is to present
the two being interrelated, below or beyond two rectangular bars side by side.
which objects become invisible. A good 4. Determine the smallest black dot on
many tests of visual acuity exist and a a light background which can be reported
great many experiments have been per- as there.
formed on the conditions which affect it. 5. Determine the thinnest black line
The artificial scenes which have been set on a light background which can be re-
up to measure this kind of fine discrimina- ported as there.
tion are interesting, and they may help to 6. Determine the finest or most dense
specify what has here been called visual grid of parallel black lines on a white
texture. ground for which the lines can be dis-
These are some of the ways in which tinguished as horizontally or
running
acuity can be measured: vertically. Such a device is an Ives
grating.
7. Determine the finest or most dense
All the above adjectives are properly ap- lack-and-white checkerboard which can be
plied to perception only. They do not ade-
quately describe the external flux of light, distinguished from an equivalent area of
they certainly do not describe a process in gray.
the nervous system, and they probably should
not even be applied to the retinal image when 8. Determine the just noticeable mis-
this is conceived strictly as a stimulus, that alignment of the two segments of a broken
is, when the image is thought of as a complex
of variations rather than as a picture to be straight line. This is called vernier
looked at. acuity. Our great sensitivity to such
no THE PERCEPTION OF THE VISUAL WORLD
misalignment is an advantage in the different sort. They involve deviations

reading of scales and pointers. from the straight quality of a contour.
9. Determine the just noticeable These deviations seem to be elementary
curvature or bend in a straight line (or a impressions of shape rather than elemen-
straight contour). Our sensitivity to a tary components of texture. They have to
deviation from the rectilinear is also do with the outline or margin of a physical
great. surface rather than with its texture. The
10. Determine the just noticeable dis- tenth test, of binocular disparity, is com-
parity between the image in the right eye plex. The eleventh, however, points to an
and the left eye in a setup viewed with interesting conclusion, that a difference in
both eyes. The effect perceived will be brightness cannot be sensed without an
one of depth. This is called stereoscopic accompanying impression of a margin (6).
acuity. The stimulus for a margin seems to be a
11. Determine the just noticeable dif- relatively abrupt gradient of intensity in
ference between the gray of a solid shape the retinal image. Is it possible that the
and the gray of its background, as evi- microgradient of intensity is the fundamen-
denced by whether a contour (and hence tal stimulus underlying not only the pheno-
the shape itself) is perceptible. This is menon of a margin or contour but also the
usually called brightness-discrimination, phenomena of texture, visual acuity in its
although there is evidence that it is re- different modes, the focused image, and
lated to the other modes of acuity (103,6). clear vision in general? Indirectly, if this
The Relation of Texture to Acuity. The were true, it would be the basic stimulus
sixth and seventh of the experiments or for the phenomena of edges, surfaces,
tests listed above demonstrate that shapes, and objects. Before answering
fineness or density of visible texture is a this question we must consider a number
measure of the acuteness of vision. Both of other problems.
a grid and a checkerboard are cyclical It hat are the Conditions for Clear
variations of dark and light. What do the Vision? Sharp contours and definite tex-
remaining patterns suggest? Omitting the tures in vision depend not only on ac-
first test, the second and third involve commodation of the lenses and normal
the impression of a gap or interspace. eyes (the dioptric mechanism) but also on
However a gap may be defined geometri- external or physical conditions and on in-
cally, it can be considered a fundamental ternal or neurological conditions. Ex-
component of a texture. The fourth and ternally there must be reflecting sub-
fifth tests employ a spot and a dark line stances and a clear medium, that is, one
respectively. A line, of course, is a sort which does not scatter the hypothetical
of elongated spot. Spots may be com- rays of light, as fog does. Internally
ponents of a texture and streaks or in- there must be a fully matured nervous
lines (as distinguished from outlines) may system and a dense mosaic of retinal
also be components of a texture. The receptor-cells. Disease or injury to the
eighth and ninth tests, however, are of a occipital brain, the optic tract, the nerves,
STIMULUS VARIABLES MOMENTARY STIMULATI ON 111
or the retina, will all reduce visual any data for depth-perception (121, p. 665-
acuity. The system is intricate and ap- 680).
parently does not fully develop in child- Both the bulging or flattening of the
ren until 7 or 8 years of age inasmuch as lens and the pointing of the eyes are reflex
visual acuity, poor in infancy, continues adjustments. As reflexes they must have a
to improve until then. Between the ex- stimulus. What has to be understood is
ternal light and the internal nervous pro- that this stimulus cannot simply be light;
cess there stands the mechanism of the it must be a condition of the retinal image
eye itself. Here is where the most common with respect to order or pattern. Funda-
defects are found which reduce acuity. mentally this condition is gegmetrical.
A primary function of the eye is to form What is the sense, then, in seeking an
a focused image, one in which there is explanation for depth in the adjustments
experienced a minimum degree of so-called of the eyes to a stimulus when the stimulus
blur. A focused image depends on ac- itself is something from which the explana-
commodation. Blur may result from a tion may be derived? Accommodation and
whole complex of anatomical and physio- convergence are responses of the eyes to a
logical defects, the simplest results of condition of their images (blur and dis-
which are nearsightedness, farsightedness, parity) which may concurrently produce
and astigmatism. These particular effects that inner response we call "depth".
can be compensated in part by supplement- ,blur as the Stimulus for the Reflex of
ing the lenses of the eyes with artificial . ccomrnodation. The lens mechanism
lenses. N./seems to operate on the principle of
Accommodation and Convergence as minimizing the condition in the retinal
Cues for Distance. It will be recalled that image which yields blur, not over its
as early as 1709 Bishop Berkeley believed whole extent (for this would normally be
that accommodation of the lens, together impossible) but at the fovea. Since blur
with the convergence of the lieyes, fur- may result from either too thin or too
nished a sign for the distance of the object thick a lens, the process is in all likeli-
fixated and therefore gave it the appear- hood a sort of back-and-forth or trial-and-
ance of existing in a third dimension. The error one, not unlike that of focusing a
precise cue employed would have to be the lantern-slide projector on a screen. The
muscle-sensations involved in these reflex resulting image has what is known in
adjustments. The theory has persisted to optics and the study of acuity as a maxi-
the present day. After the invention of the mum of definition or resolution. It seems
stereoscope by Wheatstone in 1833 and the possible that these terms can be analysed
discovery of the binocular disparity be- geometrically.
tween images, these muscle-sensations If we are forced to suppose that the eye
lost some of their importance as cues for is sensitive to stimulation of this sort in
depth, but the problem continued to at- its reflex functioning, there is surely
tract interest (19), Present evidence, how- reason to believe that it can be similarly
ever, makes it doubtful that they furnish sensitive in its higher functions. The
112 THE PERCEPTION OF THE VISUAL 'WORLD
elementary impressions of surface and comes a very minor matter. All of the field
edge, or texture and contour, may plausibly except that directly under the nose tends
result from stimulation of this sort. If a to be uniformly in focus; the gradient
reflex reaction can be a response to a levels off. This results from the small
geometrical condition of the image, so also aperture and small size of the eye con-
may a quality of experience. sidered as a camera; the depth of field
Is the Gradient of Blur a Cue for Dis- (depth focus in the photographer's
of
tance? The accommodation of the lens terminology) is very great. The working
which eliminates blur at the fovea neces- limitations of the camera in this respect
sarily produces- blur in outlying regions of scarcely apply to the human eye. Hence
the image when the environment is a ter- it is, probably, that a gradient of texture
rain or has a floor or walls. Physical over the field can be sensed in a single
parts of a surface at greater or lesser dis- fixation without much interference from
tances than the point of fixation will tend blur.8
to be blurred. Does this constitute a The Definite, Resolved, or Focused
gradient of increasing blur over the sur- Image. It is now possible to return to the
face similar to the gradient of density of problem of the relationship between acuity
texture and the gradient of disparity? If and texture. We are told by the students
so, might it be a supplementary stimulus of the subject that visual acuity is de-
for the impression of continuous distance? fined as the smallest visual angle that
The quality of blur does grade off from can be resolved or the smallest shape
the center of the visual field, but it does which has definition. These terms are
not seem to change character in opposite not perfectly satisfactory by themselves
directions. Lines or spots in the visual for, unless they are analysed geometri-
field have the same fuzzy character whether cally, they come down to this: that acuity
their objects are nearer or farther than the is what the tests of acuity test.
object in focus. The phenomenon may be The geometrical analysis of an un-
observed in photographs. It is probably blurred image provided by optics employs
not, therefore, a univocal gradient like the concepts of ray and point, of the focal
the gradient of crossed to uncrossed dis- point, and of the circle of confusion.
parity with which it is always associated With a perfect lens and a perfect image,
in looking at ordinary surfaces. Whether all light from a single point converges to
it could be experimentally isolated is a very minute point in the image. With
doubtful; the writer's efforts to obtain a other conditions a pencil of rays forms a
simple gradient of blur on a photograph more or less extended figure in the image
were not successful. It can hardly, there- instead of a point, the circle of confusion,
fore, be an effective independent stimulus and these overlapping circles account for
for distance. blur.
The fact is, moreover, that when the 8
eye is accommodated for any distance be- The writer is indebted to Professor Nora
M. Mohler of Smith College for the computations
yond a few feet the gradient of blur be- on which these conclusions are based.
STI MULUS VARIABLES MOMENTARY STIMULATION 113
This kind of analysis is not the only of the pair of gradients involved, that is to
one which is mathematically possible, say a focused image.
however, and it seems likely that analysis In all that has gone before, full illumina-
in terms of area and intensity within the tion has been taken for granted. Acute
two dimensions of an image would be more vision, however, depends on a bright image
profitable for the study of vision. From as well as on a focused image; the ability
this standpoint, a focused image would be to see detail falls off rapidly as the light
one in which the transitions from dark to reflected from objects grows dim. The
light were as abrupt as the adjustments of above formula is also consistent with this
the optical system permit. This analysis dependence of acuity on, illumination,
begins not with geometrical points but inasmuch as the gradients of intensity in
with gradients (in this case microgradients, an image become less steep as the inten-
page 73) and their slope. The formula sity of its brightest spots is lowered.9
would be that definition or resolution of an The outcome seems to be that the tex-
image is the degree to which the gradients ture, detail, and pattern of our visual per-
of light intensity within it are as steep ceptions, on which their spatial character
as the relevant conditions permit. depends, are themselves dependent on
If this formula will account for the steps of luminous intensity in the retinal
physically focused image for the ex- image. The large scale gradients of detail
ternal and the optical set of conditions of with which this chapter is concerned are
clear and acute vision it may help us based ultimately on small scale gradients
to understand the dependence of acuity on of the intensity of light. Although such a
the retina and the brain, the neurological formulation of the matter is tentative and
conditions of clear vision. Most of the incomplete, it has the virtue of bridging the
research on visual acuity by psycholo- gap between optics on the one hand and
gists has been concentrated on this ques- visual perception on the other.
tion (95), but it is outside the scope of our Ernst Mach discovered long ago, by
discussion. fusing a combination of black and white
According to this formula, the steepest on a rotating disk, that a regular and uni-
gradient of intensity would be the condi- form increase of light-intensity along a
tion corresponding to the sharpest visual given dimension of the retina did not
contour. A series of alternating gradients yield an impression of gradually in-
as steep as possible would be the condi- creasing brightness as one might expect
tion corresponding to a clearly perceptible (77, p. 217). An impression of abruptly in-
texture or surface, both in the case of a creasing brightness, a margin or border in
surface with gross details and in the other words, could be produced much more
case of one with fine microstructure. The easily. The latter impression occurred
ability to see a very small spot, gap, or
9
streak, as this is measured in tests of The experimental evidence, on acuity and
acuity, would also depend on the formation on the relation of small areas illtensity is
referred to in Bartley's survey (6) ènd in a
of an image with the maximum steepness recent review by Senders (95).
image with the maximum steepness recent review by Senders (95).
of an
referred to in Bartley's survey (6) ènd in a
acuity, would also depend on the formation
on the relation of small areas illtensity is
as this is measured in tests of

streak, The experimental evidence, on acuity and
to see a very small spot, gap, or

ability
case of one with fine microstructure. The

easily. The latter impression occurred
gross details and in the

surface with
other words, could be produced much more
texture or surface, both in the case of a

creasing brightness, a margin or border in
tion corresponding to a clearly perceptible

(77, p.
217). An impression of abruptly in-
as steep as possible would be the condi-

creasing brightness as one might expect
contour. A series of alternating gradients

yield an impression of gradually in-
tion corresponding to the sharpest visual given dimension of the retina did not
gradient of intensity would be the condi-

form increase of light-intensity along a
According to this formula, the steepest on a rotating disk, that a regular and uni-
discussion. fusing a combination of black and white
tion
(95), but it is outside the scope of our
Ernst Mach discovered long ago, by
visual perception on the other.

gists has been concentrated on this ques-
gap between optics on the one hand and

research on visual acuity by psycholo-
incomplete, it has the virtue of bridging the
conditions of clear vision. Most of the
not only with an abrupt increase of light- dients, known as aerial perspective. As a
intensity along a dimension of the retina, phenomenon it has been known to painters
but also with an abrupt change in the rate since Leonardo da Vinci first described and
of increase along a given dimension. This named it four centuries ago. The color of
is the meaning of the Mach rings. The a landscape which stretches off toward the
principle illustrated in Figures 40, 41, and horizon becomes bluer and more filmy with
42 appears to be a fundamental one which increasing distance. The fact is that the
applies to gradients of intensity on the retinal image of such an environment is
retina as well as to gradients of the constituted by pencils of light which have
density of a texture. travelled through differing amounts of air.
The principle seems to be that a visual At the extreme top of the image the light
margin or border is given in experience by may have passed through only a few feet of
either a step in the level of luminous in- atmosphere, whereas at the lower margin
tensity (a steep microgradient) or by a (corresponding to the horizon) the light will
step in the first differential of luminous in- have passed through many miles of atmos-
tensity (a change in the rate of change). phere. Owing to differential scattering of
These two seem to be equivalent as types wavelengths, the color of the light reach-
of ordinal stimulation. The retina responds ing the eye differs under these two condi-
to an abrupt change of change as readily tions, even though the color of the reflect-
as it responds to an abrupt change. ing surfaces may be identical. The sim-
Abruptness seems to be the critical condi- plified explanation sometimes given is
tion. Perhaps the function of the retina that just as the sky itself is blue, so also
has been misconceived in developing the the hue of the hills near the horizon is
theory that it responds to light rays and shifted from its proper quality toward blue
their directions. It seems to respond to in proportion to their distance. The
gradients and their differentials instead. gradient of this shift in color might be
The retina is probably to be conceived as considered analogous to the gradients of
an organ of the body which is sensitive
j
size or density of texture.
to grades of light, not points of light. We cannot be sure, however, that this
increasing blueness in the visual field is
Te Gradients of Aerial Perspective a true stimulus gradient unless it can be
In the course of the evolution of human shown independently to yield an impres-
vision, we might conjecture, all the exist- sion of continuous distance. The same
ing variations within the retinal image thing would be true of a gradient of hazi-
have been utilized as stimuli for percep- ness. No such experimental demonstra-
tion if they are consistently in correspond- tion has been made; aerial perspective
ence with the actual lay of the land. has apparently not been studied in isola-
Several of these projected gradations, tion. The change of color is not simple
differentials, and shadings have been and is comparatively slight as compared
described. One more which deserves men- with the full range of color variations. It
tion is the gradient, or complex of gra- is unlikely that we can discriminate as
ctu v.
7
c' IA)
11,0
c 1
ft 11
Bahnsen, from Monkmeyer
FIGURE 51. Aerial Perspective
many grades of this color change as we Perhaps its function is more nearly what
can grades of, say, texture density, and it it has long been assumed to be a cue,
is therefore doubtful that the color gradient indicator, or sign of distance. If, as seems
could prove to be as effective a stimulus possible, the gradient is not always con-
as the texture gradient. Aerial perspec- sistently in correspondence with the ac-
tive depends on the amount of haze in the tual lay of the land, it is only reasonable
atmosphere, and may differ from day to to suppose that it should suggest the im-
day. It is certainly not discriminable in pression of distance, as red suggests
the immediately surrounding environment. warmth, without compelling the impression
In the absence of evidence, it would be in the way a stimulus is supposed to do.
safest to proceed on the assumption that The question, however, is a matter for
a gradient of aerial perspective is not a experiment.
stimulus in the proper sense of that term. Aerial perspective is not intrinsically
related to the surfaces of the environment and intricate event, deserving of more
in a geometrical way as are linear perspec- respect than it has usually been given 10
and texture perspective. Neither are It is necessary only to give up the ex-
gradations of shading intrinsically con- pectation of finding in it replicas of the
nected with the geometrical shapes of experience we wish to explain and seek
surfaces and objects; they are only in- correlates instead. What right did we
directly connected by way of their orienta- ever have to assume that the retinal
tion to the source of light. The advancing image must copy the world, or that the
and receding colors which painters employ form of an experience must be duplicated
to bring an object'out from or back behind by the form of its image? We understand
the picture-plane are also not connected well enough that a visual stimulus is
with physical depth by any clearly under- neither an object nor an experience of that
stood principle. One might speculate that object, but something which stands be,
variations in hue or shading _as such do tween them. What we have failed to under-
not produce the same compelling impres- stand is that this stimulus need not look
sion of depth that gradients of texture, like either its cause, the object, or its
line, size, binocular disparity, and motion effect, the experience. It need only be a
produce, just because they are not re- specific correlate of both.
lated to physical depth by geometrical Some of the previously unexplained
laws as the latter are. Variations in hue features of the visual world are accounted
and brightness can and do produce com- for if we suppose the retinal image in
pelling experiences of outline, form, and each eye to be an array of steps, gradients,
pattern in the two dimensions of extensity, and changes in gradients of light-varia-
but their correspondence to experiences tion. Variations in texture and size, in
of solidity, depth, and distance is less shading, and in binocular disparity have
precise. now been described in these terms. They
are in exact geometrical correspondence
Summary
with the dimensions of the physical world,
The seeming poverty of the visual and there was evidence that they yield
stimulus as compared with the richness of corresponding variations in perceptual
visual experience has apparently been experience.
exaggerated. Even the static momentary
retinal images with which this chapter 10
According to a leading textbook of ophthal-
has been concerned appear, when analy- mology by Duke-Elder (30, p. 764), the defects
sed, to be adequate to account for the of the normal eye as an optical instrument are
depth and distance of the visual world few. The most important kinds of distortion
and even aberration found in ordinary lens
without necessity of supposing a
the systems are effectively corrected in the human
special mental process to supplement the eye. This fact is in contradiction to a fairly
widespread impression that the eye is a poor
images. The retinal image is an exact optical instrument.
The Stimulus Variables for Visual Depth
and Distance -- The Active Observer
The Gradients of Deformation of the Image During

Movement of the Observer . . . . The Types of
Retinal Motion . . . . Summary The Sensory
Analysis of Distance and Depth
Heretofore we have been talking about Every photographer is aware that even a
visual perception for the most part as if slight movement of his camera during ex-
the observer stood motionless in the en posure will shift the image on the film,
vironment and kept his head fixed in one for it ruins his picture. The same kind of
position. The normal human being, how- shifting of the image on the retina occurs
ever, is active. His head never remains all the time during vision, with the dif-
in a fixed position for any length of time ference that vision is enriched rather than
except in artificial situations. If he is not spoiled. The retinal image, of course,
walking or driving a car or looking from a has a very different function than has the
train or airplane, his ordinary adjustments photographic image. It is not registered
of posture will produce some change in the as an unchanging distribution of grains of
position of his eyes in space. Such metallic silver on a film, but as a flow of
changes will modify the retinal images in neural excitations and ultimately as
a quite specific way. Just as the image of visual experience which continues in
the terrain in the right eye differs from that time and changes as the image changes.
in the left eye by being horizontally Motion of the retinal image relative to the
skewed, so the image in either eye will mosaic of sensitive rods and cones is
be similarly skewed when the head is therefore a normal stimulus-condition for
moved sideways for a distance of two and vision and one which is almost continu-
a half inches. Both effects are similar in ously present during waking life. It must
principle to what astronomers know as be remembered, however, that motion of
parallax. The first is termed binocul'ar the image produced by head movements is
parallax; the second is usually called not the same as that produced by eye
motion parallax. movements from one fixation to another.
117
; ;
FIGURE 52. Successive Views of a Row of Fence Posts
The former is a stimulus for the percep- special attention, and sometimes special
tion of space, as we shall try to prove, practice, since what we ordinarily ex-
and is also a precise sensory correlate of perience is the visual world, which does
locomotor behavior; the latter has no such not manifest deformation. Only when we
stimulus function. The former motion is drive a car or ride on a train does it become
one which deforms the image; the latter is so strong as to be unmistakable. A better
one which simply transposes the image. idea can be obtained of what the retinal
Figure 52 shows successive views of a image must be like in this respect by
row of fence posts. If you are moving at holding a "view" camera with its shutter
right angles to the line of posts the open in front of one's eyes and then moving
parallax (shift) of each post decreases as it as the head would move. The inverted
its distance increases. This decrease image on the ground glass screen presents
follows the same geometrical law as does a striking contrast to the stability of or-
the decrease in size. Consequently the dinary visual perception.
image of the posts undergoes a horizontal The fact of the relative motion of ob-
skew which can be specified mathema- jects as a cue for their depth has been
tically. known for a long time and has always been
included in the list of indicators by which
The Gradients of Deformation of the Image
During Movement of the Observer
the distance of an object may be judged.
It has often been described as it appears
The visual field of an observer is alive from a moving train. Helmholtz wrote about
with motion whenever his head moves. If it as it looks when walking:
the reader will fixate an object across the
room and then move his head or change his Suppose, for instance, that a person is stand-
ing in a thick woods, where it is impossible
posture, the shifting of contours can for him to distinguish, except vaguely and
probably be noticed. Actually there is a roughly, in the mass of foliage and branches
deformation of the color patches in his all around him what belongs to one tree and
what to another, or how far apart the separate
field, and this corresponds to the deforma- trees are, etc. But the moment he begins to
tion occurring in his retinal images. Like move forward, everything disentangles itself,
blur, double outlines, and the other charac- and immediately he gets an apperception of
the material contents of the woods and their
teristics of the visual field, this lively relations to each other in space, just as if he
shifting of contours is visible only with were looking at a good stereoscopic view of it.
'
STIMULUS VARIABLES THE ACTIVE OBSERVER 119
When we walk through open country with conclusion is indicated by the fact that,
eyes fixed on the distance, Helmholtz in a motion picture view, an unnoticeable
wrote: optical distortion in the photographic
....objects that are at rest by the wayside.. image of a room becomes obvious at once
appear to glide past us in our field of view in when the camera is "panned" from one
the opposite direction to that in which we are
advancing. More distant objects do the same side to the other, being seen .as a deforma-
way, only more slowly, while very remote tion of the walls and corners of the room.
bodies like the stars maintain their permanent The distortion of the still picture of the
positions in the field of view.. Evidently,
under these circumstances, the apparent an- room from normal perspective is not great
gular velocities of objects in the field of view enough to be noticed, but the slight change
will be inversely proportional to their real dis- in the shape of the image produced by
tances away; and consequently safe conclu-
sions can be drawn as to the real distance of moving ,,,,h e camera lens is noticed easily,
the body.... (53, vol. 3, P. 295). and it,, appears as an apparent stretching
Helmholtz referred to the experience of and ,iontracting of the walls.
depth as an "apperception" or as a "safe
conclusion" about the objects in the field
N'he Perspective of Visual Motion for a
Projected Terrain. Consider an observer
of view. He was thinking only of objects who is moving parallel to the ground, as
and not of a background or an array of he would be during normal locomotion. Let
elements. The relative motion he de- us assume that his eyes are fixated on
scribed is a variable of the retinal image the horizon, that is, let us disregard for
itself when the latter is considered as a the present any movement of his eyes
whole. When it is so considered, as a either in pursuit of the gliding terrain or
projection of the terrain or as the projec- from one fixation to another. The way
tion of an array of slanted surfaces, the things appear to an observer riding in an
retinal image is not a picture of objects open car on a clear moonlight night is a
but a complex of variations. If the rela- good example for anyone who can isolate
tive motion is analysed out and isolated this experience from distracting memories.
from the complex of other variations, it When such a scene is looked at just for
proves to be a lawful and regular pheno- the sake of looking, the horizon, the
menon. Defined as a gradient of motion, stars, and all the field of view upward are
it is potentially a stimulus correlate for motionless, but the world and the ground
an experience of continuous distance on a below flow past in a continuous stream.
surface, as we shall see, and one no The flow vanishes at the horizon; but it
longer is required to postulate a process of increases downward and reaches its
unconscious inference about isolated ob- maximum on the road beneath. The flow
jects. is not like that of a river as it would ap-
We probably have a high degree of sensi- pear from a bridge but rather as it would
tivity to this kind of visual stimulation, appear in perspective from the bank; it is
that is, motion consisting of a change in a continuous deformation of the surface,
shape as distinguished from motion con- not a movement in the ordinary sense of
sisting only of a change in location. The that term. In whatever direction one looks,
forward, to the side, or behind, the flow direction of the line of locomotion are ob-
decreases upward in the visual field and jective directions which cannot be literally
vanishes at the horizon. There exists, in copied in a projected image.
short, a perspective of this motion which To be specific, the flow of the terrain is
is fundamentally similar to the perspective visually downward in the field as one
of the density of texture and the size of looks ahead from the driver's position; it
objects. is to the right as one looks to the right, or
Considering the terrain as projected on to the left as one looks to the left, and it
a plane in front of the eye, the rate at is upward in the field as one looks behind.
which any element or object flows is in- In other words, it is different in different
versely proportional to its physical dis- visual fields of fixation as these vary from
tance from the observer, as Helmholtz the forward direction of locomotion. As
noted. The motility of the world, like its the observer gets successively different
density, decreases as it recedes. The visual fields in turning his head around to
geometry of this decrease is precisely the right the direction of flow changes suc-
like that illustrated in Figure 30 for or- cessively in a counterclockwise rotation,
dinary perspective. Near elements of the and if he turns to the left the change is
ground change their angular direction from precisely the same except clockwise.
the observer (parallax) more rapidly than Descriptively, the visual field ahead ex-
do distant elements. There is, in other pands outward from a focus, the visual field
words, a continuous gradient of the velo- behind contracts inward to a focus, and
city of the ground "going by" from a the visual field to one side or the other
maximum at the bottom of the visual field is being continuously skewed. The foci
to zero at the horizon. of expansion and contraction correspond
The Direction of Flow in Visual 'lotion objectively to the points toward which
Perspective. Motion, of course, always and away from which locomotion is aimed.
has direction as well as speed. The Most of us have observed the expanding
two may vary quite independently of one visual field ahead while driving a car, and
another, and it should therefore not be the contracting visual field while riding on
surprising to discover that the direction of the rear end of a train.
flow in the projection of the ground during These visual fields are represented in
our moonlight ride does not vary in the Figures 53 and 54 as a projected terrain
same way the velocity does. We are re- viewed from an airplane during level
ferring to the visual field, of course, in flight. In Figure 53 the focus of expansion,
which direction can vary through 3600 or point of aim, is on the horizon, and the
like the hand of a clock. This visual diagram would be the same in principle for
direction of flow depends upon the physical a man on foot or for the driver of an auto-
direction of the spot in question, which mobile. The surface as projected is con-
varies like the pointer of a compass. The tinuously deformed in the manner indicated,
physical. direction from the observer of and each arrow is a vector representing
any spot on the ground and the physical the velocity and direction of flow of the
..,,
11ftim
de" .16,
s."---,.
.4.------
/ FIGURE 53. Motion Perspective in the Visual Field Ahead

N
4-.3.
......
(t........-----........" cv-
s -4. 40. 4 -4. .4. 4.. 4. 41. ....11. -11.
511.
FIGURE 54. Motion Perspective in the Visual Field Looking to the Right
If the arrows are reversed, this becomes the visual field looking to the left.
surface-element at that point. On the and yet finite, and it would return upon
picture-plane there is a gradient of de- itself. The space composed of one's com-
creasing velocity from the bottom up to the bined visual fields may be said to be a
horizon, and also a gradient of changing curved space in the sense that a point
direction from the midline to either side which traces a straight line will even-
(which itself decreases upward). The tually come back to the position from which
mountains on the horizon and the clouds it started instead of travelling off end-
are at such great distances that the lessly in the same direction. In other
velocity of deformation approaches zero, words, if human beings had a visual field
and the sky is not deformed since it is not whose width included the entire horizon
a determinate surface. If the direction of if they could see all the way around at the
the arrows were reversed to make the field same time like a rabbit the field during
contract radially inward instead of ex- locomotion would appear to open up ahead
panding radially outward, it would then be and close in behind in a rather astonishing
the field of view looking backward. manner. Such characteristics of the visual
Figure 54 represents a deformation field created a great deal of difficulty for
familiar to passengers on trains and air- the early students of perspective and for
planes. On the picture the velocity gra- painters who wished to represent a large
dient is similar to the gradient of linear sector of the visual world on a picture -
perspective and, like that gradient, it plane.
vanishes on the horizon, so long as the ob- Actually, of course, no rabbits and
server is fixating the horizon. Actually, relatively few men have ever adopted the
Figure 54 should be visualized as merging peculiar attitude of psychologists, artists,
with Figure 53 on the left side and with and geometers which enables them to see
the reversal of Figure 53 on the right to their visual field.1 They are, with good
yield a combination of visual fields which reason, perfectly content with the visual
cannot be projected on a plane picture. If world as it is normally perceived, conform-
the arrows on Figure 54 were reversed, ing to the rules of Euclidean geometry.
the scene would represent the visual field The world does not undergo any flowing de-
looking 900 to the left of the line of loco- formation; it is seen to be stable and rigid
.
motion. and what moves is the observer himself.
It is interesting to note that if we could The ground does not move in a gradient
combine all these two-dimensional proof directions which change around the
jections of a three dimensional visual clock; the observer sees himself moving
world into a single scene, we would ob- in a single direction in three dimensional
tain a two dimensional space, in the geo- space. Objects do not change position in
metrical sense, which is non-Euclidean. relation to the observer; the observer sees
It would have the properties of the
theoretical space defined by the surface 1
of a sphere considered as a two-dimen- The writer must admit that he hag had no
introspections from rabbits to justify this
sipnal surface, i.e. it would be boundless conclusion.
...-".."' ".'...2."%jr53kVE.4":+7ft. '''' .,..

.0' ----
.. ...- -.../ ,N.....
..
`... *--- "-4.%
Ns.. "....
.0"
/ ,V
....- /
\\ N
%.
\ ...... N.%
/ .0° ..'° N,.....\

.N. %%1/4
'..",.....\
///
..../
//
NA
// / /4 V. \\
/ r
44
,4 / I
//
/i
/
I I
1
I
/ I / I
/ 1
I
II // 1
I
I
FIGURE 55. The Directions of Deformations in the Visual Field during

Forward Locomotion, as Projected on a Spherical Surface around the Head
his own change of position relative to the tion of locomotion and this provides him
objects. This flowing deformation then in with a sense of a point of aim. The point
the visual fields we have been describing of aim is, in fact, implicit everywhere in
is nothing but a visual symptom of the gra- the visual field, and even when the ob-
dients of velocity and direction in the server does not look where he is going, he
retinal image. We shall try to show how can in a sense see where he is going. This
these gradients are stimulus correlates fact enables strollers, automobile drivers,
for perceived space and perceived locomo- and aircraft pilots to see a great deal
tion. more of the scenery than that which lies
The gradients of velocity and direction directly ahead of them, although admitted-
are invariable accompaniments of locomo- ly it is a performance which should be
tion if the observer keeps his eyes open. practiced in moderation. So strict are the
The focus of expansion in the field ahead geometrical relationships between physical
is -an exact indicator of the point in the motion of the observer's body and retinal
world toward which he is going; a shift of motion of the projected environment that
the focus goes with a change in the direc- the latter provides in fact the chief sensory
guide for locomotion in space. Retinal ordinal stimulation of the retina, but only
deformation is actually a kind of visual luminous stimulation. There is no texture,
kinesthesis.2 no surface, no motion, and accordingly no
The flow of the terrain if one imagined determinate world and no visual sense of
oneself to be looking vertically downward one's own movement. Retinal motion pre-
at it from a considerable altitude during, supposes the stimulus of retinal texture,
level flight would be in general a flow from as we pointed out at the beginning of this
the top to the bottom of the visual field. chapter. The air traveller who looks out
The velocity would be a maximum at the into a cloudless sky gets the impression of
point physically below the airplane, at the in empty space even though he
floating
center of the visual field, and would de- may be objectively moving at 200 miles
crease outward in all directions toward per hour.
the horizon, 90° from the center of the If there are cloud masses in the sky,
field. The direction of flow would be diver- however, these provide textured surfaces
gent at the top of the field, away from the on which to anchor space. Some of the
focus of expansion, and convergent at the most interesting spatial experiences in
bottom of the field, toward the focus of flying are provided by the motion perspec-
contraction. If the directions in the field tives of broken cumulus clouds when pass-
were plotted in polar coordinates, the ing through them. If, as frequently hap-
plot would look like a melon-shaped pens, the clouds take the form of a solid
family of curves. It should be noted, horizontal overcast or ceiling, the flow of
however, that this downward looking visual this surface will be precisely the inverse
field would have to be accompanied by eye of the flow on the ground, as represented
movements, the effect of which we have not in Figure 56. As in previous instances,
yet considered. an inverted gradient yields an inverted
If one asks how the retina is stimulated surface.
when the observer looks vertically upward The Effect of Eye-Movements on the
into a clear sky during locomotion, the Retinal Gradient of Motion. The flowing
answer is probably obvious. There is no visual field has so far been described with
the limiting assumption that the ob-
server's eyes are fixed on the horizon and
2 are therefore for all practical purposes
The difference between this description of
motion perspective and the established con- motionless in his head. During actual
ception of motion parallax is that the former is locomotion, however, the eyes are seldom
allied to locomotion of the organism whereas
the latter usually implies that the animal in on the horizon and, since all other parts
question must move its head from side to side of the world are flowing, they are seldom
in order to obtain the cue for depth. Moreover, stationary. The eyes of an automobile
the retinal gradient of velocity on a ground
surface is a stimulus-variable whereas relative driver, for instance, perform an endless
displacement is conceived only as a cue. The series of downward drifts and upward
seneral theory of retinal deformation as a type
of visual kinesthesis covers all special cases jerks as he fixates points on the flowing
of head-movement. road ahead of him, with occasional jerks
". I Ce.
, -...-
"
FIGURE 56. Motion Perspective with an Overcast or Ceiling
FIGURE 57. The Gradient of Flow Looking to the Right

when the Observer Fixates a Spot on the Terrain
in other directions to pick up objects of image might be supposed to be quite dif-

interest at the side of the road. The ferent from that when the point of fixation
eyes of a traveller seated on the right- is on the horizon, the perceptions which
hand side of a moving train make an end- result in these two cases are equivalent.
less series of drifts to the right and jerks The reader may check this observation for
to the left. The drifts are known as himself, with a piece of ruled paper on a
pursuit movements, and their function is table top substituting for the terrain. It
to maintain the image of a sele cted moving is as if the eye's movement added a
spot or object at the center of the retina. constant velocity toward the left, in the
The jerks are the type known as saccadic sense of vector addition, to the gradient of
movements, the general function of which velocities to the right shown in the former
is to establish a new fixation, and they oc- diagram. The horizon and the clouds now
cur in the act of scanning the environment, move across the retina at the rate which
in reading, and between the fixated pursuit the point of fixation, F, formerly possess-
movements just described. The question ed, but in the opposite direction. The
we now need to ask is this: What is the gradient from the horizon downward of
effect of the pursuit movements on the decreasing flow to the left is equivalent
gradient of retinal flow which we have as- to the former gradient of increasing flow
serted to be a geometrical correlate of to the right. Positive and negative
distance in the environment? velocities may be added algebraically,
The problem is similar to the one en- and the gradient of motion remains constant
countered in connection with crossed and on the retina whether the zero point of the
uncrossed retinal disparity as a correlate gradient corresponds to the horizon, to a
of distance, and the solution is of the point halfway down the terrain, or even to
same sort. Figure 57 represents the dia- a very near object on the ground. We
gram of Figure 54 as it would be modified know from physics that motion is relative
when the observer fixates a spot not on to an arbitrary zero point, or frame of
the horizon but on the ground halfway reference. The stimulus-motion of which
down. This is the scene frequently ob- we are speaking is a physical, not a
served from a train window. The point of phenomenal, motion.
fixation is indicated by F. At this point The phenomenal motion of objects is
the velocity of projected flow is zero, not ordinarily perceived as relative to an
since the pursuit movement of the eye arbitrary frame of reference. The stable
compensates for its change in direction visual world provides an absolute zero and
and keeps its image stationary at the hence there is an absolute sense of motion
center of the retina with a considerable or rest with reference to the ground both
degree of precision. The flow is also for oneself and objects. This fact is what
zero on a line to the right and left of makes the theory of the relativity of
point F; all points above this line flow to motion in physics difficult for the non-
the left and all points below it flow to the physicist to comprehend. Only in ex-
right. Although this deformation of the ceptional circumstances (such as looking
out of a train window at an adjacent train movement of the eye. The two kinds of
which fills the entire visual field) does it horizon coincide only when the eyes are
ever become equivocal whether the ob- fixed on the physical horizon in the so-
server is moving and an adjacent object called primary position of optics. The
is at rest, or whether the opposite is true. horizon of motion is not a limiting value,
In. such circumstances there is no terrain like an asymptote, but a zero value on a
surface on which to anchor the visual scale of opposites. Whereas size in the
world. The same kind of equivocal image varies on an intensitive scale with
motion can occur for an observer in a com- an intrinsic zero, motion in the image
pletely dark room who is presented with a varies on an oppositive scale with an
single slowly moving point of light (28, arbitrary zero (38 p. 223).
summarized in 32). In fact, with no visual Retinal Deformation with a Surface not
stimulation except a fixated light-point Parallel to the Line of Locomotion. The
both motion and position of the point be- foregoing descriptions and diagrams of the
come indeterminate, and a physically deformation of the image apply only to the
stationary point may appear to make ran- case in which the movement of the head is
dom excursions in any direction. This is in a line parallel to the material surface
the well-known autokinetic phenomenon. projected. Although this applies to normal
Its explanation, as Koffka understood ((,7), locomotion with reference to the ground, it
is that in the absence of ordered visual does not apply to all locomotion, nor to
stimulation the point may as well seem to movement of the head with reference to the
move as not and therefore sometimes does! slanting surfaces of objects. A particular-
The effect of pursuit movements of the ly important practical application is to the
eyes on the deformation of the retinal locomotion of a flier who is approaching
image during locomotion, we conclude, the ground at a certain angle of glide
is to add a constant to the motion of each preparatory to landing his plane. Dis-
point in the image but not to modify the tance perception, we are reminded again,
gradients which are its essential charac- is no mere visual luxury to be enjoyed in
teristics. The variable which corresponds parlor stereoscopes but a biological
to physical distance in the environment necessity, one use of which is to enable
must therefore be a gradient of motion-in- us to get about without colliding with
a-certain-direction, not a simple gradient obstacles (41).
of velocity as such. The horizon in the When an observer approaches a surface
retinal image is a line which is determined instead of moving parallel to it, a modi-
by vanishing values of the stimuli of size fication of its deformation is introduced
and texture, and it has no intrinsic rela- in that the focus of expansion is no long-
tionship to the gradient of motion. The er on the horizon of that surface but at a
latter gradient may have a kind of horizon particular spot on it the point of col-
of its own at the anatomical level of the lision with the surface. The rule is that
fovea, that is, at the midline of the retina all deformation in a forward visual field
where motion vanishes during a pursuit radiates from this point. Crudely speak-
ing, the environmental scene expands as aircraft carrier, the direction and rate of
we move into it, and the focus of expan- flow of the ground are determined by the
sion provides us with a point of aim for focus of expansion and the horizon in
our locomotion. An object in our line of combination. The velocity increases out-
travel, regarded as a patch of color, en- ward from the focus but then decreases
larges as we approach. It is not dif- and approaches zero at the horizon. Figure
ficult to understand, therefore, why this 58 illustrates the field of a flier who in-
expansion should be a stimulus for sensed tends to land on a runway. The gradients
locomotion as well as a stimulus for sens- of motion are approximately represented
ing the lay of the land. The behavior in- by a set of vectors indicating direction
volved in steering an automobile, for in- and rate at various points. All velocities
stance, has usually been misunderstood. It is vanish at the horizon. The focus is a
less a matter of aligning the car with the projection of the point on the ground at
road than it is a matter of keeping the which the glide is momentarily aimed; if
focus of expansion in the direction one the glide is steepened the focus will
must go. move downward in the field and if it is
Vrhen the focus of expansion is a spot made shallower the focus will move up-
on a vertical wall toward which a man ward toward the horizon. It is therefore
walks, the flow is zero at that spot and an indicator for the pilot as to where his
increases symmetrically around it, if we wheels will touch the runway and hence
disregard eye movements. When he ap- it is a cue as to whether he is under-
proaches it at a slant, the flow is cor- shooting or overshooting the field.Inas-
respondingly asymmetrical, the velocity much as either alternative is frequently
becoming greater on the near side. When fatal, the pilot has a vital interest in such
a pilot comes into an airfield or on to an cues. The working out of their interrela-
N\ t
-...-....
4,-----"'
FIGURE 58. Gradients of Deformation during a Landing Glide
tions was a part of the research on avia- doubt improve the stability and compre-
tion psychology during World War II (39, hensiveness of the visual world and his
Ch. 9).3 orientation in it. Of this there will be
Complex discriminations of the direc- more to say in the next chapter. Learning
tion, altitude, and angle of flight, whether to attend to novel features of the world, to
consciously perceived or unconsciously explore it, is something which psycholo-
incorporated in the reactions of the pilot, gists do not understand at present. What
are undoubtedly learned. But it is the the pilot cannot be supposed to learn is the
discriminations that are learned, not the impression of depth, distance, and alti-
stimuli on which they are based. The tude considered as an inference derived
retinal image of the novice flier is exactly from unlearned sensations.
like that of the experienced pilot; the Evidence that Retinal Deformation is a
difference is that the latter reacts differen- Stimulus Variable for Space and Locomo-
tially to variations of the image to which tion. Although relative motion in the
the former does not. The experienced pilot visual field has always been accepted as
does not see more of the visual world a factor in distance perception, the
than the novice but he sees a more dif- theory of the retinal motion gradient as a
ferentiated visual world. The effect of stimulus for distance perception requires
his training is to enable him to make fine proof. The theory makes possible ex-
rather than gross discriminations of dis- periments which will either verify it or
tance, altitude, angle of glide, angle of not. So far, only the theory has been
drift, speed of flight, the position and constructed actually no more than the
direction of everything, and therefore to outline of a theory and the experiments
see accurately a continuous visual world are lacking. What is required is a proof
in which he himself moves with precision. that an artificially produced gradient of
In this sense, and only in this sense, is point-motions on the retina, in isolation
space perception a product of learning. from other gradients, will yield an ob-
To a certain extent the pilot can be server the impression of continuous dis-
supposed to learn habits of scanning or tance on a surface. Variations in the
inspecting the environment, habits having gradient must then be shown to produce
to do with eye-movements, and these no variations in the slant of the surface.,
This demonstration has never been set
3
in general are little inclined
Since fliers up and performed as an experiment, inas-
toward introspection, the analysing and de- much as the gradient of velocity has not
scribing of such appearances in the visual been isolated from gradients of ordinary
field is almost absent in the literature of avia-
tion. A notable exception is Langewiesche's perspective. An interesting method of at-
book, Stick and Rudder, which describes and tack on it would be with frame-by-frame
illustrates the visual cues for landing in a way
consistent with the theory here presented, and photography of spots, lines and other tex-
is full of acute observations on the space per- ture-stimuli, that is, by animation of pic-
ceptions of the flier (70). Pfaffmann is another
exception in that he has analysed the tradi- tures. Such evidence as does exist comes
tional binocular cues experimentally (87). from a program of wartime research which
tional binocular cues experimentally (87). from a program of wartime research which
exception in that he has analysed the tradi-
tures. Such evidence as does exist comes
ceptions of the flier (70). Pfaffmann is another
ture-stimuli, that is, by animation of pic-

is full of acute observations on the space per-
consistent with the theory here presented, and
photography of spots, lines and other tex-
illustrates the visual cues for landing in a way
tack on it would be with frame-by-frame
book, Stick and Rudder, which describes and
tion. A notable exception is Langewiesche's

perspective. An interesting method of at-
field is almost absent in the literature of avia-
been isolated from gradients of ordinary
scribing of such appearances in the visual
toward introspection, the analysing and de- much as the gradient of velocity has not
Since fliers in general are little inclined
3 up and performed as an experiment, inas-
This demonstration has never been set
to do with eye-movements, and these no

variations in the slant of the surface.,
inspecting the environment, habits having gradient must then be shown to produce
supposed to learn habits of scanning or tance on a surface. Variations in the
To a certain extent the pilot can be server the impression of continuous dis-
space perception a product of learning. from other gradients, will yield an ob-
In this sense, and only in this sense, is point-motions on the retina, in isolation
in which he himself moves with precision. that an artificially produced gradient of
a continuous visual world are lacking. What is required is a proof

see accurately
the writer directed in 1943-1946 (39) on

the space perception of fliers and on the
use of motion pictures to represent space.
One goal of this research was to find out ,,,,,
'At ,4k.,* , :;400
how' effectively the perception of a three ,
: !
vi,14:1 4144f 1410,4
"4iitn if .t. V
Ak ,'
dimensional world can be aroused by the '''.'74

I lie
4,
leie,010,.:,i4
°'
%i, '',49i ''',$:',.'1;
flat motion picture screen and by the two 1"10qt4 4100r ;IA 1'!
dimensions of a still photograph. It be-

came evident that both still and moving ,1,),,
i' h
l'','''t 4'"'''
pictures can yield a more adequate visual

world than is generally recognized or
understood. The theories of texture-per-
spective and motion-perspective are de- Q.,14t44,44-"'
rived from this evidence.

If we take a motion picture shot, for ,
instance, of the scene ahead of an air-

plane during an approach glide (Figure 58)
and project it on a screen, the projection
can be studied in two ways. The experi-
4k.
menter can plot by rough measurement
the actual gradients of expansion on the
screen. When this is done, the results 4011 ,
are approximately those represented in

Figure 58. Also the experimenter can 7+1
,,,,1,1'09/4,,,,,,f,j11,1.1
present the screen picture to a group of

trained fliers or of untrained persons and
have them describe what they see, or 11
identify the focus of expansion, or make

other judgments. When this is carried
out, the notable result is that no observer,
trained or untrained, perceives an expand-
ing screen picture and nothing more. Each
onlooker gets a compelling experience of
*SO'
moving through space in a specific direc- 17 hh r
tion toward the ground. The angle of this

movement and the point of its aim can be
FIGURE 59. A Landing Field

during an Approach, at 1000 Feet, , h r
at 500 Feet, and at 250 Feet from

'the Edge of the Runway
judged by all with varying accuracy, it motion, so far as locomotion is correlated

is true, but nevertheless judged. The ex- with visual stimuli. It would clarify the
periment was often repeated and the ex- matter if we could classify the different
perience proved to be sufficiently reliable kinds of stimulus-motion that may occur
to justify making a series of such motion in the retinal image and specify the
picture shots into an experimental test for physical situations which produce them.
measuring the aptitude of candidates for
flying training, and even for measuring one
We have already seen that the retinal
aspect of the proficiency of trained pilots image is a two dimensional projection of
focused light on a sensitive anatomical
(39, Ch. 9).
The examination of many other motion surface. When we say that it undergoes
motion, we must always mean motion
picture shots taken from moving aircraft,
including views at right angles to the line with reference to that surface. One possi-
of flight, yielded the conviction that the bility is for the projection as a whole to
gradients of motion we have described are move reference to the surface.
with
genuinely compelling for the experience of Another possibility is for a delimited
distance. Many of the phenomena of ani- part of the image to move with reference
mated cartoons point to the same con- to the total stationary image (usually
clusion. When children observe Donald the surrounding image), or for the sur-
Duck diminishing rapidly in size on the rounding image to move with reference to
screen, moving upward slightly, and re- a stationary delimited part. All three of
taining the same shape, it is likely that these kinds of motion may be defined
they actually see him whizzing off into the geometrically as rigid motions. A rigid
distance. The experimental isolation of motion is one of translation or one of ro-
these phenomena, however, remains to be tation or a combination of these, but not
carried out. one which involves any change of shape or
distribution. A square which moves across
The Types of Retinal Motion the retinal surface, whether it moves with
Ordinarily when we speak of perceiving all its points tracing straight lines or
motion we refer to the movement of an tracing arcs of circles, remains a square.
object. The problems connected with the A total image which moves rigidly across
motion of an object's image across the the retina keeps the same distribution al-
retina have been studied for many years. though it loses old and gains new parts at
What we have been talking about, however, opposite margins of the retina. This
is the movement of the observer and the ac- rigid kind of motion relative to the retina
companying deformation of the environ- has an analogy with the motion of physical
ment's image on the retina. The latter objects solid rather than liquid sub-
has never been fully described and has stances and it is the kind analysed in
been little investigated. Our concern has classical physics and dealt with by the
been with the perception of locomotion laws of motion. But it is not the only or
rather than the perception of ordinary even the most frequent variety of motion-
in-general which the retinal image may object in the physical environment moves,
undergo. The image or any part of it may with reference to the ground, at right
be expanded, contracted, or skewed in a angles to the line of sight.4 Motion of the
number of ways. All of these ways are surrounding image occurs with the eyes
definable in terms of gradients of velocity in a pursuit movement when an object in
and direction along one or another axis of the physical environment moves at right
the retina. They are also definable in angles to the line of sight. (The percep-
terms of mathematical transformations of tions resulting from these latter two modes
one image into another. This is the kind of stimulation are surprisingly equivalent).
of motion which has been termed deforma- Deformation of the total image occurs
tion. As with rigid motion, deformation with eyes fixed on the horizon when the
may apply to the retinal image as a whole head of the observer moves in relation to
or to a delimited part of it. the ground (locomotion). Deformation of a
These types of generalized retinal mo- delimited part of the image occurs with
tion can be put into a very simple cor- stationary head and eyes when an object
respondence with the physical situations in the physical environment moves in
which produce them. Motion of the re- depth in any direction not at right
tinal image as a whole occurs with sac- angles to the line of sight. The rela-
cadic eye movements from one fixation to tionships can be summarized in the form
another, the head and body being station- of a table.
ary. Motion of a delimited part of the 4 Actually, the retinal motion is non-deforming
image occurs with fixated eyes when an only when the object crosses the line of sight.
Mode of Perception of
Retinal Motion. Physical Situation Objective Movement
Rigid motion of the Saccadic eye-move- None, Perception

total image ment in stationary of a stable world
environment
Rigid motion of a Stationary eyes, ob- Object moving fron-

delimited image ject moving frontally tally in stable world
Pursuit movement of
Rigid motion of the eyes with object Object moving fron-
image except for a moving frontally tally in stable world
delimited part
Movement of head in
stationary environ- Movement of self
Deformation of the in stable world
total image ment
Stationary eyes with Object moving in

Deformation of a object moving in depth in stable
delimited image depth world
These modes of stimulation are to be gether, they need not do so. The clearest
understood as individual cases and not as demonstration of this fact is the negative
alternatives. They may coexist in many after-image of motion. One gets it after
combinations, and the resulting percep- fixing one's gaze at a waterfall for ten or
tions will also apparently coexist without twenty seconds and then staring at another
interference or confusion. A man driving part of the scenery. It is sometimes ob-
an automobile, for instance, can see his served out of a train window after the train
own movement and at the same time per- has stopped. A better method is to watch
ceive the movement of a car approaching the apparent expansion of a rotating disk
on a crossroad without obvious loss in on which a spiral line has been drawn so
the clarity of either perception. The de- as to make the line appear to move out-
formation of the total image, considered ward from the center. The after-image of
as a set of gradients, appears to lose none motion is confined to the stimulated area
of its stimulus function even though a of the visual field (like an after-image of
number of other motions be summated with color) and it consists of a vivid impres-
it. sion of motion in the opposite direction,
as an lhdependent Variable of
illotion persisting for some seconds and only
Experience. The theory that all visual gradually dying out. The fact to be noted
experience was made up of elementary is that an unmistakable motion is per-
sensations of flat color, each spot having ceived but that it does not involve any
an elementary quality of location, implied change of position. When we look away
that motion was a perception which could from the waterfall the foliage appears to
always be analysed into a change of loca- move upward but it is not displaced. Out
tion of a color in time. Visual movement of the train window the ground seems to
was not a simple but a complex experi- move forward but not to shift position
ence, the sense of location being the pri- relative to the window. After the rotating
mary fact on which motion depended. It disk has been stopped it seems to con-
seemed to be only reasonable that one tract, but it does not get smaller (although
could not see a thing as moving if one it may yield a queer impression of reced-
did not see it in successively different ing or retreating from the observer). One
places. cannot avoid the conclusion that here is
The localization theory of space-per- an experience of motion without an ac-
ception, it has already been argued, was companying experience of displacement in
inconsistent with a good many facts of the frontal plane.
experience but nowhere was it more false The same type of experience may be ob-
than in this corollary. For, strange as it tained by having an observer fixate the
may seem to someone who has not ob- center of a semicircular screen (a peri-
served it for himself, an impression of meter) and presenting motion in an area or
motion may be got without any impression window at the far periphery of his visual
of change in location. Although under field. He may sense the motion correctly,
ordinary circumstances the two go to- as compared with no motion, but without
any impression of what the moving pattern jacent spots at successive moments with
might be, and even without any impression a considerable separation between the
of what the direction of the movement is. spots and the moments. There must be a
An even more notable example of this kind combination of adjacent and successive
of experience, because of its historical order but neither the adjacent series nor
importance, is what has been called the the successive series has to be "dense,"
phi-phenomenon (118). Successive ex- or "compact" as the number-theorists say.
posures of two separated black spots will In short, the essential stimulus for motion
yield stroboscopic movement, one spot is not physical motion.
jumping into the position of the other. The rule to apply is that the stimulus
When the timing of the exposures is not for a spatial impression need be only a
optimal, however, the impression reported correlate, not a copy, of the corresponding
is one of movement without any moving physical variable. Physical motion, like
spot, or "pure phi." Taking all these physical shape and physical depth, does
instances together, it seems certain that not have to be duplicated in the retinal
there must exist a visual quality of what image in order to yield phenomenal motion.
might be called movement-as-such. The paradox of the stroboscope and the
A combination of successive and ad- motion picture that we can see motion
jacent order over the retinal mosaic would in a situation where nothing moves is
seem to be the fundamental stimulus condi- no longer a paradox if this rule is ap-
tion for this impression of motion. The plied. One has only to relinquish the as-
stimulus may be complex, in a mathematical sumption that a stimulus must be a re-
sense, but the impression is simple. If plica.
motion is a variable of experience which
can exist in its own right independent of The Perception of Acceleration as Dis-
the displacement of an object, there is tinguished from the Perception of Notion.
probably also a variable of stimulation Normal Locomotion. It should be noted
to which it will prove to correspond. The that the relationships of the types of
exact definition of this stimulus variable retinal motion in the table refer to the
remains to be established. One fact about perception of motion as such, which must
it, however, can be stated with some cer- be distinguished from the perception of ac-
tainty: the stimulus for motion is not celeration or force. Perceived motion of
necessarily motion in the retinal image. one's own body seems to be mediated best
To put it more precisely, it is not neces- by vision, whereas perceived acceleration
sary to have the stimulation of a continu- and perceived force, which are actually
ous series of adjacent points at a con- the same thing, are mediated by stimula-
tinuous series of successive instants. tion from within the muscles (the muscle
(Points and instants are fictions, in any sense) and stimulation from within the in-
event, as the paradoxes of Zeno ultimately ner ear (the labyrinthine sense). The
demonstrated.) Instead, the stimulus retina is insensitive to forces acting on
condition for visual motion may be ad- or within the body, and also insensitive to
an acceleration of the body, that is, a and many kinds of tool-using would all be
change of motion with time. On the other included in the category of active loco-
hand the inner ear is extremely sensitive motor behavior, for what we mean by it
to any force acting on the body, gravity is behavior accompanied by the percep-
for example, and to any acceleration of tion of moving from one place to another.
the body, but is wholly insensitive to As experienced, it includes sensory im-
uniform motion. The air traveller who pressions of exertion or muscular action
floats in the sky at a uniform speed of and of visual motion, the two being co-
two hundred miles an hour is witness variant. The product of these stimuli is
to this latter fact. Stimulation of the something neither wholly motor nor wholly
inner ear and of the receptors in muscles visual; it is locomotor action in a visual
do not provide a kinesthetic sense in the world. As we shall try to show in the
literal meaning of the term, but a force next chapter, the perception of a stable,
sense. Visual stimulation, with the upright visual world also depends on co-
possible addition of stimulation of the variation of the visual sense with the so-
joints, is a better mediator of the motion called body senses, and subsequently we
sense.5 may have a better understanding of why
When a man walks, or, as we say, gets locomotion and the stability of space are
about under his own power, he experi- intimately connected.
ences both motion and force. The per- Not all locomotion, however, is active
ception of voluntary locomotion, as dis- or voluntary. Driving a car or flying an
tinguished from passive locomotion, is airplane is a relatively passive action com-
jointly determined by two sources of pared to walking, and for the passenger it
stimulation, stimuli from the retina, on it still more passive. Modern man has
the one hand, and from the muscles plus gone to great lengths to save himself
the inner ear on the other. In the study of effort during locomotion, and special
active locomotor behavior, an extremely problems arise in learning to use these
important branch of applied psychology, locomotor machines. The muscular ac-
neither of these sources of stimuli can be tions involved in steering, accelerating,
neglected. What has come to be called or balancing an airplane with the stick are
psychomotor behavior, for example operat- artificial and highly reduced actions as
ing a crane or a lathe, is only a special compared with walking or balancing one's
form Walking, running, athletic
of it. own body, and the visual stimulation be-
achievements, automobile driving, flying, comes proportionally much more important
than the bodily stimulation in these
5
This conception, which admittedly does relatively passive types of locomotion.
violence to the accepted meaning of the term Visual Motion during Rotation. The
"kinesthetic" will be extended in Chapter 13.
The above distinction applies only to bodily distinction between motion and accelera-
acceleration and motion, not to the motions and tion also makes possible a better under-
forces perceptible among visual objects. The
latter have been studied by A. Michotte in La standing of a kind of movement of the
Perception de la Causalite. observer which has not so far been men-
tioned rotation of his body and head. for example, are aroused not only by the
This is seldom, properly speaking, a retinal stimulation of movement but by
form of locomotion. Prolonged passive the inner-ear stimulus of acceleration.
rotation, as distinct from active explora- Either stimulus will produce eye move-
tory rotation of the head, is an artificial ments in isolation from the other, as many
situation not met with outside a laboratory. experiments have shown. During any
Motion of this sort has nevertheless re- initial rotation of the head, the accelera-
ceived a great deal of experimental study, tion and the retinal motion are concordant
usually employing a rotating chair in stimuli producing the same response. But
which the observer is seated. The eye during the stopping of this rotation, after
movements which go with it are called an interval of uniform rotation, the ac-
compensatory and they are collectively celeration is reversed and becomes nega-
known as nystagmus. When the head is tive, although the retinal motion continues
rotated the eyes move oppositely in such in the same direction during this period.
amount as to preserve an unchanged visual The compensatory eye movements may then
field, thus compensating for the rotation. cease to have a compensatory function;
As rotation continues, the eyes jump with they may be reversed in response to the
a saccadic movement to a new fixation negative acceleration, and this after-
and then follow the environment once nystagmus may even persist after the
more, repeating these slow and fast move- head is physically at rest. If the eyes are
ments as long as the rotation continues. open this kind of eye movement inevitably
Under the circumstances the observer produces an illusory motion of the en-
should perceive a stable visual world vironment. The world then appears to
with a sense of his head moving in it. Ac- rotate in the direction opposite the bodily
tually this is the result, but only if the rotation since the compensatory eye
rotation is not prolonged or rapid. When movements are compensating, as it were,
an observer is whirled in the usual ro- for a nonexistent motion of the head.
tating-chair experiment, the compensatory Other phenomena, however, may com-
movements soon cease to yield precise plicate this result.
fixations and the stability of the visual The illusions which occur when a sub-
world breaks down. It then appears to ject is rotated in partial darkness or in
rotate as a world, and the observer be- an environment showing only a few points
comes disoriented and dizzy. of light should be particularly strong, and
The visual appearances of the world they are of considerable practical im-
during the rotating-chair experiment are portance since this is the situation that is
complicated by the fact that the subject encountered in night flying. Graybiel and
is stimulated by accelerations which affect his associates have studied this situation
the inner ear as well as by retinal motions. and described what they call the oculogy-
In this artificial situation the two kinds ral illusion which, like the after-nystag-
of stimulation necessarily come into con- mus, is produced by compensatory eye-
flict. The compensatory eye movements, movements not having any compensatory
STI MULUS VARIABLES THE ACTIVE OBSERVER 137
function (45). Together with other illu- skyline or, in other words, the amount of
sions due to acceleration in night flying ground between the observer and the ob-
maneuvers, it may help to explain the oc- ject (19, p. 270).6 Sometimes these fac-
casional disorientation to the objective tors are called signs or criteria of dis-
world which a pilot suffers sometimes tance or, more frequently, cues. Whatever
with fatal results. the term used, it is clearly implied that
they are not precise geometrical correlates
Summary. The Sensory Analysis of Dis- of physical distance but probable indica-
tance and Depth
tors, symptomatic rather than exact.
The traditional list of clues by which In the attempts to describe these fac-
the mind is believed to infer a world of tors it has often not been clear whether
three dimensions usually includes the fol- they referred to sensory experiences, or
lowing factors: linear perspective and the characteristics of the retinal image, or
decreasing size of similar objects with facts about the physical object. These
distance, the apparent size of objects are three quite separate classes of facts
whose real size is known, superposition which should not be confused with one
of one contour on another or the covering another. The covering of a far object by
of a far object by a near one, the distri- a near one, for instance, is a description
bution of light and shade over an object, which mixes physics and experience. It
the relative motion of objects or monocular cannot explain depth perception, as
parallax of motion, aerial perspective and phrased, since it presupposes the pheno-
the loss of detail with distance, binocular menon which it seeks to explain one
disparity, or stereoscopic vision, degrees object behind another. The joint contour
of ocular convergence and of accommoda- in the retinal image is two-dimensional.
tion of the lens, and occasionally the How do we see depth at a contour so that
factor of angular location of an object be- one side of it appears near and the other
tween the bottom of the visual field and the far? That is the fundamental question.
Brightness is sometimes listed as a cue to lum in ated surfaces, therefore, brightness is
distance, the presumable assumption being that not an indicator of distance.
an object necessarily appears darker as its dis- The fact is, however, that in a dark room
tance from the eye increases (19). Apart from with no other sources of stimulation the more
the little-known effects of atmospheric condi- highly illuminated of two equidistant and
tions on visibility aerial perspective the otherwise equivalent surfaces tends to look the
assumption has no basis in physical fact. It nearer. This is an empirical fact which has
is true that a point-source of light yields an nothing to do with optics. It is not easy to
intensity at the eye which decreases in account for. It does not imply that brightness
proportion to the square of the distance. But is a clue, indicator, or sign of distance; it only
an illuminated surface (an infinite number of poses a problem. So far as the writer knows,
theoretical points of light) yields the same this empirical fact has never been observed
intensity per unit solid angle at the eye when except under darkroom conditions where dis-
it is far as when it is near, and hence possess- tance is relatively indeterminate and where
es the same brightness, within limits, under presumably the impression of distance may be
both conditions. Each theoretical point be- affected by minor determinants which are in-
comes theoretically dimmer, but the density of operative in a normal environment of il-
points becomes theoretically greater in exact luminated background surfaces. The experi-
proportion. In the ordinary environment of il- ment needs to be repeated.
The cues for depth have not been reduced have to be given a more arbitrary name;
to their components and precisely speci- they will be called sensory shifts. The
fied. They need to be analysed in terms first correspond with gradients of adjacent
of (a) the retinal image and (b) the cor- stimulation on the retina; the second with
responding impression in the visual field. abrupt rises or falls in such stimulation.
In this and the preceding chapter those The varieties of perspective can be listed
variations of the retinal image which somewhat as follows:
underlie the principal cues for depth have 1. Texture-perspective. This is a
been described. The theory is that they gradual increase in the density of the fine
are retinal gradients and steps of ordinal structure, the spots and gaps, or the ex-
stimulation and that they are geometrical- tended pattern of either a part or the whole
ly precise. As stimuli, they can be tested of the visual field. There is a great variety
for exact correspondence with impressions of textures for which no adequate names
of distance and depth in the perceptual exist. The increase in density may run in
experience we have called the visual any direction but very often it runs upward
world. If they are stimuli, it should also in the field. The impression of density
be possible to put them in correspondence turns into an impression of depth or dis-
with impressions in the relatively depth- tance during ordinary vision. This type
less visual field. In common terminology, of perspective has been illustrated in
they should correspond to sensations as Figures 23 to 26 and 32 to 35. It merges
well as to perceptions, although the into the next type.
meanings of these terms have been re- 2. Si ze-P erspective. This is a de-
versed by the argument that the perceptual crease in the size of the shapes or figures
impression is the primary one, immediate in the visual field when it is considered
and independent, and the sensory impres- as an array of color-patches. It presup-
sion the secondary one, obtainable only poses contours, or figures on a background,
by analysing the perception. What are each of which may have its own texture-
these secondary sensory impressions, perspective.
then, which depth and distance look like 3. Linear Perspective. This is size
when they are not seen as depth and dis- perspective when contours are rectilinear.
tance? What are the features of pictorial It is a gradual decrease in the spacing
vision which parallel the three-dimen- (the size or dimension) between either out-
sional features of normal vision? A lines or inlines in the visual field. Since
classification of them should serve as a the edges of things in man-made environ-
useful substitute for the list of the cues ments are so often straight and since
for the third dimension. straight lines are easy to draw, this kind
The sensory impressions which go with of perspective is the one we have learned
the perceptions of distance or depth over a to notice and pay attention to. It has been
continuous surface might all be called illustrated in Figures 36 to 38. All these
varieties of perspective. Those which go perspectives can. decrease to a zero limit
with the perceptions of depth at a contour of size or spacing(or to a maximum density
FIGURE 60. A Road Slanting Down and then Up
Can you find two gradients of converging lines in the drawing in addition to the gradient
for horizontal converging lines? The picture was constructed with three different vanish-
ing-points. (Adapted from H. Buckley, Perspective. London, Sir Isaac Pitman & Sons,
Ltd., 1947)
of texture). A horizon in the visual field of double imagery for the elements of a
is, introspectively, a line at which these texture, therefore, goes unnoticed. It is a
limiting qualities are reached. change from a maximum of crossed double
When these three kinds of perspective, imagery toward uncrossed double imagery,
usually in combination, give way to the as illustrated in Figures 48, 49, and 50.
perception of a continuous surface such as For any given plane surface there is a line
the ground, the decrease in spacing or of single imagery which is perpendicular
size (or the increase in density) gives way to the direction of change and which al-
in turn to the perception of a constant ways passes through the center of the
spacing, size, and density. There results visual field. The stimulus to which this
an experience which will be described in kind of perspective corresponds is a gra-
Chapter 9 as an impression of constant dient of the horizontal skew of one retinal
scale. image relative to the other, i.e. of the rela-
4. Binocular perspective. This is the tive disparity at a given point. Both the
perspective of double-images in the visual gradient and the graded double imagery
field and it is almost impossible to observe disappear when the observer closes one
except at contours. The continuous change eye.
' .44-W4telovA1}A., kf* 44$4,48012
; Orkt 4,141At,4*.48f08 I 44, la 4 Y.,/ 1' 10 I 414i4.)1'

'
kki4414.., 1 '1
*A*
0,4
ykr 4
ILL
.8.1!,° t.'
Wk
41644' .4
k
A0.4,4k, kk5 k kick,
" A
V.M14t"
0404"."44i/k kV
- 4? 40,1
i4 4.4 14
1.
0..74 0
15 y
/1.
44
k"Vi
+IL
'kkt
L'
"48 44 't op. go, 0.6,

44 41,
004,4*;,'FA',441t4+'tS444k7,i "'4
.4t, .4
Or'
'
..,-,1T:44,0'.,1;,."
/ /
,ory
itrI, ,,.. Z.. . .704..4,
.4 4,., . k 1/4.4''.1141:r1 .. " a.7/1iC l'

44 "re 4 4 r4,t;t4' 0 `,104. i' -40
I .. ; I.:8,,, :: 'GriP: ,....:... ''''
YAK :: 4 ....r.:;; rit; '-...1.44. 1; i,t 4'..441.'
. 1, I'''''' 1(4.
,14 4.-
,04* Of el....r4
-.<' ' 'j-
It .
*.
4, 484
1414
y
.. 4,4 * '' *
4,4*. .'''': :4'.:::::1:*:: 14'"gs,'4.
14°4..
.. 71::?::: 'II's::: 7
e t*, % :0"). r'4,14+04. 15.1.AO '41.414*
k ,,k, ,..... 1111
..4 470 itillr '
Arc..11t .-04-90 1,,:,/ 444;07...4 ,A,14ta St 4*
..hN1144d4it S ,k4.*'
47..11'"'jj
y1
4", .'"Ple "re'4;
'i4 '''''' ;' 1°,oe. ., '.4: 1,174*'''''': :41...., 1 f' ' ilklf*::''
FIGURE 61. Two Scenes at Five Feet From the

Ground and at Twenty-Five Feet From the Ground
The views on the left were taken at the height of a man standing, those on the right from a
high ladder, but at the same spot on the ground. Note that there i s a steep gradient of
density in the left-hand photographs and a moderate gradient in the right-hand photographs.
Note also that the more moderate gradient of density, in either case, produces an im-
pression of a surface less inclined to the line of regard than the steeper gradient. Since
the surface is interpreted as being level, the resulting impression is that the point of re-
gard has moved upward. The right-hand photographs locate the observer high in the air,
looking down. AnOther instance of the relation between the point of view and the slant of
the surface is given later, in Figure 70.
Motion-perspective. This is a gradual

5. When the vanishing line of motion coin-
change in the rate of displacement of tex- cides with the horizon of texture, size,
ture-elements or contours in the visual and spacing, the perspective of motion be-
field. The change is from motion in one comes easier to specify and describe. In
direction through zero-to motion in the op- that event the rate of deformation in the
posite direction, and it also has a vanishing visual field simply decreases from the peri-
line, at right angles to the gradient, which phery toward the center of the field. The
passes through the center of clear vision. directions of the motion radiate from and
passes through the center of clear vision. directions of the motion radiate from and
phery toward the center of the field. The

line, at right angles to the gradient, which
visual field simply decreases from the peri-

posite direction, and it also has a vanishing
that event the rate of deformation in the

direction through zero-to motion in the op-
field. comes easier to specify and describe. In

The change is from motion in one
and spacing, the perspective of motion be-

ture-elements or contours in the visual
change in the rate of displacement of tex- cides with the horizon of texture, size,
When the vanishing line of motion coin-

5. Motion-perspective. This is a gradual
the surface is given later, in Figure 70.
looking down. AnOther instance of the relation between the point of view and the slant of
in the air,
gard has moved upward. The right-hand photographs locate the observer high
the surface is interpreted as being level, the resulting impression is that the point of
re-
pression of a surface less inclined to the line of regard than the steeper gradient. Since
Note also that the more moderate gradient of density, in either case, produces an im-
density in the left-hand photographs and a moderate gradient in the right-hand photographs.
on the ground. Note that there i s a steep gradient of

high ladder, but at the same spot
on the right from a

The views on the left were taken at the height of a man standing, those
Twenty-Five Feet From the Ground

Ground and at
at Five Feet From the

FIGURE 61. Two Scenes
47..11'"'jj
4",
.'"Ple"re'4;
'i4
'''''' ;' 1°,oe. ., '.4: 1,174*'''''': :41...., 1 f' ' ilklf*::''
toward a pair of opposite poles (Figure 55) which are sometimes still listed as a cue
which are specific to the physical motion for distance.
of the observer himself. The resulting per- 8. Relative upward location in the
.
spective of expansion, skew, or contrac- visual field. It has occasionally been sug-
tion as the case may be is not difficult to gested that the amount of background the
observe, but it strongly tends to pass over angular extent between the lower margin
into a perception of continuous distance of the field and a given object is a clue
or space. This perception is vivid and to its distance. The rule seems to bold
compelling. for objects represented in pictures. Ob-
Binocular and motion-perspective might viously, however, the clue is valid only
be called perspectives of parallax whereas when the background is taken to be the
texture, size, and line-perspective are terrain rather than a wall or a ceiling. It
perspectives of position. Next to be listed serves mainly to illustrate the fact that
are three which are independent of the ob- the effective stimulus gradients in outdoor
server's motion or position. vision are usually those produced by the
ground. Upward location in the visual
Aerial perspective. This is an in-
6. field does not correspond to any gradient
crease in haziness, blueness, and de- on the retina but only to a dimension of
saturation of colors over the visual field. the retina on which a gradient of stimula-
It has not been measured or precisely tion might occur. We shall recur to this
described. Unlike the other forms of phenomenon in Chapter 9.
perspective it is variable with the condi- The foregoing eight varieties of per-
tions of illumination and it does not rest spective all have reference to distance
on the geometry of optics. For this reason over a surface or an array of surfaces.
it seems improbable that it will ever prove There is another fundamental kind of dis-
to be a stimulus for the impression of distance perception, however, which was
tance, although it may be an indicator. called depth-at-a-contour. The visual
The perspective of blur. This is a
7. field contains only patches of color. What
decrease toward the center of clear vision are the differences between these color-
of the quality of blur. Blur depends on areas which yield depth?
texture elements and contours in the visual 9. Shift of texture-density or linear
field. It is difficult to observe since the spacing. This is a change in the density
out-of-focus quality is never at the center of texture or the spacing of inlines which
of vision for a normal eye. When the lens is sudden rather than gradual (Figure 41).
is accommodated for any considerable dis- It is usually coincident with a change in
tance, however, the gradient of blur tends brightness or color in the visual field such
to level off so that one may doubt whdthet as to produce a contour or a segregated
it could serve as a univocal stimulus for form (p. 65). The parallel perception in
the impression of distance. It is impor- the visual world is a recession in depth
tant only because it is more fundamental on the side of the increased density or
than the sensations of accommodation decreased spacing. The effect is like
looking at a valley beyond the edge of .a move across the background that part ap-
cliff on which the observer is standing. pears to be in contact with the background,
10. Shift in amount of double imagery. or resting on the ground. The shift in rate
This occurs when the texture-elements on of motion at a contour goes with a per-
one side of a contour are seen less ception of one surface behind another, and
doubled than those on the other side. The the amount of retinal shift is theoretically
contour itself may be seen in single im- an exact stimulus for the perceived dis-
agery (as any contour is when it is hori- tance between the surfaces.
zontal in the binocular visual field). When The phenomenon of the superposition of
the observer takes a normal perceptual objects is actually not a clue to the depth
attitude he sees that side of the contour of objects but a perception which requires
more distant which manifests a relative explanation. A man knows that a near ob-
shift toward uncrossed double imagery. ject can partially obscure a far object but
11. Shift in the rate of motion. This his retina does not, and the retinal ex-
occurs during a movement of the obser- planation should be sought first. The pre-
ver's head and consists of more rapid ceding three factors supply such an
displacement of texture-elements on one explanation. The drawings of Figure
side of a contour than on the other. If 62, however, suggest that there are
the contour is closed, the shape appears other factors, involving the shape of
to move across the background. If part of the contours, which help to determine
the contour (the bottom usually) does not superposition. There is no texture, double
FIGURE 62. One Object in Front of Another
STIMULUS VARIABLES THE ACTI VE OBSERVER 143
imagery, or relative motion in these draw- of the physical surface at the corresponding
ings. They suggest the principle that the point.8 An abrupt change of rate is the
more complete, continuous, or regular out- projection of a corner (Figure 40). A
line tends to be the one which looks near. gradual change of rate is the projection of
Is completeness, then, a sign or clue for a curve. If this formula holds for the
distance? density of texture it should also hold for
12. Completeness or continuity of out- binocular disparity and rate of motion;
line. We can reasonably assume that, if the slope of these gradients is also geo-
objects tend to have regular outlines, com- metrically linked to the slant of the sur-
pleteness, closure, or continuity7 tends to face projected (cf. Chapter 9). A sup-
be associated with the near side of a com- plementary explanation for the modelling
mon contour and incompleteness to be of the visual world, however, is provided
associated with the far side. At the visual by the relation of light and shade to the
contour between an intercepting object convexities and concavities of the en-
and an intercepted object the side belong- vironment.
ing to the completed outline usually has 13. Transitions between light and
the coarser texture, the greater relative shade. An abrupt shift in the brightness
motion, and the greater crossed disparity. of adjacent regions within the visual field
Consequently, a visual contour in isola- produces a contour, which is the necessary
tion from these sensory shifts suggests condition for a segregated shape or form.
greater depth for the incomplete outline A shape in the visual field is coordinate
and lesser depth for the complete one. with an object in the visual world. This
The connection could be learned. rule is complicated, however, by the fact
Besides continuous distance and depth- that there may occur shifts in brightness
at-a-contour there is another major feature which produce not contours but the model-
of three-dimensionalitythe shape of an ling of a surface. These are the bright-
object in depth. The solid modelling of a ness differences which we call differences
surface, its protuberances, indentations, in light and shade.
corners, curves, or flatness, is something Light and shade, for reasons imper-
which applies to backgrounds as well as fectly understood, are not the same quali-
to things. The principal explanation of ties in perception as white and black. The
this modelling of the world is probably the color of a surface and the illumination of
formula advanced in Chapter 6 for the it are perceived separately, although both
slant of a bounded surface: that the rate of must ultimately depend on the light-
change of texture-density at any point in a energy of the retinal image. Whatever may
projected image is proportional to the slant be the explanation of this paradox, the
7 8
Ratoosh has given a mathematical formula- The slant of a surface, in this context, is
tion to the phenomenon of the continuity of an its slant with respect to the line of sight not
outline at its intersection with another outline the line of gravity. It is the angle of con-
(89). frontation of a surface.
transitions between light and shade seem variable of the retinal image should be
to be capable of giving a surface the considered a stimulus-correlate for depth
quality of shape in the third dimension as or not.
distinguished from its shape in two dimen- The traditional cues for depth, to sum-
sions. This fact has been illustrated in marize the last two chapters, can be re-
Figures 43 and 44. stated as variables of the retinal image.
The relationship between these transi- They can also be described as they appear
tions of shading and the corresponding when one observes his own visual field,
depth-shapes is not well enough known to and can even be called sensations so long
be specified in psychophysical laws. A as one is careful to remember that they are
gradual transition, it is true, yields a only the visual symptoms of stimulation
curved surface and an abrupt transition an and not the causes or elements of percep-
angled surface, or corner, but whether this tion.
depth-shape is concave or convex depends Eight of the thirteen listed can be
on complex factors. The geometry of light thought of as stimuli for perceptions of
and shade changes as the direction of the space. The remaining five are better con-
illuminating light changes. Consequently ceived as probable signs, secondary to
it is impossible to decide whether this the others, or as having doubtful status.
The Problem of the Stable and
Boundless Visual World
The Stable Visual World .. .. The Problem of

the Unbounded Visual World
The distance, depth, and solidity which retinal images are the basis of the visual
characterize our visual perception have world? The problem has been encountered
been the subject of the preceding chapters. before. The assumption that the visual
The visual world described in Chapter 2, stimulus was a static momentary image
however, has not yet been accounted for. was only priovisional and it should now be
Our experience of things is stable, up- abandoned for good. We need to consider
right, and unbounded although by rights it images as affected by exploratory eye-
should not be so since neither the retinal movements.
images nor the visual field has any of
those qualities. The Stable Visual World
The feature of the visual world which, As we asked in Chapter 2, why does the
as much as its depth, impels most of us to world not go shooting about as the ob-
the conviction that it is there and is not server shifts his fixation from one object
an illusion or a picture is its stability. to another? Helmholtz was aware of this
Samuel Johnson is said to have refuted problem, as he was of nearly all the other
Bishop Berkeley's doctrine that the world problems in vision, and formulated it as
was all in our head by the simple argument the question of how one recognizes sta-
of kicking a stone. He implied an ultimate tionary objects as such, in spite of the
trust in his muscular and tactual sense. shifting of their images over the retina
An equally good reason, however, for during eye movements. Helmholtz' answer
confidence that things are sensed as they was typical of his general theory: one sees
are is that they stay where they are. In objects as not changed because one learns
view of this stability can the instability to regard the retinal shift as merely the
of the retinal image, shifting over the "sensory expression of the ocular move-
retina with every movement of the eye, be ment" which actually corresponds to no
reconcile d with the assertion that the change of the objects (53, III, p.63). This
145
146
. / , ......., /
! .00
, .1m...........
.0
/ / .....!mlOm..........
'4%.
\\
jr \/
i Now is the ;time for all ;good men 1
0 1
/ \\
I
/ .
I 1
. / . . . . . /
, . .
1
.....__...
2
, 3
. ........... ..... _.... 4,
Now is thNottithe Chu iiinteltdothisktigeofflomeall good men

FOVEA -÷ 0
FIGURE 63. Successive Superpositions of Retinal Images
amounts to saying that we learn to see the suit would appear as shown in the second
change as not a change. line. The numbers in the first line
The problem can be formulated in differ- indicate fixation-pauses. The dotted el-
ent ways and it involves a set of related lipses represent the region of clear vision
problems all equally puzzling. It might immediately adjacent to the fovea. During
seem that a quick and easy way of solving a single second of reading, three fixations
it would be to reply that the question asked of the eye may occur, and the retina will
is meaningless. To speak of the world, or have been stimulated by a complex of three
of objects in general, as being displaced overlapping images. The fact to be noted
does not make sense, for what could it is that successive retinal excitations
possibly be displaced with reference to? during eye-movements do not fuse with one
This reply, however, is no solution. The another, as do successive exposures on
visual world is a response of the organ- a film or retinal after-images, but are in-
ism amazingly complex but still a tegrated in a very different way.
response and it could be displaced with The patches of color which make up the
reference to the tactual world, that is, to visual field continue to have a fixed direc-
the perceived surface felt by one's feet. tion-from-here when the corresponding
Anyone who pushes on the outer side of patches of focused light are displaced
his open eye (p. 31) can see the visual over the retina. The absence of blur during
world being displaced, and anyone who saccadic eye-movements might be explain-
has ever been dizzy knows that it can ed as due to a momentary central inhibition
move. Why does this kind of perception of visual experience during the interval
not occur during normal exploratory eye- (56; 121, Chapter 23), but this would not
movements? account for the absence of displacement.
Consider the diagram of Figure 63. There must exist some compensatory effect
If the series of displaced retinal images produced by the eye-movement itself.
which occur during the reading of the Whatever neural process initiates the jerk
printed phrase were to be thought of as of each eye from the old to the new fixation
persisting after each fixation the re- point must, at the same time, shift what
THE STABLE AND BOUNDLESS VISUAL WORLD 147
used to be called the "local sign" of every experimental observations. In terms of

anatomical retinal unit. This innervation visual experience, the theory states that
of the eye-muscles might be supposed, in when the eye actively rotates to the right
geometrical terms, to add a constant dis- the scene should appear to be displaced to
placement of opposite sign to the whole the left, but that it is not displaced be-
anatomical pattern of excitations or, what cause there is a compensatory shift to the
is the same thing, to transpose the axes right which cancels it. (Only a single eye
of reference which give it a right-left is referred to, but the argument applies
(and up-down) quality in experience. equally when both eyes are functioning.)
What the physiological correlate of local (1) If the eyeball is mechanically ro-
signs or of reference axes might be we do tated toward the right the scene should ap-
not know. But we do know that the eyes pear to move to the left and will do so,
are mobile relative to the head, the head according to the theory, because the com-
relative to the body, and the body relative pensating shift is absent. The reader may
to the ground, and we know that the pos- be able to verify this prediction for him-
ture of all these organic systems with self by pushing his eye with his finger.
respect to the ground is maintained by re- (2) If the eye actively rotates to the
flex mechanisms. A chain of reference is right after a clear negative after-image has
thus provided by which an absolute stand- been established in the center of its field,
ard of "straight ahead" might be estab- the after image itself should not appear to
lished. move at all because it is not displaced on
the retina; according to the theory, how-
The requirement of some such theory as ever, it will appear to be displaced to the
this has been recognized by psychologists right because the compensatory shift is
for a long time (53, III, p. 570). As here present with no retinal motion to be can-
formulated, it implies that a saccadic eye- celled. This result is a matter of common
movement involves both the neural activity observation.
producing the eye-rotation, which varies (3) If the eye is innervated to rotate
between opposites (right-left; up-down), to the right but actually does not move
and the neural activity produced by the because of a paralysed eye muscle, the
shift of the image over the retina, which scene should not appear to move on any
also varies between similar opposites. grounds of retinal stimulation but, ac-
The movement of the eye and the move- cording to the theory, it should move to
ment of the image are by necessity re- the right because the compensatory shift
ciprocal. A combination of two processes to the right is present, as in the last ex-
each reciprocal to another is obviously a periment. That it does just this is ap-
constant, and such a constant product parently well known to ophthalmologists
might then account for the non-displace- and the phenomenon was described by
ment of the visual scene. Helmholtz (53, III, p. 245). If the central
Some evidence for this theory may be neural mechanism is normal, a patient who
obtained by trying it against a number of tries to move his eye to the right in these
circumstances has an illusion that things world has the useful but unappreciated
move to the right. property of remaininghorizontal and
(4) One final observation may be given. vertical in visual appearance even when
If the eyeball is mechanically rotated to we lie on our side or tilt our head and
the right after a negative after-image has thereby turn our retinal images around
been produced in the field, the after-image their centers like a couple of radio dials.
should not appear to move on account of Although the contours and edges consider-
any compensatory effect. Only if one as- ed as color-patches within the visual field
sumes some such nonsense as that visual usually appear to tilt under these circum-
sensations are projected outward from the stances the visual world remains linked
eye like the beam of a magic lantern to the vertical and horizontal axes of
should it move at all. As nearly as I can gravity. The eyes, it is true, do roll in
observe, the after-image in this experiment their sockets a few degrees to compensate
is stationary. The scene on which it is for slight tilts of the head to right or
superposed, however, is displaced to the left but they do not counter-roll enough
left, as in the second observation. Helm- to put a twist in the optic nerve. Why then
holtz, as one might guess, observed this does the world not tilt when the image
phenomenon too, although he did not in- does?
terpret it, and his report is that the after- The answer seems to be that when a man
image does not move although the screen lies on his side or puts his head in an ab-
on which it is projected does (53, III, normal posture he gets two forms of sti-
p. 244). mulation which vary concomitantly, first,
The implication of all this is that the the rotation of the retinal image around
directional stability of the visual world its center, and second, the off-balanced
might be a product of activities which are pull of gravity both inside the inner ear
inverse to one another the activity of and on the bilateral musculature (42,
the motor centers for eye-movement and p. 303). The gravitational and the visual
the sensory centers for retinal motion. The stimuli vary in a reciprocal relation to
visual world then could possess a stand- one another; the product of these two
ard straight ahead direction-from-here and variations would be a correlate for the
a constant direction at all other points. upright visual world.
Kafka has discussed the problem in the Evidence for such a theory as this is
light of its history (67, p. 384 ff.) and sug- no more than indirect since the existing
gested an explanation essentially similar experiments on the problem have not been
to the one given above. based on it. Most of them have sought to
prove either that visual stimulation was a
The Upright Visual World. The visual primary and posture a secondary deter-
world is stable in another respect in addi- minant of upright vision or vice versa.
tion to direction. The world always ap- Although asking whether posture is prior
pears upright and, little as it may seem so, to vision or vision prior to posture is like
this is actually a very curious fact. The the juvenile puzzle of the hen and the
egg, a good many students of the problem Koffka's application was to the ordinary
have been guilty of it, including the writer case of covariant or reciprocal cues.
himself (42). Insofar as Koffka was demonstrating
Koffka, (67, p. 213 ff.) believed that the with this theory that anatomically fixed
upright character of the phenomenal world axes on the retina the nervous connec-
despite a tilted image was explained by a tions of a hypothetical pair of lines en-
tendency for the main lines and contours graved on the retina are no explanation
of the image (floor, walls, horizon) to be- of why we see things as vertical and
come the reference-axes on which an im- horizontal, he was surely correct. The
pression of tilt would have to depend. He theory of Cartesian coordinates on the
based the theory partly on a report by retina is a myth. But our visual sense of
Wertheimer (118) that if an observer keep- the vertical and horizontal, unlike our
ing his head upright looked through a tube visual sense of texture or contour, can-
at the surface of a large tilted mirror, the not be ascribed to the prevailing main
tilted scene gradually righted itself in per- lines of retinal stimulation alone. Pure
ception and, after a period of time, looked order in the image might explain a contour
normal. This observation has been ques- but not the orientation or direction of a
tioned, however, by the writer who re- contour. The visual vertical and hori-
peated but could not verify it. He pro- zontal (when apprehended correctly) have
duced an optical rotation of the visual reference to the direction of gravity. The
world by Wertheimer's mirror-and-tube direction of gravity is reliably indicated
arrangement and on another occasion by by postural stimulation, in ordinary life.
a pair of reversing prisms placed before Neither the uprightness of the visual.
the eyes, but the tilt did not disappear world nor its stability can be understood
(42). The observation, therefore, remains if we confine our attention to a shifting,
in doubt. Koffka's theory, in any event, swivelling retinal image. These features
went somewhat beyond this phenomenon to of space are inseparable from the feeling
assume that the principal lines of the of the ground under our feet and the feel-
visual field when the head and eyes were ing of standing up, of moving about, and
tilted but the physical environment was up- of looking. The tactual and kinesthetic
right immediately determined the vertical stimuli which arouse these feelings or-
and horizontal axes of the phenomenal dinarily co-vary with the visual stimuli
world. The main lines of organization are and the product is something which is
seen as the main directions of space. The neither visual nor postural. It is not that
implication is that vision, and vision we stand with our eyes or that we see
alone, accounts for the vertical and hori- with our muscles, but rather that we both
zontal appearance of things, and that stand in and see a steady world.
only the laws of sensory organization The writer's discussion of the problem,
need be invoked. It should be noted that written in collaboration with Mowrer, has
Wertheimer's phenomenon was a case of already been referred to. It made the er-
discrepant or conflicting cues, whereas ror of assuming and trying to prove that
postural cues were logically and geneti- visual world. When they are concomitant,
cally prior to visual cues in determining as they are in voluntary change of pos-
how we maintain postural equilibrium and ture in a physically normal environment,
perceive an upright visual world. A motor perception of the vertical remains correct
theory in opposition to Koffka's visual and equilibrium is maintained. When,
theory of the perceived vertical and hori- however, they are set in conflict with
zontal was proposed. The evidence con- one another by an ingenious experimenter,
sidered was taken from experiments on or in the "haunted-swing" of an amuse-
conflicting visual and postural cues, the ment-park, or during the act of banking
unrecognized assumption being that an airplane, the perception of the vertical
when two kinds of determinants seem to tends to be unstable and is likely to be
underlie the same kind of experience one objectively incorrect. In this situation,
of them must be the true basis for the ex- the organism is forced to search for re-
perience and the other must be secondary. liable cues to the direction of gravity, and
The props for either a motor theory or the perception is objective only to the ex-
a visual theory have recently been knocked tent that reliable cues are discovered and
out by the studies of Asch and Witkin used. The flier, for instance, usually has
(2, 3, 4) who repeated the observations to learn that the objectively reliable cue
made by Wertheimer and the writer with in banking an airplane is the visual hori-
more subjects and under more varied zon. He always has to learn that the ob-
conditions. Several of their experi- jectively reliable cues in night flying are
ments made use of a boxlike room open the instruments.
on one side, into which the observer The question of the reliability of the
looked, and which could be physically cues for space perception, including the
tilted while the observer himself stood cues for depth and distance, arises when
upon a level floor. He had to adjust a there are discrepancies among them. In
movable rod within the room to what he the preceding chapters there has been no
thought was the true vertical. Their ex- mention whatever of this question. It is
periments show that when the direction of a valid question, of course, and it might
the postural vertical and the direction of just as legitimately be asked about
the main lines in the visual field do not stimuli for space perception, as it is asked
coincide, the usual result is not a com- about cues or clues for space percep-
plete domination of either but a somewhat tion. Brunswik gives it great emphasis
unstable compromise between the two in his treatment of the objectivity of
directions. Whether one or the other set visual perception (15). The neglect of it
of discrepant cues tends to dictate the in the present book stems from an inten-
observer's perception of the vertical de- tion to concentrate on the theory of those
pends on the circumstances and on the spatial perceptions for which the deter-
observer. The implication is that postural minants are supplementary to one another,
and visual stimulation are both deter- not discrepant, and for which the stimulus
minants of the upright character of the conditions are optimal rather than im-
poverished or inadequate. fundamental than the problem of form as

something possessed by objects. Actual-
The Transposition of the Retinal Pat- ly, what is transposable on the retina is
tern. The theory of a reciprocal effect the whole panorama of the retinal image:
upon the excitations set up by a moving the steps of texture and shading, the
image on the retina still leaves another gradients of variation, and the inflections
paradox unsolved. A moving image as an of these gradients. As contours, sur-
event of physics and optics can be under- faces, depths, and objects they yield a
stood by analogy with the displacement stable visual world. As stimuli, therefore,
of a camera film relative to the image they must be equivalent when one set of
focused on it. But a moving image as retinal receptors gives way to another.
something which initiates vision in the The paradox goes deeper than the equi-
retinal receptors can be understood only valence of a square; it is the identity of
by such analogies as our moving electric the retinal image itself which comes into
sign (p. 56). The pattern of excitations question. What equivalence can there be
in the retina, the optic nerve, and the between two sets of nerve-cells which
brain implies a set of localized excitations bear a purely anatomical relation to one
which are anatomically displaced by an another?
eye-movement. The corresponding shift The answer suggested in Chapter 3
of the patchwork of colors in the visual was simply that there exists an ordinal
field relative to its margins also implies a pattern of excitations. Steps, gradients,
set of localized excitations. But then why and inflections of a gradient are mathema-
are the steps and gradients of excitation tical facts which are the same no matter
unaffected by the displacement? The dif- where they are located. In an array of
ficulty was first formulated with reference adjacent nerve-cells an increase in ex-
to geometrical forms, and the question citation from one cell to the next is a kind
was why a square could still be seen as a of serial order which can be transposed
square after a wholly different set of without causing a permutation or an in-
retinal elements had been excited. This version of the order. The set of numbers
is the problem of the transposability of ..345.. are similarly equivalent in order to
forms. It led first to the doctrine of form- the set of numbers ..789.. but not to 987
qualities or special sensations of form or 879.
and, when this was recognized as naming Can one suppose that the organism is
but not solving the problem, it provided a capable of reacting specifically to an or-
point of departure for the whole of Gestalt der of excitations? Can a perceptual im-
theory (Chapter 2). The solution offered pression correspond to what seems to be
by Gestalt psychology was that a process a mathematical abstraction? We do not
of configuring or organizing the excitations know, but the suggestion may be. more
could be assumed in the central nervous reasonable than might at first appear.
system. What, exactly, is a stimulus? Physiolo-
Evidently, however, the problem is more gists and psychologists often assume that
152
it is a kind of physical energy acting on a

single receptor. We shall name this the
one-stimulus-to-one-receptor theory and CD
we note that it gets into trouble when a 4--
sensory surface composed of adjacent

receptors is considered. Moreover the
physiology of single receptors, or what
is known of it, suggests that the effective
stimulus is a change of energy acting on
the receptor. Some psychologists are led
to the assumption that the organism
responds differentially to variables of
stimulation but not, strictly speaking, I I
to isolated stimuli. In psychophysical

I I
I I
experiments, what the observer reacts to

I
is the relation between two things or \

events and almost the whole of our knowl- I
I I
edge is founded on judgments of "greater",

JO
I I
I I
"less", or "equal" (102). Change, varia-

tion, and relation are no less abstractions
than is adjacent order. Perhaps the con- ,
cept of ordinal stimulation only makes

i
( ) ,
,
C_)
explicit what has often been assumed.
We must not forget, of course, that there
is a fixed anatomy of the retinal receptors
Li i \ ,
CD
,
and that the appearance of the visual field

Li
I
.
bears witness to it. A transposable order I
is not, then, an abstract, timeless form of -4-
the sort that Plato believed to underlie

the world. A transposable order would be
meaningless unless it were embodied in
an order of adjacent anatomical elements. CID
The Transformation of the Retinal Pat-
tern. The retinal image, as we have seen
in the first part of this chapter, may under- t-
go not only a transposition but also a de-
FIGURE 64. Successive Transfor-

CD
mations of the Image of a Square
of Pavement while a Person
is Walking along a Street
formation. As a whole and in every part, destroy the pattern as one might at first
it undergoes deformation whenever the head suppose. A transformation is a regular
moves, but the stability of the visual and lawful event which leaves certain
world remains unchanged. The only effect properties of the pattern invariant. The
on perception is an impression of subjec- study of the invariant properties of geo-
tive locomotion. Mathematically, this metrical forms which have undergone trans-
change of the image is analogous to formations is known as topology and the
a projective transformation. This is study of projective transformations is
a kind of transformation which, as it known as projective geometry. The writer
were, stretches but does not tear the form. cannot claim to be an expert in either of
Point-to-point and line-to-line corres- these branches of mathematics. The
pondence is preserved, but the shape is general application of such principles to
altered. Transformations are usually re- the deformation of the retinal image during
presented on a plane, however, whereas locomotion by the observer has not been
the retinal image is a projection on a attempted, so far as I can discover, by any
curved surface. As a matter of fact, the psychologist or mathematician. At the
actual retinal image on a curved surface crudest level only, then, it seems clear
is related to the hypothetical image on a that not every pattern can be regularly
picture-plane only by such a non-rigid transformed into any other. A doughnut
transformation (see Figure 28, Chapter cannot be transformed into a cube by any
6). The geometry of transformations is continuous process. But a transformation
therefore of considerable importance for can be applied to a given pattern without
vision, and it is conceivable that the clue affecting certain of its general properties.
to the whole problem of pattern-perception The skull of a chimpanzee can be trans-
might be found here. If we add to the formed into the skull of a man, as D'Arcy
classical problem of the transposability Thompson first suggested, by afairly
of the retinal image the requirement that simple geometrical operation (105,p. 1085).
it must be transformable as well, the Moreover a series of transformations
problem may emerge in a new light.). can be endlessly and gradually applied to
The transformation of a given pattern, a pattern without affecting its invariant
mathematically defined, does not simply properties. The retinal image of a moving
'According to Courant and Robbins (25), any transformations provides a solution for the
mapping of one figure on another by either a puzzle of perceptual constancy. But he assumes
central projection or a parallel projection (or a that the process of perception must necessarily
succession of such projections) is a projective, involve a search for constancy, an objectifica-
transformation. In such transformations straight tion of sense data, or a discovery of the in-
lines are invariant, intersections are invariant, variant properties of shapes which have under-
and the order of points is invariant. Lengths gone transformation (21). What he suggests is
and ratios of lengths are altered, but the ratio that the mind transforms the retinal images,
of two ratios of length (the cross-ratio) is in- whereas the suggestion above is that transfor-
variant. mations of retinal images are equivalent as
Cassirer has suggested that the geometry of stimuli.
154
FIGURE 65. Human and Chimpanzee Skulls
(Redrawn from D'Arcy Thompson,Growth and Form (Macmillan, 1942)
observer would be an example of this locomotion, as we have noted, but during

principle. Perhaps the clue we are seek- passive locomotion little or no bodily
ing lies in the invariant properties of such stimulation exists, and hence we probably
a continually changing retinal image. Only cannot fall back upon it for the whole of
these properties would be capable of pro- the explanation. Certain features of the
viding the stimulus basis for a stable and retinal image are preserved during a
unchanging visual world. A man who walks locomotor transformation order, con-
about in his back yard never has the tinuity, points to points, and straight lines
same retinal image of it twice unless he to straight lines while certain other
comes back to the same spot, puts his head features are not preserved angles, the
at exactly the same point, and fixates in congruence of shapes, and the metric
the identical direction. Nevertheless he properties of lines. The features that are
perceives the same environment throughout preserved may be the mediators of a
his wanderings. He also, of course, per- stable visual world and the features not
ceives his wanderings. If the retinal preserved the mediators of the visual im-
images have constant properties and also pression of motion.
undergo a continuous transformation per-
haps we can account for both his visual Evidently the retinal image needs to be
perception of the environment and his considered as a changing event in time,
visual perception of locomotion in it. and the analogy with a static picture is
The suggestion is that we need to thoroughly misleading. The order into
understand the geometry of the transforma- which the image can be analysed ought to
tions of the retinal image in order to ex- include not only the adjacent order which
plain why its successive changes can all has mainly concerned us but also the
be equivalent for perception. The changes successive order which it equally pre-
co-vary with muscular action during active supposes.
The Problem of the Unbounded Visual The Function of Saccadic Eye-Move-

World ments. The movements of the eyes have
The problem of why the visual world is been studied for many years. In the half
stable is actually related to another prob- century since Dodge invented a photo-
lemwhy it is unbounded. Ordinary visual graphic method for recording them objec-
perception is not delimited by an oval-shaped tively and described their fundamental
boundary, nor does it have a clear center types (26), a great deal has been learned.
and a vague fringe. These are the charac- The shifts of fixation when we look at pic-
teristics of that unusual kind of visual ex- tures, for example, have recently been
perience, the visual field, which we get carefully studied by Buswell (17). A
when we fixate a point and take note of typical record of how the eyes behave is
the experience, concentrating on how it shown in Figure 66. Note that the fixation
feels to see. As we noted in Chapter 4, point moves all over the picture, but not
the center, the fringe, and the boundary in an orderly fashion. There are frequent
are reflections of the optic anatomy: the zigzags and irregular jumps. No simple or
fact of the fovea, the thinning out of clear relationship appears between the
sensory cells, and the margins of the order of fixations and the order of elements
retina. which compose the picture.
Prolonged fixation interrupts the nor- Some theorists have believed not only
mal course of vision by inhibiting the that eye-movements could account for the
exploratory movements of the eyes. perception of locations and distances in
As has been emphasized, the eyes space but also that they might explain
are extremely mobile organs. In the ac- the perception of patterns and two-dimen-
tivities of everyday life the center of clear sional forms. The eyes might be sup-
vision will shift as often as a hundred posed to trace the outlines of things and
times a minute, and during reading or while thereby provide cues to their shapes.
driving a car the rate of fixations will ex- But the actual records of eye-movements
ceed this figure (121, Chapter 28). Can have never supported the theory. They
we find an explanation in the facts of tend to be more or less like the record
ocular movement for the absence of the illustrated. A related but more plausible
above characteristics in the visual world theory is that the composition of a paint-
its lack of boundaries, its more nearly ing is something which enforces a parti-
uniform clarity and its possession of what cular sequence of fixations in the onlooker,
we might call a panoramic character? The the eyes being drawn from one point to
depth and distance of the visual world can another by the spatial order of lines and
be accounted for. Its stability or un- masses in the painting. It is widely be-
changing direction-from-here one can at lieved by artists and art critics. But the
least struggle to explain. But why is it theory cannot be literally true in the face
unbounded when the stimuli consist of of the illustration given, and if there are
fragmentary images? obscure relations between pictorial corn-
5 5, 5
' a
01
'00''r \ L\
,;01.:44,410,
t: a
v,a ,a, a' .4 af,'.,141',.. "I
1
I
,a45k 'at 'al
+45'5, alarv,
V;:'
''O,11Yhth, '4, 'a ta11,155A1',14,"1,5;
41'ea5,,,
-
a;55'
14
mf 11111kr.,
PP 1
4.44444.44AL,
FIGURE 66. Successive Fixations of the Eyes in Looking at a Picture
(From How People Look at Pictures, by G. T. Buswell (Chic-

ago University Press, 1935). By permission of publishers.)
1`.1
S4
55 41 'Caartpalle , 40,114 a'vak
i "1, '',; ." s aa' { 5 'r 5' .51. 5 A 'a

55, ,5,5%)
'401 4,§, 44v A 'h 5

44,4,4% r a ' 1 r 5, 5 155 SSS 554%4'44, *4
5,5, 4, 06 1' " 555 IA , 55 .5 55

5
54'14'1'1,4 '44
., ,, 4-; ,,4'0,, ,,,4, ..4, ,4,54,1,,,4 ,1 04i,
4AV1'i4,k Ir, , ,VA I
1: ' 5 a, * 'itiaa, 5A4 ' 405,i':.4,k'5,,,,i1.5,,,t,,N0a 45,,',a1AFt545" 4,5,51 ,,,,,, 't" ., , ,, ,, , , 5-' MT, S 5 5
'
,tt a
C 4.:''' , , ,, + , : 5.,1 a .
,
114 ' v
4
,
' ^ ,15 ' a 'a V 5 . ' 1
0, 5 ''5,," , ,s' ',',',':, ,, ,4

''' '' 5,5 _54,5,
5S
'' '' / 5 50 ' t , '5 5 .5 a , , 55 5 55
''"t 5 s ''
Va
5'' 5 a'
14,41' ..,,..s.eass0 a. 41' .5

v '
5 , 5,5 , V, 55 5 5
55j5sH55s5 " 5'1
5555 5 4
'155 5 ' . 55 '44-
5 5 5 5
, S 555
s "5 555 5 ,5 '55 ,5-5,1 av1t; 5 'sea 5,
5 '5 , 5
5
, A'on, 1'5
Vaa
55 , aa,
1,
.6 4i 5 .11.
55555S
" ,v5a, L'
5n ,
rt V1,5 4,4 1aVim4,14 :,1,4454. nas'

, 555 S55 55 ,1a a 5 5, 4 '1,4V
Ppr
,15e 50 a 55 ,
5
IA* '.#4,4;:a4
,r, ,',.;61plef,,. a ,.a.a.Alta,.444/4atak a 54.
V 5'5; 15 5:05 5 :5 11,1,. 55
5S , 45, 04;1. Yra 4:
,
y
t
;154 .4 4515 1, 555S ,5v, 5 . S
sa!,:a55^05,54044155,15,5,,4 ';2'0
,,, ..,,,,, ,, , o,
, tt ,at , 55
4.,,554 \..05,..4' ,,r)e.5 '',',41.,"7, sV:'lA145
1
tIV, ^
,,,,,, ,,5,1 5,#:4,.., .?,, ,* 5A, a o 0,, ,A0; I \ , 1'g';
, SS 41%1, . ,,1, ' 1.1 , 5' ..., , -'5 , , , ,1 is,,,i, A ,,, s' 4 v4 4a, 964,,4aas 1.1
"5 5-H, ''11.1144'q\r"441;
, 4; 1,k *4'5 s ,A4,, v,i, 144 , , av .a, 'k4 ,,aa.a
al:';044''', 4101 ; ''
, ,,,z,,, , 44., , 111
atv, al, , , ,ts, av ,.'', :T 5.5.5'1,45 (0,14N t ,A at Si444 V ,555,,a) as', 4,v lai 4:',i5 a.',N,.'1'").44,,,,p0)40444s 1,404'4140
,,,,,,,, , a 0,0,,s , , 45,$, s,o. s %,.104,4, *51, ''', 5" 445,a4,4s.1144, sa ,s' a'V,v
, )Y ,,,!0,AAiftV 1,01 'Or0

414'1
1 '
. 04,10,1)Orn, ......,, ,5
ss l5, 40:4
5 ;444,7,4v karì.501sa 1 N Ara0i'av,
a'a1A4- ,sakisaa, 's
, 45' A 0,4"
44:4,-4 ,,141'11' 44 4.55,t5,I AA,
'IA
51 a 454 '0, 1
!tkio*'itklo: 0,0 '4,4r

a",
ife;W4 t15,',5
4s 1.kor,,Ii1,44555's,44,.4,444'61a5
; 1',4',40,040sv:* '5; " 5 5,45
%')1sCasaa664*,7,,4,,,4 to1s0a,,a,44.,4 .4`40i,s

in
4'611" "Y
Ltt
va,44,arithettoW+4,P,Ait'A. 555A,5454,5,01,
.114,A1c.550
i 4",
position and fixation sequence they have rotates. All we can do is to let our as-
not yet been discovered. The evidence tonishing ocular reflexes take care of the
suggests, for vision in general, that the intricate task of looking.
sequence of fixations is more or less ran- One slight but important exception to
dom, and that each single fixation tends to Helmholtz must be taken. We must not
fall on an object of attention (rather than forget the total retinal image. What gets
the background) and on a point of interest centered on the fovea are successive small
within an object. regions of an extended image covering
The function of saccadic eye-move- 180° which, as an image of optics, is
ments is probably not that of perceiving or equally distinct everywhere. These
appreciating any kind of order in the reti- regions correspond to the so-called ob-
nal image. It is more probably to enable jects of attention or points of interest.
us to register the environment in two ways It is therefore not true to say that what
which, in the use of a camera, are mutual- happens on the retina is a succession of
ly exclusive. Eye movements enable us to fragmentary images as we did in the be-
see the environment over a very wide ginning. There occurs a succession of
angle; fixation gives us narrow angle overlapping images, 180° wide, only the
vision for fine detail. The eyes function centers of which are registered by the ner-
so as to obtain the advantages of both a vous system in fine detail.
very wide-angle camera lens and a tele- The Problem. We are forced to conclude
scopic or narrow-angle lens. \X ith the that the visual world cannot be perceived
same eyes, we can either look all around all at once. The process of perception
or look at a single object. cannot rest upon the image of a single
Helmholtz understood some such func- fixation such as yields a momentary visual
tion as this. The intent of vision, he said, field. To see more than this takes time,
is to see as distinctly as possible, with and requires a succession of images.
both eyes, various objects or parts of an The product of these successive impres-
object in succession (53, III, p. 56). The sions, however, is such that, paradoxical-
sole purpose of the mobile eyes is to per- ly, all awareness of the succession has
mit a kind of light-absorptive pointing. been lost. Unquestionably the panoramic
Any kind of ocular movement which does visual world depends on a temporal series
not produce the distinct imaging and of excitations and just as unquestionably
centering on the retina of successive the succession of the excitations is not
parts of the environment in each eye can- represented in the final experience.
not be performed. We cannot move our Evidently, the abstractions which we
eyes evenly and slowly across the en- call space and time are not as distinct as
vironment. We cannot move one eye up- they have been assumed to be, for space
ward and the other downward. We cannot cannot be apprehended except in time.
converge the eyes without accommodating These abstract ideas, however, arise out
the lenses. We cannot hold the eyes still of concrete experiences, and it is these
with reference to the head when the head experiences that concern us. At the
stimulus level we have a succession of in order.

overlapping images. At the level of ex-
perience we have a panoramic world, all The Integration of Successive Images.
parts of which are concurrent. In ordinary Successive excitations of the retina must
visual perception there is no sense of be integrated by memory. How else could
either the sequence of fixations or the our present perception of the extended
time-lapse between them. This is to be world arise except through memory of just-
contrasted with auditory perception, where past glimpses? The kind of memory re-
the sequence of experiences is identical quired to explain perception is not, of
with the sequence of stimuli. A heard course, that commonly understood by the
sound begins, proceeds, and ends, but a term. It is often called primary or im-
visual object does not, although its patch mediate memory. It is the kind of memory
of color enters, remains, and leaves the which makes possible the apprehending of
foveal region. At the circus, for example, a melody. Its similarity to perception has
you may watch the tightrope walker, then been emphasized by calling a melody a
look at the performing seals, pause to ob- temporal Gestalt, and there are some
serve a clown, and return to the tightrope striking parallels between such experi-
walker. Although you have had a suc- ences as visual patterns and auditory
cession of impressions the events are rhythms (67, ch. 10). Primary memory is
perceived as coexisting. The stream of the kind which enables us to hear a
your consciousness is more like the series of seven or eight spoken digits
multiple stream of events than it is like and repeat them immediately as if we were
the stream of fixations. reading them from a page. Beyond seven
The successive retinal images, it must or eight, however, we begin to have trouble,
be remembered, are not distinct entities and the limit for a given individual is
like the piCtures of a comic strip. They known as the memory span. Similarly, a
overlap and, to that extent, are trans- certain number of strokes of a clock can
positions of the just-preceding images. be apprehended without counting them;
Instead of saying that the visual world is five o'clock does not have to be counted
based on a succession of images, there- whereas twelve o'clock does (10, ch. 5).
fore, it is possible to say that it is based All the experimental evidence implies
on a continuous but changing image. The that successive stimuli somehow endure
retinal image may be defined as a pattern and are integrated to a single experience.
or as a process. The problem, according- Introspectively observed, the conscious
ly, may be formulated in two ways. First, present does not seem to be a point where
how are successive patterns on the retina the future meets the past but a consider-
integrated to form an unbounded visual able range of events. Although the concept
world? Second, can a compound of ad- of primary memory has been derived from
jacent and successive order be defined the study of auditory perception, it must
which would provide a stimulus basis for apply with even greater force to visual
the visual world? We shall take them up perception, where successive integration
is so complete that the observer can be a few seconds, is the unit of the motion
wholly unconscious of his fixations. picture camera, and a spliced series of
We could assume that each fixation is shots comprises a scene, which is the
followed by a primary memory image which unit for telling a story. It is no accident
fades very slowly, each segment of each that a series of shots is called both a
image being fitted to every other by the scene and a sequence in the terminology
compensatory displacement mechanism al- of films. A good film editor can make us
ready described. The result would be a see a scene by means of a sequence. A
panoramic world which, at any given concurrent situation combining, for in-
instant, will have faded least in memory stance, persons and places can be created
at the point of regard and most at the by successive camera views if their se-
point directly behind the head (depending quence is consistent with that of natural
on how recently one has looked around). apprehension rather than that of some ar-
The fading would not be a decrease of bitrary logic. Just as the visual world
clarity like that observed from the center must be scanned in order to be seen, so
to the periphery of the visual field but the situation must be represented in some
only a decrease in reportability or that little understood natural order. Each
quality of experience which goes with camera view persists in memory, but that
recency, and hence the visual world can be is not sufficient for a clear understanding
said to be everywhere equally clear. of the situation; the fixations of the camera
What the physiological basis of memory on the points of interest must conform to
might be, we do not know. Each momen- certain rules.
tary excitation might be assumed to leave A traditional film sequence consisting
some after-effect or trace in the brain, of a long shot, a medium shot, and a
but the working out of a complete theory of close-up reflects merely the tendency of
successive traces is faced with many dif- an onlooker to look around in a new situa-
ficulties (67, ch. 10-13). Even a theory tion before he narrows his attention. The
comprehending the primary memory-image, instantaneous cut from one shot to another
the eidetic image, and the image of im- is analogous to a saccadic eye-movement.
agination is lacking, and the relation of Scenes, on the other hand, are separated
these to the after-images and other after- by fades, dissolves, or other visual de-
effects of prolonged fixation is unknown. vices for representing a time-lapse. Be-
tween the heroine bound to the railway
Sequence and Scene. It is interesting to tracks and the hero galloping across
compare the integration of the visual world country to the rescue occurs a cut which
with the way in which a series of motion- makes the time of the two events simul-
picture shots builds up a scene. The first taneous. The alternating order of shots
kind of integration is an unconscious in a chase sequence is easy to under-
process. The second kind is one whose stand; where the difficulty comes is in
rules have puzzled even the greatest of arranging the sequence for a scene of
film artists. A shot, which may last only complex human interaction.
Our hypothesis implies that one guide have a succession of retinal images, the
to motion-picture cutting would be to de- image of a moving observer undergoes
termine the sequence of fixations which an continuous transformation. Must we
onlooker would unconsciously perform if analyse the latter into a succession of
he were taking in the scene as it would distinct patterns and assign a specific
really occur. What would he look at first, memory-image to each? How many memory
what next, what would he then examine? images shall we assume one for each
The formula would not be literal, since ten seconds of locomotion? One for each
the camera cannot look in the way that second? One for each thousandth of a
the eyes do, and the limitations of the second? It would be simpler, if this
camera must be allowed for. Such a for- were possible, to define the retinal image
mula might serve as well, however, as the as a continuous process in time and leave
obscure intuitions and traditional prac- the memory images entirely out of account.
tices which govern most film-making, and Consider the perception of simple
it has the virtue of being subject to ex- motion the movement of an object across
perimental study.2 a visual field, for instance. If we were to
The problem of how to produce a scene be consistent, the integration-by-memory
from a sequence is not confined to the theory would have to be applied to this
motion pictures. The novelist, and in fact experience. Only if each retinal element
all artists but graphic ones are confronted successively excited leaves a trace in
with it. Abstractly, the question is how immediate memory could the past excita-
to produce a series of impressions which, tions be held together to yield the impres-
for the reader, the hearer, or the observer, sion of motion. We would have to con-
creates an objective world flowing in ob- clude that motion is never perceived,
jective time. A successive order of im- strictly speaking, but only remembered.
pressions must be made to yield an ad- Contour, color, surface, depth,
form,
jacent order of things. and distance, being instantaneous, are
The Compound Order of the Changing perceived, but not motion. This conclusion
Image. The foregoing explanation for comes near to reducing the memory theory
the panoramic character of the visual to an absurdity.
world is satisfactory for a stationary ob- A simpler assumption would be that
server but the case of a moving observer motion corresponds to a variable of retinal
puts a bad strain upon it. Although the stimulation. The variable is a correlation
stationary observer may be considered to of the adjacent and the successive order
of excited elements. This assumption
raises no questions as to whether the im-
2
Dtscussions of film-montage, such as that pression of motion is based on memory,
of Eisenstein (31), have been concerned with inference, or the interpretation of eye-
the dramatic effects of successive shots rather
than the effect of a coherent world in time. muscle sensations. We need not assume
The former, however, surely depends on the any short-circuiting in the brain such as
latter, and the simpler problem should be
studied first. Wertheimer proposed to explain strobos-
copic movement and the phi-phenomenon when it is focused as an image. An eye

(118), and we need not decide, in an out- can explore the flux of light at a given
moded terminology, whether motion is a position like a blind man feeling an ob-
sensation or a perception. We only sup- ject on different sides in succession, and
pose that it has a stimulus-correlate. the panoramic image is just as immovable
This assumption is consistent with the as a fixed object. The complex order of
evidence that motion is a simple quality steps, contours, and gradients of this
of experience . and it has been implicit potential 360° image is unchanging, and
throughout the last two chapters. If it is the momentary images merely sample it.
valid for the movement of a patch of color If now we suppose that the eye moves
across the retina, however, why should it from one point to another in the environ-
not also be valid for the continuous de- ment (as its possessor goes about his
formation of the retinal image? If it is, business) the image becomes a continuous
we are committed to the position that suc- serial transformation which, as a series,
cession is just as much a property of is unique to the path travelled. The series
visual stimulation as distribution. More- has no real beginning and no end during
over, if successive order participates in waking hours; any momentary cross section
and determines visual experience over a is specific to the momentary position and,
short interval of time, why should it not over a long period of time, the serial image
do so for a long interval of time? will have sampled the light flux at a great
Considerations such as these point to a many positions. Assume next that the
general theory of space perception which panoramic series and the locomotor series
has interesting implications. Let us first are combined, as they must be if an ob-
define the physical environment in which a server both scans his environment and
man lives as the one in which he gets moves about. The combination yields a
about. Depending on the criterion of range of images in two dimensions which
mobility selected, this might be his house, corresponds to the whole of a three-
his neighborhood, or his city. At any dimensional environment, independent of
given physical point in such an environ- any given point of view. It will not only
ment there is one and only one ocular be stable, panoramic, and unbounded, but
image which a standard human eye will it will approximate the range of images
produce when it is pointed in a given possessed by another man who lives in
direction. This image is unique. (Since that environment in the degree to which
the principle holds for either eye we need they have been in the same places. It
not refer to the second image.) If the eye will be something very much like objective
rotates at that point in its peculiar sac- space.
cadic fashion the images are individually An image of this sort is extended in two
unique, and the 360° panorama of images dimensions with respect to distribution
is a unique collection. Note that this and is extended in a third dimension with
fact is not psychological but physical; respect to sequence or time. It might be
the flux of light is unique at that point called a train of momentary images, but we
must remember that the panoramic series although it is abstract, and if it is not a
(related by overlap) and the locomotor pure illusion there must be a basis for
series (related by continuous transforma- apprehending it. To the abstraction on the
tion) must be combined in the same train. side of experience and behavior there must
This image-in-time is a conception so correspond an equivalent abstraction on
abstract and so strange that to call it a the side of physical stimulation. Tracing
visual stimulus may strike the reader as the latter may be difficult but the attempt,
absurd. A boundless, concurrent, and at least , does not involve us in an ab-
public world is not an absurdity, however, surdity.
The Constant Sizes and Shapes of Things
Why Do Things Look as they Do? . The

Constancy of Perceived Objects with Respect
to Color .... The Constancy of Perceived Ob-
jects with Respect to Shape ....
The Perception
of Foreshortened Surfaces . The Constancy
of Perceived Objects with Respect to Size .
How is the Distance of an Object Seen? . 0
The Perception of Distance and of Scale . .

The Rigidity of Visual Dimensions Does....
Size Constancy Break Down at Great Distances?
.... Conclusion: The Objectivity of Experience
For a long time psychologists have been that this explanation is not the true one.
puzzled about what has been called the Birds, for example, can discriminate cor-
approximate constancy of visual things. rectly between a large and a small object
Objects look much the same size whatever when the larger object is so much farther
their distance from the observer, and the away that its image is smaller. Does this
same shape at different angles of regard, or imply the memorizing by the bird of the
from different points of view. It is true, of true sizes of all the objects it sees?
course, that when you attend strictly to Animals in general do not behave as if
the appearance of objects the rule does what they saw was at first a flat visual
not hold, but then, we have argued, you are field and subsequently a three dimensional
having a different kind of visual experi- visual world. The behavior of animals and
ence. Will a psychophysical theory of birds suggests that they react differently
perception throw any new light on this from the outset to objects at different dis-
problem? The traditional explanation for tances. We may suspect that the so-called
this constancy of objects in ordinary un- problem of constancy is actually only one
critical perception has been that we cor- aspect of our larger problem the problem
rect our sensations of the size and shape of the perception of the visual world with
of things by remembering their true size all of its objective characteristics. The
and shape. Even apart from the objection aim of this chapter is ultimately to show
to original sensations there are indications that the question of why things retain their
163
sizes and shapes under different circum- solidity, and any theory of space percep-
stances is a false question. Only if it is tion must at least recognize this fact.
believed that perceptions begin as patches Of these five general properties, which
of color in a visual field does the question can we now account for with a stimulus-
arise. theory of perception? We may begin at the
end of the list and work backward. The
Why do Things Look as they Do? meaning of things, surely, cannot be ex-
To what extent can we now account for plained solely by their optical stimuli.
the perception of a phenomenal world Nearly all, if not all, meanings are learned
which is adequate for behavior? This is by experience and therefore depend upon
the question which we set out to answer memory. So much do they vary from one
in Chapter 2. The visual world, it will be individual to the next and from one culture
remembered, differs from the visual field to another that we are accustomed to say
in a number of ways. First, it has depth that different people do not even perceive
or distance, and it includes the experience the same world. We usually realize, how-
of solid objects which lie behind one ever, that such a statement is an exaggera-
another. Second, it is Euclidean in the tion. The world we refer to is the world of
sense that neither the objects nor the values and meanings rather than the world
spaces between them appear to change of space and shapes. The purest examples
their dimensions in perception when the of such meanings are incorporated in those
observer moves about. This is a general visual objects which are used as symbols
way of saying that they tend to remain printed letters and words. Although a
constant. Third, it is stable and upright; Chinese character will be seen as the
things as seen have constant directions- same shape, at least approximately, by all
from-here when the observer moves his men, it has meaning only to someone who
eyes and the pefceived ground remains has learned it. We shall return to this
horizontal when the observer tilts his head. problem in Chapter 11. Before we can in-
Fourth, it is unbounded; our experience of vestigate why things have meaning, how-
the world does not have any visible mar- ever, we must know why they are seen as
gins or limits such as the visual field or material things. We assume, for the
a picture has. Finally, it has a charac- present, that the solidity and separateness
teristic to which we have scarcely referred of things must develop in the vision of the
but which, in a way, is the most important infant before meanings can begin to be
of all: it is composed of phenomenal things attached to them. It seems probable that
which have meaning. Even if the visual only as the child can identify things by
world is primarily an array of spaces, sur- shape, size, and color does he learn their
faces, and contours, it is secondarily an significance for his needs or their use as
array of familiar objects, persons, and conventional signs.
symbols. We have been trying to account The unbounded or panoramic appearance
for the perception of the material world, of the material world, according to the
but objects have significance as well as suggestion made in the last chapter, is
THE CONSTANT SIZES AND SHAPES OF THINGS 165
one which might have a basis in retinal complex stimulation which would normally
stimulation. This possibility, however, be in correspondence with an unchanged
depends on the assertion that time is a scene.
variable of the stimulus or , more accurate- The distance,. depth, and solidity of the
ly, that sequence and arrangement are co- visual world have been considered in de-
ordinate dimensions of stimulus variation. tail. They tan be explained by stimulation
This is a plausible but abstruse assertion, if the world is conceived as an array of
stimulating to the imagination but not im- surfaces. Continuous distance is in cor-
mediately productive of clear-cut experi- respondence with certain gradients in the
ments to prove it. The alternative hypo- retinal image of a surface or the combined
thesis is to rely on the more familiar con- images of two eyes. Depth at the con-
ceptl of memory. The explanation would tour of an object which stands out from
then be that successive fixations or its background is in correspondence with
glimpses of the environment are integrated an abrupt step in these gradients. Solidi-
in primary memory to yield a panoramic ty, or depth-shape, is in correspondence
visual world. Stimulation, conceived as with the slopes of these gradients and
momentary, fails to account for the pano- their variations in slope. If these types
ramic world. of psychophysical correspondence are up-
We have seen that the stable and up- held in future experiments, the gradient
right character of the world can be suc- theory can be said to explain the tri-
cessfully explained by stimulation if we dimensional properties of visual percep-
take note of the body as well as the eyes. tion.
One may assume concomitant stimuli There remains one more item on our
which are reciprocal to one another. The list, the second. The visual world might
shift of the retinal image and the movement be described as rigid. A given object
of the eye to its new fixation are then in tends to remain constant in size as the
opposite correspondence. Likewise the observer moves toward or away from it,
tilting of the retinal image and the devia- and constant in shape as the observer
tion of the body or head from its normal moves around it. Perceptual objects, in-
posture are correlated but opposite. In stead of being deformed as the color-
both cases, the visual and postural patches of the visual field are when the
stimuli would yield a joint variable of observer changes his viewing position,
remain approximately the same, dimension
I
The fact that physicists have found it use- for dimension. This constancy of size
ful for certain problems in astronomy and in and shape also appears to hold true for
the structure of matter to assume that time is the ground or the floor, and for any seg-
a dimension of space the fourth dimension
is suggestive for psychologists but is apt to ment or part of the background. We shall
lead into high-sounding guesses. There may find some evidence that it also holds for
be a significant parallel between problems of
relativity physics and problems of visual sti- the distances between objects the
mulation, but until the latter have received a shapes of the intervening spaces even
mathematical formulation the tracing of the
parallel will be speculative. though the corresponding color-patches in
the visual field be shrunken, expanded, or light energy is perhaps the most basic fact
transformed. This tendency for visual of the science of vision, the sensory
dimensions to remain constant is a fact of colors can be seen only under the artificial-
perception. It is also true, of course, ly controlled conditions of the laboratory
that measured dimensions remain constant, or by the practiced eye of the painter.
but this is a fact of geometry and engineer- Such are the hues seen in a spectroscope
ing, not of perception. Perceptual es- or the color transmitted through a pane of
timates are less accurate than physical ground glass in a dark room: they are.dis-
measurements, but both are facts. The embodied colors floating in a visual field
axioms of Euclidean geometry, we shall rather than the colors of objective sur-
propose, are abstractions from both of faces in a visual world. They look filmy
these facts. When it is asserted that the and insubstantial and appear at an in-*
world we tend to perceive is a three- definite distance, in contrast with the
dimensional Euclidean space, the asser- colors of objects in daylight illumination
tion really means that the dimensions of which appear to be localized on and to be
things and of their interspaces tend to re- part of the surface of the object in ques-
main constant in perception. tion. The latter are known as surface-
The constancy of visual objects with colors, whereas the former have been
respect to size and shape needs to be variously called film-colors, expanse-
considered in detail, and it will be the colors, reduced colors, and the like (61).
main subject of this chapter. By implica- Only when color is thus disembodied or
tion, however, the problem is not limited separated from a localized surface does
to objects in the literal sense but ap- its brightness correspond with the in-
plies to the whole of a visual scene, to tensity of the light on the retina.
the ground on which objects normally rest,
and even to the dimensions of abstract The colors of objects in the ordinary
space. If our approach is correct, the sunlit visual world are not the same as
problem is a corrolary of a much more the colors of the patchwork in the cor-
general question: How do we perceive a responding visual field. The untrained ob-
world which is consistent with our actual server cannot see these differences, but
behavior the visual world of ordinary a landscape painter can, for he has had to
experience? learn that the disembodied color of an ob-
ject is the color which must be repro-
The Constancy of Perceived Objects with duced on his canvas. Just as a table top
Respect to Color cannot be represented as a square, so a
As a preliminary to the main problem, we white surface in shadow cannot be re-
might consider briefly the nature of color presented as white. In pictorial vision
perception. Although- colors are tradition- the former is a trapezoid and the latter
ally supposed to be sensations, and al- is gray. The surface-colors of everyday
though the correspondence of hue, bright- perception remain fairly constant despite
ness and saturation to the variables of changes in the illumination of different
1, ov,
14'007 ,
41'
rir u.
WA" I
opt
'Itf
.7"
,11), '
r, , Arj '4 Gir
"° ).{ 110'
,""j,,l'r,
I .1 N
FIGURE 67. Brightness Constancy
Fritz Ilenle, from Alonkmeyer Camera Guild
dry
, , k
;1;P:Vir,"-
h, 46*,
'h
) ,
14'4C' tio''
parts of the world and despite changes in impression of depth and therefore cannot
the illumination of the whole world be- at the same time yield an impression of a
tween sunrise and sunset. It is as if the difference in color between the adjacent
colors of objects in perception corres- surfaces. The bright-dark stimulation on
ponded to the chemical and physical the retina could yield the disembodied
properties of their physical surfaces in- colors white-gray, but in that event the
dependently of how well these are lighted, perception would be depthless. (So it
and hence in seeming independence of the would actually be if the protuberance were
light-intensity of the retinal image. Per- looked at through a tube which hid its
ception seems to deviate from its stimulus outer margins.) The alternative is for the
and, in the words of Thou less (106), to bright-dark stimulation on the retina (in
show "regression toward the real object." combination with texture and binocular
This is, of course, a valuable achievement disparity) to yield a protuberance, but in
for the perceiving individual insofar as it that event the perception is of both sides
helps to identify objects at dusk or in deep with the same color.
shadow. The conception of two kinds of seeing,
A variety of theories (79, 67) has been the natural kind and the introspective kind,
proposed to explain this tendency in per- is confirmed by some of the experimental
ception, based on the assumption that the results on brightness-constancy. Both
disembodied color is first aroused in the MacLeod (79) and Henneman (54) have
retina and that the object color is then demonstrated that two different attitudes
perceived by some added process. This is are possible for an observer in such an
not, however, the only possible assump- experiment, the objective attitude and the
tion. One might suppose that a disem- subjective attitude. The former is naive
bodied color corresponds to light stimula- and is directed toward the real object.
tion without ordinal stimulation and that a The latter is critical, analytic, or photo-
surface color corresponds to light stimula- graphic, and is directed toward the sti-
tion with ordinal stimuli for a determinate mulus. The two attitudes are not mutually
visual world. Both types of correspond- exclusive, for there may exist intermediate
ence would be strict, but the latter would stages. MacLeod found that "a shift in
involve more variables and would be ex- the observer from the objective to the sub-
pressed in a more complex function. As jective attitude is sufficient to reduce
an example take a protuberance which, considerably oreven to destroy com-
as we argued in Chapter 6, may be con- pletely the phenomenon of color-con-
sidered an elementary depth-shape. Al- stancy" (79, p. 45). Henneman demon-
though both sides are physically white, strated that the objective attitude yielded
the left side is lighted and the right side a high degree of brightness constancy
is shadowed. Why does it appear to have (62 per cent) while the subjective attitude
the same color on both surfaces? Possibly yielded a great reduction in constancy
because the high-to-low step in the (25 per cent). The writer's interpretation
brightness of the retinal image yields an of these facts would be that in the first
THE CONSTANT SI ZES AND SHAPES OF THI NGS 169
case the observer tended to see a visual have handled it and erhaps even measur-
world whereas in the second case he tended it or, if not, that we know the laws of
ed to see a visual field. physics about all material objects, in-
The problem of color-constancy is much cluding unfamiliar ones. Assume that
broader than the problem of perceiving the what we know modifies what we see.
whiteness of the shaded side of an ob- These familiar assumptions will explain
ject. It includes the perception of the a great many of our perceptions, including
whole field of colored surfaces seen at an object with constant dimensions, but
different levels of illumination or in arti- whether it is necessary to call upon so
ficial or colored illumination, and the per- intellectual a process to explain this
ception of surface hues as well as black- particular kind of perception is question-
white qualities. The experimental evi- able. Conceivably the rigid object has a
dence is voluminous and there are difficult correlate in retinal stimulation even though
questions such as how the general level it certainly does not have a copy.
of illumination over the visual world is The retinal image as a whole and in
sensed by the observer. An introduction every part undergoes a continuous trans-
to the subject is provided by MacLeod (79), formation as the observer moves about.
Katz (61), and Koffka (67). The images of objects, moreover, are de-
The implication of this digression into formed when the objects move with re-
color-constancy is intended to be this: that ference to the observer (page 34). As
even colors the supposedly pure quali- you walk up to a mailbox to post a let-
ties out of which vision is built tend to\ ter, for instance, the projected shape of
be intimately connected with surfaces, the object goes through the series of
slants, and edges. Color as it is em- transformations shown at the top of Figure
bodied in space is affected by spatial 68, and when you face a door that is being
stimulation even though it is true that opened, its image is transformed as in the
spots and grades of color provide the drawings just below. Nevertheless, the
basis for spatial stimulation. The innate mailbox and the door retain the same shape
attribute of extensity which color has been so long as you adopt a natural attitude
supposed to possess turns out to be not toward them or, in our terminology, so
the simplest kind of space but merely in- long as you see them as part of the visual
determinate space. world. On separate occasions when you
see them at novel angles of regard, they
still possess their proper shape in three
dimensions. Is this entirely a matter of
The Constancy of Perceived Objects with knowing the mailbox and knowing the door?
Respect to Shape An answer can be given only by experi-
Is the reason for our seeing an object ment and only by simplifying the situation.
as possessed with constant dimensions As described above, the conditions are
from whatever pcsition we view it simply complex. Constancy experiments in the
that we know the object? Assume that we psychological laboratory, therefore, ab-
cj
FIGURE 68. Constancy of Shape
stract from the perception of mailboxes and object would be a rectangle or an ellipse
doors in several ways. First, the object cut out of cardboard. Third, the shape to
whose shape is to be judged is isolated be judged is reduced to the flat outline of
from the background of surfaces in an a single surface which is then slanted
ordinary environment and is placed on with respect to the line of sight. What the
an artificial background. The surfaces observer sees is not a true solid object
surrounding the object are often flat card- but an isolated face of an object. The
board screens, or something equivalent, observer has to judge the particular dimen-
and frequently the environment is darken- sion of this outline which, being slanted,
ed. Neither the object nor the observer is optically compressed for example
is allowed to move. Second, the object the altitude of the rectangle or ellipse.
to be judged is itself artificial. The A scale of graded rectangles or ellipses
chosen shape is arbitrary and therefore is provided for the judgment. Fourth, the
unfamiliar to the observer. A typical outline to be judged is chang.ecLlayt-lae,
THE CONSTANT SIZES AND SHAPES OF THINGS in
experimenter on successive trials so that the objective shape and the shape as it
tks...obaerver cannot become familiar with would be projected on a picture-plane. If
the object, and the slant of its surface is
,...0,1* Myk,
he is asked to pay attention to the pro-
altered in a randEirinicanner from one trial
r
to the next. The conditions of perception
jected shape his judgment is still more
compressed, but it seldom or never reach-
are simplified and reduced as compared es the true projected shape. If the cues
with ordinary vision. Nevertheless, the for depth perception are abs the judg-
significant result of all these experiments ment may approach the perspective shape.
is that the observer can fairly success- These compromise judgments are usually
fully match the slanted object with an taken to be the typical results of the
objectively equal but unslanted object, experiments, and they are expressed by
so long as he takes a naive attitude computing an index of the amount of
toward what he sees and so long as cues constancy, or the degree of phenomenal
for the depth of the object are visible. regression toward the real object (106))
The most obvious implication of this The index is zero for a perspective judg-
result is that we do not have to be fami- ment and 100 for an objective judgment.
liar with an obje:t in order to see the Is it not true, however, that this index of
same shape at different angles of view. constancy begs the question? Using such
Knowledge and past experience of the ob- an index suggests that one is measuring a
ject in qu n are not essential for con- perceptual process and that this is super-
stancy. The constancy of its dimensions imposed on a primary sensation of shape.
must depend, instead, on our ability to The process is assumed to be one which,
see it in three dimensions. For the parti- figuratively speaking, corrects the sensa-
cular object used in the experiments tion. It implies that the visual field is the
the flat face of a solid object this means sensory basis for experience, while the
our ability to see the slant or tilt of the visual world is a perceptual accomplish-
surface. ment of the organism.
There is another result of these experi- If one questions this traditional ex-
ments which has seemed even more signi- planation of the experimental results, what
ficant and which has therefore received theory is better? The basis for an ade-
more emphasis the fact that constancy quate theory has been laid by Koffka
is often incomplete. In general, if the (67, p. 228 ff.). Apjperceptiorj_afthe
observer takes a critical attitude toward stimulus object involves two components,
what he sees, or if he is asked to judge the shape and the orientation. These
the apparent shape of the object, his two aspects of the percept are, as he
judgment becomes a compromise between says, coupled together. The shape is
not experienced in isolation; it is always
a shape-in-a-given-orientation. We can
2
These conditions are fairly typical for ex- suppose that the perceived orientation
periments on constancy of shape. The most combined with the apparent shape yields
recent is that of Stavrianos (100), whose re-
port summarizes the earlier work. a constant shape. If the orientation is
seen correctly, the constancy will be pression which the square face of an ob-
complete; if the slant is not visible, ject would undergo if it were slanted
there will be no constancy. backward from a frontal position.3 The
An effort to verify this theory has re- outline is said to be foreshortened, that
cently been made by Stavrianos (100), is, it is shortened along the fore-and-aft
using tilted cardboard re ctangles. The dimension but not along the dimension
idea was simply to determine whether ob- which serves as the axis of rotation. The
servers who failed to perceive the tilt texture of this surface is represented as a
correctly also failed to perceive the tilted checkerboard, the size of the units of tex-
dimension correctly by a geometrically ture being much exaggerated, in order to
equivalent amount. It could be predicted illustrate that the same optical compres-
that if an observer judged the degree of sion which affects the outline also affects
tilt as less than it was, he should then the texture. How is the slant of such an
judge the rectangle as just so much short- outlined surface sensed by an observer,
er than it was. This determination proved and what is the relation of the slant to
to be not as easy as it might seem, and the the foreshortening of the outline?
prediction was not perfectly confirmed. basis for slant is provided y the grin
The trend of .the results, however, es- stated in Chapter 6, thattb.e..-catinal_gra.
pecially in one experiment, was consistent dient of density in the image of a physical
ith it. surface......,wal,...........amm.m.m.
bears a constant relation ...........,....
to the(
slant of that surface. ..-Thteepness.....of
he Perception of Foreshortened Surfaces I the gradient is proportional to t fie_dgre_e
_- ......... .... ...................-...
In formulating the problem of shape of slant, ---ra-..--EB-6----artection in which the

constancy the tendency has been to think density increases is related . .,...
to the direc- .4.7 ...,...............,..
, , .... ..*..' ...'.
of shape as a kind of disembodied geo- tion toward which the surface facia,,s. The _ .....
metrical form and to think of depth as a principle was_intenriPri toapply --to- any-
kind of disembodied third dimension. The variety _of_.yisual textuze.-.-. .from ...the. finest
kind of shape which manifests constancy, microstructure___too......,..n....*
a gross structure- of-in _
however, is an outline attached to a sur- lines or repeatedkattefns. The steepness

face, and the kind of depth which is of a texture gradient in a certain direction,
relevant to constancy is the slant of a then, is a potential stimulus correlate for
surface.i The perception of a surface is perceived slant. The steeper a gradient of
the central problem, then, if we want to density along a projection, the greater
'understand the seeing of a shape in depth. will be the compression cf the texture, as
JExperimenters have simplified the general one can note in Figure 69.
phenomenon of shape-constancy so as to
deal with the outline of one flat face of an
object, but they have not thereby eliminat- 3The slant of a surface may be defined with
ed the surface in which both the shape reference to a line of regard and a surface per-
pendicular it. Slant is any departure from
and the impression of slant are embodied....I this plane y rotation on a horizontal or a ver-
Figure 69 illustrates the optical com- tical axis.
173
FIGURE 69. One-way Compression of Tex-

ture and the Foreshortening of a Contour
A slanted surface manifests a one-way tually what occurs when the experiment is
compression of texture an increase in tried.
mean density along one dimension of the Other gradients than that of texture-
projected image relative to the other. The density are, of course, supplementary
texture the actual surface quality of stimuli for the perception of slant on a
the face of an object is foreshortened when delimited surface, particularly the gra-
the outline is foreshortened. Although the dients of binocular disparity and of motion.
compression of outline is much easier to The cues of disparity and motion are sup-
observe than the compression of texture, posed to yield depth rather than slant, but
especially in the case of surfaces with a these two concepts have been shown to be
fine microstructure, the latter is also fundamentally inseparable, and the rule
important for the perception of slant. If relating the gradients to the slant of a
the outline of the tilted object in Figure surface can be stated. The steepness of
69 were an ellipse of unknown dimensions a gradient toward crossed disparity run-
instead of a square, the one-way compres- ning from the far to the near margin of a
sion of texture would still yield a percep- physical surface is proportional to the
tion of slant although the foreshortening of slant of that surface (Chapter 6, Fig. 49).
outline would be indeterminate. The ob- Likewise the steepness of the retinal
jective dimensions of the ellipse could gradient of motion over a physical sur-
then be perceived, and this result is ac- face bears a constant relation to the slant
of that surface. In general, the greater the when the two are equal. This is true be-
slant of a physical surface the steeper cause the elements are compressed along
will be all these retinal gradients, and with the dimension. The amount of sur-
the steeper they are, we may postulate, face perceived, therefore, will be in
the greater will be the impression of slant. proportion to the dimension of the surface
These two gradients of disparity and and not to the dimension of the image.
of motion, however, are not primary ordinal If you ask, "Why is the retinal com-
stimuli. The first is dependent on different pression not seen?" the answer is that it
simultaneous projections registered in two can be seen with special attention and a
eyes. The second is dependent on differ- special effort to disregard the slant. (Some
ent successive projections registered in a impression of slant, however, is normally
single eye. One is a gradient of the rela- compelled and the observer therefore can-
tive skewness of the two binocular images not see all the compression.) Without the
and the other a gradient of the deformation special effort, however, the retinal com-
of the image. Both presuppose the exist- pression yields an experie of slant,
ence of a retinal image possessing an out- not of compression. Is quality of
line and usually possessing texture. Both slant is a correlate, not a copy of its
gradients can be eliminated in a shape stimulus. When the conditions of the ex-
constancy experiment if the observer looks periment are such that the texture of the
with one eye and holds his head motion- surface becomes indeterminate (in a
less since this leaves only the bare out- darkened room, for instance) and when
line and the texture as the stimuli for the gradients of disparity and motion are
slant. When they are thus eliminated the also eliminated, a dim outline on the
tendency toward shape constancy should background may come to be the only re-
still be evident if the density of texture is maining variable in the retinal image, and
visible. in that event the slant should become
The tendency to perceive the dimen- zero. The observer no longer sees the
sions of things as constant, therefore, dimensions of a surface, but the propor-
can be ultimately reduced in the shape tions of a pure depthless shape an
constancy experiments to this question: abstract shape pr s umed to lie in the
how can we judge a height relative to a frontal plane.
width when one of these dimensions is
foreshortened on the retina? Figure 69
suggests the following kind of answer: Constancy of Perceived Objects with
4h41(he
a dimension is judged as the amount of Respect to Size

surface perceived, and a surface is The size of the retinal image of an ob-
composed of texture-elements. The gross ject is a very poor indicator of the size
number of elements, or amount of surface of the object, just as the shape of the
perceived, along the compressed dimen- retinal image of an object is a poor indi-
sion is just the same as the number or cator of the objective shape (15). So far
amount along the uncompressed dimension as the light which composes the image is
amount along the uncompressed dimension
as the light which composes the image is
sion is just the same as the number or cator of the objective shape (15). So far
perceived, along the compressed dimen- retinal image of an object is a poor indi-
number of elements, or amount

of surface of the object, just as the shape of the
composed of texture-elements. The

gross ject is a very poor indicator of the size
surface perceived, and

a surface is The size of the retinal image of an ob-
a dimension is judged as
the amount
of
Respect to Size
suggests the following kind of answer:
4h41(he Constancy of Perceived Objects with
foreshortened on the retina?

Figure 69
width when one of these dimensions is
frontal plane.
how can we judge a
height relative to a
abstract shape pr s umed to lie in the

constancy experiments
to this question:
tions of a pure depthless shape an
can
be ultimately reduced in the shape
dimensions of a surface, but the propor-

sions of things as constant,
therefore,
zero. The observer no longer sees the

The tendency to perceive the
dimen-
in that event the slant should become
visible.
maining variable in the retinal image, and

still be evident if the density of texture is
concerned, the object might be either of 1) egradient of linear perspective.

something small ,a.9.4 near or something e more obtuse the angle at which
large and far off. The cone of light rays the lines of a retinal image converge
which enters the eye and is focused as an to their vanishing point, the steeper
image does not differ substantially whether is the gradient of convergence. The
it comes from five yards or five hundred steeper the gradient of convergence over a
yards. Figure 70 illustrates this principle. surface, according to our hypothesis, the
The photographs show two charts for greater will be the impression of slant,
measuring visual acuity, one chart being that is, the more it, ill look inclined to
enlarged to four times the size of the the line of sight.4)bserve the floor of the
other and placed at four times the dis- corridor in the two photographs and com-
tance of the other. They subtend exactly pare the angles made by the edges. In
the same angle on the retina just as they the second scene one appears to be look-
do on the photographic film, and this fact ing dawn at the floor, relatively, whereas
may be noted when the two charts are in the first scene one appears to be looking
projected side by side. The letters on one more nearly along the floor. These gradi-
chart, consisting of E's in various posi- ents, in combination with others on other
tions, are just as distinguishable as the surfaces, establish a point of view in each
letters on the other chart, for the two scene, one being approximately from a
charts are optically equivalent. standing posture and the other from a
In the right-hand scene, when the obser- kneeling posture.
ver puts his head in exactly the position to
align their two sides, the charts come die non- variation in the perceived size
together and appear to be of identical of things in spite of variations in their
size and to be at some paradoxical but distance has frequently 13w studied under
identical distance. The two surfaces ac- laboratory conditions.7The experiments
tually seem to be continuous with one on the problem generally involve the set-
another. In the left-hand scene, however, ting up of an unfamiliar object such as a
the chart on the right appears to be much stick or a cardboard square at a consider-
larger than the one on the left and at a able distance from the observer and a
much greater distance, although the series of varying comparison objects at a
legibility of its letters remains the same. conveniently close distance. The ob-
When it goes back into the distance it server's task is to judge the size of the
looks automatically inescapably
and far object in terms of the near objects.
larger. The photographs illustrate, there- With a naive attitude, and under favorable
fore, both the law of the v
...............................
---TOI- conditions for seeing the distance the
the retinal -tovrard estimates are fairly accurate. Constancy
size -14=-4b4-4.orxespanding,41eris. is then said to be complete. This accura-
cy, however, is only possible if there are
By way of digression, Figure 70 also adequate cues to the distance of the far
illustrates two degrees of steepness object. When the cues are reduced or eli-
(-22 '1 lyI4pol,t
7 47
'0
,2412424201
1' t r 224" ',
r 2 I
w0010422****2-
2;2 ' 02 1,222ft
'r 4k; " 41'22' P2
MMM 4,11
,,r,p/kiktiA44)40 y2,2V22,
PICTURE 1
FIGURE 70. Two Objects with the Same Visual Angle
minated from the scene, correct judgments object, it must be produced by the light
of size no longer occur (57). rays which constitute the image of the
These results suggest that there is no background of the object.
such thing as an impression of size apart
from an impression of distance. In the How is the Distance of an Object Seen?
case of an unfamiliar object, its size is Figure 71 illustrates the two definitions
necessarily a size at- a-given-distance. of the problem implied by the theory of
One ought really to speak of size-distance cues and the theory of gradients. In the
perception, for the two are "linked to- upper drawing the near object projects a
gether," in Koffka's words, both optically retinal image twice the size of the far
and perceptually. The question is, how is object. How can they be seen equal? The
the impression of distance obtained? Since different cues for the distance of each ob-
it is not produced by the cone of light ject must correct or compensate for the
rays w 1 ,constitutes the image of the different retinal sizes. It is easy to under-
176
0 04' 4'
15, A o
4,
rr
4o
P^
474O4444
4,
S4
ti
iffWt
*14 .42/ Qrl
à1 41,44 14?
r' 22
o
441
,C4
t
0F44,0
1}7
it
O eV.
3a 47IM6 IFIire*rg Ire.4
"t 1-
PICTURE 2
stand why cues like accommodation, con- relative skew provide a continuum of dis-
vergence of the eyes, binocular disparity, tance in which the distance of each object
and superposition received all the empha- is then fixed. The gradient theory under-
sis in this theory, for the objects are re- takes first to explain how continuous dis-
presented as if suspended in empty space. tance is visible and second to account for
How the empty space is seen gets no ex- the distance of objects.
planation unless one assumes that dis- Granting that a longitudinal background
tance is computed in the brain by a is necessary, how is the distance of an ob-
mechanism similar to that of the optical ject fixed on the background? Let us as-
range-finder every act of perception sume that an object is seen where its con-
being an exercise in trigonometry. In the tour interrupts the background at that
lower drawing the same objects are re- distance and no other except when depth-
presented as if resting on or attached to at-a-contour brings it forward in distance.
a surface. The spots of the surface pro- This latter effect is produced mainly by a
ject as a background for the objects; the step in the rate of deformation or disparity
gradients of density, deformation, and at the contour. We are assuming that in
177
omr.. *ow
. GME.
I1
....._-
''siwr7..
\ Z.N.,
N, `,. -------.
rA,-----L-.--4'-471.- --.7,--- -:,..7.,,,,E,
\ \ \ \ -' NS:,::`,.........., .Z-...:::-------- -L----!.-k-77--- 7-::c.......--- ^,
\ \\ ., :..,-. _,-... z..,. _-....-__ E-i--_---------,.... :-..:----_ ?--------E4_.--
\
,---7,._.
=--
, ..1,, ",.., ..._ SN. ...._ . 1..... -

....= ---' -----
\ \ \ N, N. ,,, ...... ." . --, .".

\ \ \\
\
N, N.,
''''' '''`. ,... ..
'",.
-, ',..
'`,.
\ N,,,
s,.
\ \
N. N.
\ \ \ N
N.
N.
1-1 X 1-1 rs,
FIGURE 71, Two Objects of the Same Size but Different Visual Angles
the absence of what is called relative terned cover on which two white oblong ob-
motion or stereoscopic depth a contour is jects appeared to rest, one large and far
seen on the background. This would ex- on the left, and one smaller and nearer on
plain why, in the monocular vision shown the right. This impression is clear even
in Figure 71, the object appears where it in the photograph. The observer looked
physically is instead of nearer and small- through a small hole in a large screen at
er within its cone of rays. the near end of the table, using one eye
One is free, of course, to assume in- and keeping his head motionless. His
stead that an object is seen resting on or angle of view was approximately as re-
attached to its background because we presented.
know from experience that objects are Actually, the object on the left is the
either at rest or physically attached to a smaller and nearer, as can be seen in the
surface. Although the effect of past ex- second photograph. It is simply a card-
perience no doubt contributes to the ex- board rectangle raised about three inches
planation, it is not a necessary assumption. from the table-top by a rod which is in-
The former hypothesis can be tested ex- visible to the observer. Its distance is
perimentally. Figure 72 represents a set- given by the optical contact of its base
up for doing so. The upper photograph is with the background. The dimensions of
approximately what the observer saw in the rectangles are such as to project
the experiment: a long table with a pat- identical images at their respective dis-
FIGURE 72. Distance as Dependent on Contact with the Background
+OK*
apkw* I OW 001,.! An.
.406,4,4444 ,,,,,,
44g '44""' 0401
0,44r.14
$4°
1r
V*.
kr,
*11.7110'
71,
11
44'
1:"tr.61k
41,
*42 ,0114,;.;,,
PS`
tances, so that the two images in the up- cue is studied in isolation, as it would be
per picture are physically the same size. in an empty frame, it should be regarded
To an observer unaware of the physical as an associated variable, an inference,
set-up the illusion is astonishing, but or a probable indicator of distance, not as
when he is permitted to look at the ob- a true stimulus, since although the sti-
jects with both eyes and to move his head, mulus gradients on the ground produce an
he suddenly sees the left-hand object as increase of distance upward in the visual
much smaller and suspended in the air. field, the stimulus gradients on walls and
Even with full knowledge, however, the ceilings do not (Figure 26, page 71).
original appearance returns, or strongly
tends to do so, when monocular motionless The Perception of Distance and of Scale
vision is reemployed. The conclusion is The evidence of the size-constancy ex-
that a perceived object under these condi- periments is not entirely summed up in
tions recedes within its cone of light rays the conclusion that the size of an object
until its surface is continuous with the tends to remain the same at different dis-
background surface. Other things being tances. Some of the experiments point to
equal, two optically adjacent surfaces a more general statement; they imply that
tend to be adjacent in depth a pheno- the dimensions of things, large or small,
menon which was also observed in the are comparable at different distances. This
acuity charts of Figure 70. general formula comes closer to describing
Painters have always known that the the ordinary experience of the world in
higher up an outline is placed within the which we live since, under favorable
frame of a picture, the farther away it will conditions, we can estimate a range of
appear, even when no background what- sizes at a range of distances. As an ex-
ever is represented to give perspective. treme case we might take a flier looking for
This is the recognized cue of relative up- an emergency landing field. He will es-
ward location in the visual field. Of two timate the sizes of all the cow-pastures
objects in a perfectly blank frame, the for ten or fifteen miles in any direction in-
upper will appear to be farther away. The cluding the one directly below, all being
explanation is probably that a blank back- unfamiliar and no one being the same as
ground suggests a terrain or floor the another. ese facts suggest that we
fundamental visual scene of Figure 19 erceive a in theuaLataxid--_,
more strongly than it suggests a wall or a wh ch might be called scale.
ceiling. A terrain or floor is always For objects at any given distance we
present to vision whereas a wall or ceil- possess a subjective scale of sizes from
ing may not be. "Upness" is therefore a very large to very small such that we can
faiily reliable cue to the distance of an judge immediately and rather accurately
object in the visual field, given the prin- not only when two trees are equal in width
ciple of contact with the background and but also whether one tree is half, or twice,
granting the great frequency with which or three or four times the width of another.
the ground is the background. When this Size perception, to describe it accurately,
involves a scale of sizes in relation to ground at the point to which it is attached,

which any given size or dimension is and that is why its apparere is linked
unique. Such a scale is psychological; to its apparent distance.
it is something we carry around with us and The impressions of scale and distance
it is implicit in the very process of per- are so related to one another that with in-
ception; it is therefore not to be con- creasing distance there goes an unvarying
fused with the conventional scale of scale. This is the rule for ordinary per-
meters or feet which depends on a set of ception of the visual world. If the ob-
operations using carefully constructed server tries to see the world in perspec-
pieces of wood or metal. The conventional tive like a flat picture, however, the re-
scale is a psychologically complex affair, lation between them is the same but the
depending as it does on the learning of impressions are different: with unvarying
techniques and concepts, whereas the distance there goes a decreasing scale.
implicit scale of visible size is a primi- The reason for the landscape-painter's
tive fe ure of perception. 4 ability to see perspective in his visual
T e size constancy of objects, in the field is that he has retinal stimulation
light of this conception, is a by-product which ordinarily yields joint impressions
of the constant scale of the visual world of both distance and scale. When he makes
at different distances. Scale, not size, is an effort not to see the distance, the
actually what remains constant in percep- scale is correspondingly diminished, and
tion. The gradient theory can account this altered impression is what we call
for this kind of constancy, for one can seeing the perspective.
assume that the perceived scale of the Figure 73 illustrates what is sometimes
background is a function of the same sti- called the perspective-illusion. All three
mulus variables which yield the continuous cylinders are the same size on the page.
distance of the background a different It is not an illusion at all but a demon-
function, it is true, but equally dependent stration that apparent size depends on ap-
on stimulation. The size of any particular parent distance.n illusion ma be de-
object is given by the scale of the back- fined as a_p 110
consistently not in asseemeat

wìt me s retneats_af---th.emubrerr---gtving
rise to the perk
4T he study of psychological scales, or clear enough for objects; it is when pic-
subjective scales, is still in its infancy.
Enough is known, however, to suggest that tures come into csastherration......tharzur.
they permeate all forms of perception and that ITlinking about illusions is apt to become
the traditional emphasis on perceived objects
has too long diverted psychologists from con- Eon triser--- Insofar as this picture is a
cerning themselves with perceptual scales. A substitute for objects (and our chronic
program of experiments on various perceptual habit is to see pictures thus) the increas-
scales (area, length, angular size, numerous-
ness, and so on) is currently being conducted ing size of the cylinders' is not illusory.
under the direction of Dr. John Volkmann at Insofar as this picture is itself an object
Mt. Holyoke College, but this research has
not yet reached publication. (consisting of black lines on white paper)
FIGURE 73. Size as Determined by Distance
the increasing size of the cylinders is stancy was the rule. The significant result,
illusory. however, was that the interspaces could
be estimated without notable uncertainty.
The Rigidity of Visual Dimensions
Dimensions in the air were judged with
It was proposed at the beginning of this less accuracy and with a tendency to-
chapter that constancy tends to hold for ward underestimation as compared with
the distances between objects as well as dimensions of a solid object, but never-
for objects. If this is true it would help to theless with an approximate constancy of
explain why we find it so easy to conceive size. The implication is that the dis-
of abstract geometrical space as we do. tances between things tend to be visually
Although the tendency toward a constant rigid as well as the things themselves.
size and shape of interspaces with varying
distance and angle of regard is suggested Does Size Constancy Break Down at Great
by ordinary observation, the phenomenon Distances?
has been neglected by experimenters. One more wartime experiment is relevant
During the war R.H. Henneman and the to the kind of theory being developed. Al-
writer set up an exploratory experiment though the perceived size of an object
which, incomplete though it was, may be which recedes in the distance has been
worth describing here. The observer was recognized not to diminish at the same
seated at the end of a thoroughly cluttered rate as its retinal image, no one has ven-
room containing tables, cabinets, boxes, tured to suppose that it does not diminish
shelves, and furniture. Among these ob- at all. At some eventual distance the ob-
jects he had to estimate 20 specified di- ject ceases to be visible, and what is
mensions, some being the dimensions of easier to suppose than that it does so by
solid things and some being dimensions way of becoming smaller? It has therefore
in the open air between them. The dis- been assumed that size constancy neces-
tinction was not as clear as it sounds, for sarily breaks down at large distances, and
there was always a background surface perceived size then tends to become per-
behind any dimension. They varied be- spective size. The implication is that in
tween extremes of two and forty inches. outdoor space or aerial space, as con-
The subject made his estimates by pulling trasted with the room-sized spaces of the
out a steel measuring tape to match the psychological laboratory, the appearance
specified dimension. It should be noted of things necessarily tapers of toward the
that the length, width or height was al- horizon and the features of the terrain per-
ways optically diminished by distance, and force look smaller than they are.
might also be foreshortened, relative to The experiment to be described pro-
the tape. The estimates of the fourteen vided a test of this assumption, for size-
subjects varied among themselves and estimates were obtained out to a distance
from one dimension to another, as would approaching bare visibility of the object.
be expected, but the over-all mean error The situation is shown in Figure 74, which
was a slight overestimation. Size con- represents several trials of the experiment
;
,
.41.,14g1hiA
4,00j
A yris t 4r
tip '1,"4?;j4;',c ,;(40040044411,0

d'At 41441'40, ,114.,,;44'1 w
44nc1,,,4;11,0!14 r
'11}1'.',;1:''b hr0 6644

t'F' 44V
Y41;,
*114041-
c."
4A' ""
krt 1t 5Y,
FIGURE 74. Judgment of Size-at-a-Distance
, ',0 ,
. , , ,,,,, } 4 ,' , f ,
'
"'
I
lr' .(1,7 .1. A116

.,, ,6 4414 .4,1, O't ljt...;,k,ONY ,1,101# r
"'" , ,r 4;4. V: 4'',(1544,4011, h
01 y,k I
,7,144, VekTWO."
41 4 q, § Oft ; !4514
',;,,,itoot,..41( 0 4
Wt.
,Argt40'44,
'4, I '4'4 ,M
IM, kni
yi
41,4*ft:
t!
' , , ' ,' 1, A 1 0 ,' '''''' ' 1 '1,'' ',SO'. ft's11,,;(i411.. Oil, '44',;, '',.."14t,q1';A '4F4',,,i'1,4' 'i,e,411',44q,, ',' ' : r'i''' 1, L"''' ,'"! ,'',,' t'' ',e) ", '4"".l'Alvt'444."' 4.,...:'0 ,,PS4',, '4',4"14 ;;t 4 :'%, 4 ;,4'; IA '''' 1 ':'' t,"::t '4g '1'n'''''i, l'tI V"r14''S"'`," i '4'4 '''')4". ''';;;S:1'1":',";''''''s. :i'44',;"' ' ' I"r'ottAX';'4'W '''1' 4 0 '44.tiLi,,,'N'1,4't4lell'I's4tIkief,:titi,'IVtriTt*:444,t ,Sq.,*
, .",%.,1.4,,,,,,,,,,,I., ...t'irr, 4,.,04,,4,,,,,j/,44,f,,,, 144,44*,044,,,,;4,4p1, I ti7,40,4, ,r.14,, ,,,,11 y,y;,;,,, i,, ,,,,, ,,,,,, , ,,4',p,,,,0,7,1, '.44, 0,s, 4,,,i,,,,,,, ,A4,,., 4 ,,,,44.414'As ,,,j,,,,' ',,`,'1,4", r`V, l',/v,i4",.;!,R4Vt 11,04,4:;ie,,,144'sit s`,4,1,1Y frir,,$44;44,4.;sksi, `,41:1:','-t,,,,,,` -,,it s, ', os'''',;11$4,, , ' '4 '4,44k )','")"`si';', Ast41404is';'' t'`,
,,, t,..s. ,,...:,,,,, . :,,:A0, .,,,,,'.,4,, 's, , 4 44 4sFs '," ..,p , , ,, ,, 14',.' , S, Ls' ''46,' '1,C' ''',f41.'444t1gSj;$40.4 .4'''''th.#9.0.4'4,14 ,"4.-.,t,,,,,,, 44 64,4,1,1,,,,i, s,..'14'',' 4 ,:, 'r :' 4 45,4,11444 aortt- '0,'" -S. , ,§,.'
rAl*S',1441041,0,4?'/1"11 !;,',"'' '41 F 4;4 ' sA. 4444, ,F Fs

,Ft44,041N00,0;,
,,,,±"', , ',t'";,I%,,',,;'i,',4';'1 ,
"400441/1744i4J}t4e 1'44* 4,0'44; F' rF4,:; 4',44 V tre, ts2`4ANY,r4s'Aii,, ,4 4tx ' ' 4
'"4
, ,,
4441': ;''Pgirr,xet,F4',,,f,.,s4,t4;;:kr4re'tt`k.,140
i.,,4'%,*:,60,0,,,lort,p -40Sti,,ss44,40,'
9,,
404404,4V4s, '
;44:
, *,'4444,,,,14.;',040444f, 44,
,6
, ss,,,,ffr,k4'4Vqt
101 P(14 k'I'4011474W
,"4 SS,1,4;00,01,
4'4
,01,q,e4Of
sl
114
, , ,
440,/ C 44,4%
S,40
'
i, 4444444'/
14L 44-
I,:
:1404W 4,4 .4 \ 404/44344,,, "O.; s.

,
Irk
41','0;111',4,4' 1 ' e''' 1

.,'_ ee:',::,,, ,,,,. ,.. 01
. r.
,,, ,,L,.., .
(
.' ,e:. t 4,.-,
14,1 4,414ti; ' , ,,,
' , 4,440^1t "P',; ,4 4 401,10$ ,, ,, oft
...
,, .1t,.,,
.
.-,, ,
). ,.
,,,,,..1., ,, .....,,,,.1....
, .
f44,4 ; .40A '''''','; 1 ',
'' ;1,Y ' *" ' 1,04' ; ' _,4.4 ,' 4:,,,,r14,k,4;i4/,' '' 4.4 ;4# 04', ..,.4,4 ',
,AA
"
'' ; (:', ;%44'j'.
4
' ' ,,, ,s' AJivot ,,::441 4., it' ,,, i''Ir -, ,' ' ' S,'0,^ 4,4 4.14 4414,'r
FIGURE 74 (Contd.) The bottom photograph is the same scene as the one opposite
it except that the camera is higher, "looking down".
4 r 4 ' ' I .'

4 "1 :- ,, ;II '. ' ., 4 1'1' ' ,
, , ,; I ,I , ,,,,. ,,, ,,4 I ,,,, 1, ,,epk:4k44 .',4' r 14*$ 0, v4

i ; . '' ' ' 14 ;', ' 4, vl ' 0 44` 4'; j
' ,
,
v 1 ' , , , ',`'''
,I ' r , ,' , ,, , , ,,, 1,4 44 ' 4, . ' 4I 444 ' 4,,,, ' '4.4t I ; , , 4, 4 4 , 1 '' ..' t.,,k I ' tle 4 4 ,4, 4 ,,44,,, 41:0
( ', ,,414 4,,%3444 4% V;1,,,211;11,/a.1!1fr.:1' 1/,'11;,',,,,rill 140.,),t,ti,..,
. , 4,.4,4., ,,,, , 4,S +.N11 4 ,,,; ;,, , ft; 44 \v" "r ,.. ' ' '' ' ;'104,;) ';;'.04 P: '' .4t*4 "44 ;""4140%/;,4114 4 .4),`,24".is,NA, Iv t4p 10t4.104,41`,°4414,11Nq
". e
' 1 ;
3,
6t r
' '4';' " 0' '.4i;',44to' '''''''''''' '''''''';;44'1;":.4;4'4; ' 4 '',V ;4°W 1,'`;,'; %,;;,,14 4 41'0,,tvi '41,14,4e,V;14;0141441; ' "-ve.14;14v)4:1
'' , 4
./Sk ''' , '
' ' k:" '. r' .4 v ,,, , '',0404i*. "o4..
, ,,'...,
' `
, ' ;
4, ,4 s ,t , F
, , '''' ','' 't''" .
,, ,,,,04..y.., AOIAI),?"A'Al17" ;1V,",04 ''Y.0.;,:" 111,0',-,.,44,'4,'Ao.i.A4''oe'.%i'' 74*V Y541k*1. ,A ,,,ti . .. t, ..

'_...,....,..A.
,444.
stoo.s.1
.14. 6 441;
4,4:7",411V 0 ` 4,,,,R,
' t`4
r rr r 'Zrro
,r444.1,14,
!,''''' ',''' 4 tct cio,,:;t,t,-,4,1%,,, , ,..,,,,, . 4#411 r, , ,7 ' , ,, ,, ,4 . , f
' '''''r""' ' ' ''
,
,
,
'
/ '," { tOyiNIVIO 4 , , , 4 ,: ,,r. ,44 , , 4414-s?, ,....s 411 4j nps (, rs 444, 41 , , tri4 41 4,3,,
,s4x...44''4 t' kr , n ,Il 446, 1 , oC724k ,..,',,V.,:,..",
- 4
,
,' OM, :4!,,ster:440,4*Ftt
alMk4 '147;
A
4,04,is 44t, At, 410 -I, 4' 4,,4444,4, 1, 1442434 4s AC4 ess gFF4F
a'
40V* k4
iffekt,44:"
,I44K
44" '`44/%41,104
ION 0*
*144415(4,"4
'44 44
4. V4444:4,4"
4,4 r ;*440
4,47, 40,,
4.
c'.3
.
,
-.4044'PootoAr. 14
s44 Yr
4101S ' t'140
4,4,4
4'0
4,o44
441.,14414'
' rA " A/ S 544'4'
'4
4414
4,44/i I
1
`4t.: 14
4
bs
11'11;4
'1%44* t,t5 '4 0
.4 'itr040's
' I4
'44:4 St, ,,,
4,
(36, p. 201 ff.). From the observer to the parable. This outcome is surprising, for
low hills in the background the distance at 784 yards nearly a half-mile a
is a half-mile or more. The field is the man-sized object is beginning to be
sort of perfectly level cultivated land difficult to make out.
found in the coastal plains of southern Our question as to whether sizes neces-
California, and it was selected after much sarily become smaller in perception before
exploration. Its texture was fairly even, they reach a vanishing point seems to be
but without any furrows to give linear answered in the negative. Under favorable
perspective. The procedure was as fol- conditions for seeing distance, as these
lows. A wooden stake, from fifteen to were, an object can apparently be seen
ninety-nine inches in height, was planted with approximately its true size as long
at a previously measured distance by one as it can be seen at all. Its size does not
member of the experimental team who then become smaller but only more indeter-
hid in a conveniently invisible irrigation minate. In this experiment there was no
ditch. The observer, in complete ignor- question of an index or amount of size
ance of its size, faced about and esti- constancy, for that was complete, but only
mated its height, either in scale-numbers, of the increasing variability of the judg-
or by saying "smaller than one" or ments with increasing distance.5
"greater than fifteen." These judgments The critical reader may notice in Figure
were repeated for different sizes of the 74 that there are one or two ways of
stake and at different distances, 150 judg- consciously inferring an approximate size
ments being obtained from each of fifteen for the distant stake. Inferences of this
observers. The averages of the estimates or a similar sort are always possible to
were then computed for the different sizes, some extent in judging objects on the
distances, and observers. ground. With the exception of one ob-
Let us examine the results for the server who was a professional psycholo-
seventy-one inch stake at different dis- gist, no one became aware of these cues
tances, since they are typical. The cor- during the experiment. What the average
rect match was with scale-number 12, subject perceived was merely a size-at-a-
which is also seventy-one inches high. distance.
The mean estimate when it was planted
at the same distance as the scale (14
yards) was 71.9 inches (S.D. 1.8) show- Conclusion: The Objectivity of Experience
ing great accuracy. The mean estimate The general implication of this chapter,
when it was planted at 224 yards, a three- taken with the preceding ones, is this:
minute walk down the field, was 75.8 the objectivity of our experience is not a
inches (S.D. 7.3). The mean estimate paradox of philosophy but a fact of sti-
when it was planted at 784 yards, a ten- mulation. We do not have to learn that
minute walk and nearly to the end of the
field,' was 74.9 inches (S.D. 9.8). The 5
See (39) p. 210, for other details of the
four intermediate distances were corn- experiment.
things are external, solid, stable, rigid, not require for its explanation a process
and spaced about the environment, for of construction, translation, or even or-
these qualities may be traced to retinal ganization. The visual world can be
images or to reciprocal visual-postural analyzed into impressions which are
processes. object-like, and these impressions are
Such a conclusion upsets completely traceable to stimulation. The fundamen-
the traditional interpretation of percep- tal impressions obtained by introspection
tion. The experiments of Gestalt psycho- are not colored bits of extensity but varia-
logy undermined it, but they did not over- bles like contour, surface, slant, corner,
throw the conviction that somehow we motion, distance, and depth, in addition
construct our world of things and events to color, all of which correspond to the
out of impressions which are themselves variables of a distribution of focused light.
not thing-like. The conception of sensory These impressions do not require any put-
organization implies a putting-together of ting together since the togetherness
non-objective elements in perception. On exists on the retina. The suggestion is
this theory, the data of sense still have to that, philosophers and estheticians to the
be translated into an awareness of objects contrary, order exists in stimulation as
and events. According to the present ar- well as in experience. Order is just as
gument, however, the objective world does much physical as mental.
Geometrical Space and Form
The Space of Geometry The Problem of

Visual Form (--.0sychophysical Approach
to Form Perception
In Chapter 5 it was argued that the ab- point of view holds true for interspaces as
stract space of points, lines, and planes well as objects, it is possible to imagine
was a poor conception with which to be- that something like matter or an ether is
gin the analysis of how we see, for no one also constant in these respects. If the
has ever seen it. The investigation of ground tends in perception to have the
space-perception got off to a bad start by quality of scale, so might the boundless
taking the space of geometry as the pheno- firmament above and the earth beneath. If
menon to be explained. We chose to study this scalar quality seems to extend as far
the visual world instead. Now that we have as we can see, perhaps it never ends.
a theory of the concrete phenomenal world, All these features of visual perception can
however, do we also have any insight into be abstracted, and this abstracting was
the conceiving of abstract space? Why do what the Greek geometers were the first
we find the postulates of geometry satis- men to do. If the world is emptied of
fying and why is empty space so convenient objects and only their ghosts in the form
and easy to imagine? of points, lines, and planes are imagined,
Similarly, we chose to study objects one thereby simplifies visual thinking.
with surfaces and edges instead of geo- Size constancy then finds expression in
metrical forms. Triangles, squares, and the postulate that two straight lines in the
circles, however, can be drawn on paper same plane can be drawn which, however
and can be perceived. How they are per- far they are extended, remain equidistant.
ceived is a puzzle. What is the status of This is a form of Euclid's famous parallel
these abstract geometrical figures in postulate, which is one of the characteriz-
psychology? ing features of "self-evident" geometry.
Empty space as thus conceived has been
The Space of Geometry described as analogous to a box without
If the tendency toward invariant sizes sides. The unvarying sizes and shapes of
and shapes of things in spite of variations things also find expression in the abstract
in their distance and in the observer's idea that a geometrical form can be shifted
188
GEOMETRICAL SPACE AND FORM 189
about and made to coincide with another non-Euclidean geometries, correspond-

identical form. This leads to the axioms ingly, developed later in the history of
of congruence and the identity of dimen- mathematics and are harder to understand.
sions and angles in Euclidean geometry. The space of Euclid, the Cartesian co-
Empty space is rigid, as are solid objects. ordinates, and the Newtonian universe
The self-evidence of these postulates, were based on the visual world of human
however, was somewhat shaken when, cen- behavior. We are, after all, terrestrial
turies after Euclid, geometers began to an and our actions presuppose the
study optics and perspective and to ex- ground, upright posture, and forward loco-
amine their visual experience . In the motion. These abstract to three dimen-
visual field, parallel lines meet at a sions, and to a rigid space with absolute
vanishing point, and this suggests that a location and absolute motion. This rigid
location called infinity is thinkable. What space is perfectly adequate for terres-
are the consequences if parallel lines are trial measurements and the analysis of
assumed to intersect in a specific way? terrestrial events, just as the visual world
Objects change shape in the visual field as is adequate for locomotion and our dealings
they move. How then can one describe with ordinary objects. If it is not adequate
these transformations? Things expand in for astronomical events and measurements
the composite visual field from one pole involving the speed of light, as the theory
and then contract to another pole 180 of relativity suggests, it is at least the
degrees away. What kind of geometry does primitive conception of space from which
this imply? These were not, of course, the the more erudite notions are derived.
literal questions which the non-Euclidean Whether the space of the physical uni-
geometers asked themselves, but the converse as a whole is or is not Euclidean is
nection between the study of vision and the a problem for physicists and astronomers.
new geometries is too suggestive to be ac- The suggestion has sometimes been made,
cidental. In recent times these geometries however, that perceived space is non-
have flowered and, in their applications to Euclidean, and such a conclusion has
physics and astronomy, have been popular- recently been reached by Rudolf Luneberg
ized. Everyone has heard of curved space on the basis of a mathematical analysis of
and nearly everyone is puzzled by it. Why binocular disparate images (76). Most of
is such a conception at once reasonable us are bewildered by such a conclusion.
and unreasonable? The first difficulty with it is that it fails
If geometry is an abstraction from ex- to distinguish between an imagined space
perience, then Euclidean geometry is an and a perceived world. To say that a
abstraction from our experience of the space is not Euclid's space may be in-
visual world whereas non-Euclidean geo- telligible but to say that the visual world
metry is an abstraction from, or is at least does not follow Euclid's postulates vio-
suggested by, our experience of the visual lates common sense. A more important
field. The latter is a more sophisticated difficulty, however, is that it rests on a
experience and is harder to attain. The confusion between pictorial seeing and
everyday seeing. The visual field, to be seem to expand as we move forward in the
sure, is non-Euclidean in the sense that environment, although the retinal image
its geometry is based on perspective. The does. The expansion is a stimulus cor-
visual world, however, is an experience of relate for the sense of moving forward. An
quite a different sort. Its geometry is object does not appear to contract as it
presumably based on the practical biolo- recedes, for the contraction is a stimulus
gical necessity of estimating the dimen- for its recession. The face of an object
sions and judging the shapes of the en- does not look compressed when seen at a
vironment we live in, for only if things keep slant ,since the compression is a correlate
a certain rigidity of appearance can we of its looking slanted.
identify them for what they are.' The changes of the retinal image which,
it always seemed obvious, ought to pro-
The Problem of Visual Form duce a change in the size and shape of the
The so-called constancy of the shapes object are actually stimulus variables
and sizes of things seems to be a corol- which yield changes in the distance and
lary of the perception of a visual world the orientation of the object. The size
whose surfaces have the quality of slant and shape of physical objects are not re-
and the quality of visual scale. Under- presented the retinal image although
in
they are specified by it. The natural
lying this visual world is always the
assumption has been that an outline on the
primary surface of the ground. We can now
retina yields an outline in perception and
better understand why the world does not
hence, that a deformed outline should
yield a deformed percept. This assump-
1When
Luneberg suggests that perceptual tion is reinforced by our interest in the
space is the hyperbolic type of non-Euclidean lines and forms which we can draw on
geometrical space it only confuses me. The
obstruse and theoretically unclear set of facts paper, for which it holds true that a modi-
which it seems to account for (the alley experi- fied drawing yields a modified percept.
ments) are not obtained in a situation with full
illumination and optimal conditions for depth When the retinal deformations of outline do
perception. The argument is based entirely on not have the expected effect in perception,
an analysis of binocular disparity of images,
leaving out of consideration the geometry of we are faced with a paradox. How can the
perspective as it applies to size, texture, shapes and sizes of objects be constant
motion, and other types of stimulation. Per- in experience? Perhaps the fundamental
ceptual space as we get it under optimal condi-
tions with constancy of size and shape is error lies in making the original assump-
so plainly and simply the space from which tion. Conceivably, a deformation need not
Euclid abstracted his geometry, and this con-
ception is so illuminating for all the constancy yield a change in the perceived form.
experiments which yield 100 per cent con- But if that is true, what enables us to have
stancy, that to deny it for the sake of the alley
experiments seems unjustified. The application the experience of a visual form in two
of mathematics to space" perception is a fertile dimensions?
field, but the conclusions will be no better
than the assumptions with which the mathema- Form, as we rerer to it here, means pro-
tician starts. jected form a silhouetted shape as con-
trasted with a shape in depth. This ab- all of which can vary continuously along a
stract geometrical form is not, we have scale or dimension. If we cease to think
argued, a primitive spatial impression at about forms as a set of geometrical en-
all. The primitive impression is a form- tities and concentrate instead on the
in-depth, and its two aspects of projected transitions between them, as we have
form and depth as such are abstractions learned to do for colors, our thinking about
never experienced in isolation. What we the visual process may be clarified. This
are concerned with is a conceptual experi- endeavor will be made in the remainder of
ence represented by outlines on paper. the present chapter, but first, we may ask,
The main stumbling-block in the whole what stands in its way?
history of our efforts to understand percep- Against this conception there stands the
tion, one might venture, is the tendency to tendency to think of a form as unique.
think of form as two-dimensional only. How can there be more or less triangular-
The form-on-a-plane, the geometrical form, ity in a triangle, we ask? A form, accord-
the drawn form such forms are literally ing to the emphasis of Gestalt psychology,
nothing but the shadows of things, that is is something more than the sum of its
to say, their projections. Although Ges- parts. "A shape," to quote Koffka, "is
talt theorists have recognized that depth itself and nothing else" (67, p. 175). As
is as fundamental in perception as mere an assertion that configurations are not
extensity, their central problem and guid- reducible to elementary point-sensations
ing concept was the shadowy form, not the this emphasis is surely correct, for the
substantial one. Might it not be that the triangular quality of a triangle cannot be
dynamics of projected forms, so diligently derived from the points as such. It does
studied, consists not in the laws of form not, however, tell us how to analyse the
as such but in the laws that relate them perception of form or how to understand
to solid objects the laws of projection the ways in which we see and discriminate
and transformation? forms. It suggests that form is unanalys-
This way of considering size and shape able. If every form carries its own law
points toward a conception of visual pro- there can be no laws common to all forms.
jected form which is very different from As a scientific hypothesis this emphasis
the traditional one. Perhaps a closed is of no value whatever, once the principle
outline is not an independent entity as is accepted that we need to study forms
we have tended to think but some kind of and not retinal points. Although unique
variable. Perhaps we should conceive and unanalysable forms play a leading
form not as a thing but as merely one of the role in Gestalt theory, the actual ac-
variables of things. The projected shape complishments of this theory consist of
of a perceived object would then be only attempts to find the laws of form.
one of its visual qualities among others The only laws of sensory organization
such as the slant of its surfaces, its size, discovered, however, have been principles
its color, its texture, and its distance, such as proximity, similarity, symmetry,
and good continuation (119). The criti- in which a subject is required to draw (or
cism can be made that they are intended recognize) a visual pattern from memory or
to explain only why the perceived form is from just previous observation yield no
different from the retinal form and that information whatever about why the sub-
they are not relevant to the main question: ject's response is like the original pattern.
Why is the perceived form specifically re- That is a problem of psychophysical cor-
lated to the retinal form? Only if there is respondence, analogous to why a certain
a specific relation is the perception good wave length looks red. What these ex-
for anything, since only thus can it be periments do yield information about is
related to the outside world. This rela- how a subject learns to discriminate
tion cannot, it is true, be one of simple between similar patterns and how we learn
pictorial correspondence but it does not to conceptualize objects. Such learned
need to be pictorial as long as it is discriminations are of the greatest im-
specific. The Gestalt theorists, however, portance but they must be founded on un-
being aware that the perception is not a learned discriminations, and it is the lat-
copy of the retinal image, assumed that it ter with which we are now concerned.
could not be wholly specific to the retinal
A Psychophysical Approach to Form-Per-
image (like other theorists before them) and
ception
went on to account for this discrepancy
by a theory of dynamical processes in the ow do we judge the shapes and esti-
brain. The laws of sensory organization mate the dimensions of the environment
were the expressions of such processes. we live in? This is the question too often
These principles of physiological self- forgotten in discussing the perception of
arrangement gave a simple explanation for abstract projected form. For a mobile
errors in form perception. In some respects animal like man, the very essence of en-
self-arrangement in the brain seemed a vironmental shapes and dimensions is that
better explanation of the errors than a dis- they are successively transformed on the
torting effect of well-remembered or fre- retina. If these locomotor transformations
quently experienced forms, but in other all yield perceptions of the same objects
respects it did not, and a controversy why could not other transformations yield
which was never settled arose over the the perception of all possible different
role of sensory organization versus past visual forms ?rtain regular transforma-
experience in form perception (121, Chapter tions (expansion, contraction, a one-way
4). Neither a dynamical brain process nor compression, a certain kind of skew, and a
an interpretive brain process, however, is simple transposition over the retina) go
relevant to the primary question of the with a perception of the same shape. They
psychophysical relation between retinal are related to t geometry of perspective
and perceived shape. and parallax. Certain other transforma-
The writer has concluded, after puzzling tions, not experienced as continuous during
t visual form perception for a good locomotion, go with the perception of dif-
1> i
lcars (35), that all the experiments ferent shapes.` They are not confined to
the geometry of perspective and parallax. classes or categories except the so-called
Conceived thus, visual outline forms are transformation-groups, and the closed
not unique. They could be arranged in a visual outlines now being considered are
systematic way such that each form would all in the same transformation-group (25,
differ only gradually and continuously from Chapter 4 and 5). Any closed form can be
all others. Such dimensions of simi- transformed into any other closed form by
larity among contour forms have never a perfectly gradual or continuous change.
been explored (and there must exist an The psychological quality of shape which
enormous number of such dimensions) but changes with such a continuous transforma-
the fact that modern geometry can specify tion is what needs investigation. The nor-
general modes of transformation suggests mal or standard shapes familiar to every-
that the exploration ought to be possible. one are no more than special points of
"The concepts of modern geometry," ac- anchorage on a continuous dimension of
cording to Cassirer (21, page 8), "derive variation. They are norms or standards of
their precision and true universality only reference for shape, not entities of shape.
from the fact that the intuited particular Figure 75 is a chart of a few very simple
figures are not considered as pre-given dimensions of variation. It illustrates the
and rigid, but rather as a kind of plastic modes of transformation which comprise
material capable of being moulded into the the manifold of all rectangular figures. The
most varied forms." square lies at the center of the chart, but
The seemingly infinite variety of visual there is no unique square, for the equal-
forms is the crux of the difficulty. This sided shape varies in size along the dia-
manifold has usually been reduced to order gonal. The discriminable qualities of
only by classification, beginning with tri- 'shape are very evident but they are not so
angles, squares, circles, and the like. The easy to name as one might suppose. Thin-
result is a set of groups or mutually exclu- ness to thickness and tallness to short-
sive categories analogous to the classes of ness are perhaps the most obvious terms to
individual things and persons implied by apply. The dimension of variation running
Aristotelian logic. The Greeks, and es- from upper left to lower right is a quality
pecially Plato, thought of geometrical forms of shape which, however it be named, ap-
in this way and the tendency has per- pears to some observers as simple as the
sisted. Classification is not, however, the supposedly elementary quality of warmth
only or the best way of ordering a mani- to coldness.
fold. Serializing is more apt to bring out The chart may be conceived to include
the fundamental relations between things either transformations of the same object,
(75). If we are ever to understand exactly which we get when we move about in the
what yields a perception of shape we must environment, or transitions between dif-
study the dimensions of variation of ferent objects, which we perceive as
visual shapes. qualities of outline-shape. The dimension
Geometrically considered, visual forms from lower left to upper right is a change
do not fall into any mutually exclusive in projected size such as accompanies
IMMINO.I. IIIMOIN.........,
194
lam 1111111
..........
111
6.
11111 =1111.
am. III
MR*
111. db.
10
U
11111=1,,,
0 0 4... 1
I.
1111111111 1 1 r 1 i 1
FIGURE 75. The Transformations of Rectangular Shapes

These transformations are based on parallel projection rather than on polar (or perspective) projection.
getting progressively closer to a given method of systematically varying the
object. Whether it is seen as a change in stimulus, one can determine the cor-
distance (with size constant) or a change responding variations in the perception,
in the size of objects (with distance con- as psychologists have done for the vari-
stant) is indeterminate; only with a normal ations of color, sound, and the whole
environment does one or the other alterna- gamut of the qualities and intensities of
tive become realized. experience. In so doing, one can hope to
The main significance of such a chart uncover the nature of the specific relation
is that it makes possible psychophysical between the stimulus and the impression of
experiments on the variables of outline shape.2
shape in the frontal plane. Having a
2
This suggestion, like much of the chapter, shape on paper never yields a determinate per-
is vague and speculative. As this book goes to ception; it only yields a presumption of a real
press the writer has begun a set of experiments shape. Perhaps shape cannot be studied intel-
based on the hypothesis that a flat outline ligibly apart from the slant of its surface.
V IP 'r'' ,
1 :. , , h,
(CC
FIGURE 76. The Qualities of a Simple Line
The writer suggested some years ago direction and curvature of a short visual
(38) that a visual line or border has two line has one not specified the entire ex-
variable qualities, besides length. One is perience?
left slant...zero slant...right slant, and the The evidence for this unorthodox analy-
other is convex...straight...concave. A sis of border sensations came from two
line looks as if it had those phenomenal sources. First, it was possible to vary
properties and behaves in perception as if line segments in these two systematic
it had them. The two dimensions of varia- ways and observe the corresponding vari-
tion are as much sensory as are the hue ations in the qualities of direction and
and brightness of color. They could be tilt. A psychophysical correspondence
termed the quality of direction (linear could be established (38). The variables
slope) and curvature (linear shape). The could be isolated and the qualities could
names are inadequate, but the geometri- be discriminated. Second, it was a fact
cal variables to which they correspond are that tilt and curvature changed in the same
exact; the stimulus variables are the first way as do sensory qualities like white-
differential and the second differential of a gray-black or warm-neutral-cold during and
curve in analytical geometry. The first and after prolonged stimulation. When a line
second differential are sometimes explained was stared at, the tilt (or curvature) tended
as the slope of a curve and the sense in toward neutral, and thereafter an upright
which the slope is changing. Mathemati- (or a straight) line looked tilted (or curved)
cally, these two variables determine a the other way. In short, the familiar after-
curve at all its points. Phenomenally, the images of color were paralleled by after-
two corresponding qualities determine a images of tilt and curvature, slight, but
visual line or border in all its (convenient- definite in amount (36, 37, 38, 43). The
ly chosen) segments. If one specifies the after-image of color could be understood as
a shift in the psychophysical correspond- Moreover, the groups or outlines which

ence, and so could the after-image of tilt tend to occur depend on (1) the proximity
or curvature. and (2) the similarity of the elements to
If curvature and direction are the variable one another, on the tendency of lines (3) to
qualities of a border, there is a possibility make outlines (closure), and on the tenden-
that a closed border a form may be cy of outlines (4) to be smooth (good con-
reducible to variable qualities. The tinuation) and (5) to make simple forms
variables of a rectangular figure have al- (good shape). These factors were inter-
ready been illustrated. The changes of preted as laws governing the appearance in
curvature and direction around a contour perception of a figure on a ground. The
which determine its shape may become implicit assumption was that points and
enormously complex. They seem to be inlines have to be unified for a figure to ap-
tegrated or organized to yield qualities of a pear in perception. The factors could
higher order. The quality of closure it- therefore be taken to indicate forces of
self, one must admit, does not appear to neural organization operating among the
be a variable quality. It goes with "thing- elem ts.
ness" and suggests a theory of dynamic he laws of organization have already
organization which is difficult to analyze. een criticised on the grounds that they do
When a line becomes a contour, the varia -' of account foraccurate perception,
tions of shape appear to jump to a higher Koffica acknowledged the determining role
level. But there is no reason to assume of the retinal image only in the postulate
that shape cannot be reduced to its varia- of external forces of organization which
tions and that a form is unanalyzable. set limits on the internal forces of or-
The theory of visual form which this ap- ganization (67, p. 138). In addition to this
proach suggests is rather different from criticism it is also possible to doubt
that of contemporary Gestalt theory. The whether the particular arrangements of
latter is based on the celebrated laws of points and lines which Wertheimer devised
visual organization discovered by Wer isolated the fundamental types of visual
theimer (119), expanded by Koffka (67), and stimulation. The images of texture and
often generalized as the laws of all per- /contour need to be understood before the
----------
ception. These principles were based on images of point-groupings and drawn out-
drawings, chiefly of points, lines, and lines are studied. Wertheimer's drawings
curves, in different arrangements or re- were nonsense patterns of the extreme
gularly varied compositions. Elements type, far removed from the images of a
of this sort can vary in their spacing, material world. His laws are applicable,
curvature, orientation, and modes of inter- therefore, to some kinds of abstract draw-
section. The fact is that the points and ings and paintings, as Kepes in The
lines look like groups-or suggest complete Language of Vision recognizes (63), but
outlines in these artificial arrangements. not so much to ordinary visual stimulation.
,v-
aa
Meaning
Meaningful Perception .. . . How Much Percep-

tion is Learned? . . .. The Possibility of Spatial
Meanings . . Detachable Significance
Are There Unlearned Meanings? . . The Altera-
tion of Spatial Perceptions by Meanings . ... The
Literal Visual World and the Schematic Visual
World . . . . Conclusions
Let us consider the vision of one of our We can hardly say that he was conscious
very early ancestors some five or ten of geometrical space but we can be sure
million years ago on the plains of Asia. he discriminated the various distances of
lie was no longer living in trees and, al- an object. For example, one conceivable
though we know little about him, he would object to which he must have been sensi-
probably be classified as a member of the tive was a sabre-toothed tiger or some
genus porno. His eyes were probably just beast of equal ferocity. His conduct must
like ours and, if this is true, so were his have been rather nicely adjusted to dis-
retinal images. The shades, borders, and tance when he encountered one in open
gradients of light which composed these country, varying as the retinal image
images were specific to his environment varied in a precise way. To the tiger at a
and were the immediate cause of his see- mile he could react by going about his
ing the environment. lie was probably business. To the tiger at 400 yards he
a sharp-eyed creature for, as we know, he should have reacted by going in another
survived a rather risky existence. Since direction. To the tiger at 10 yards he
he probably had little or no language, he must have reacted (if he was one of our
had no names for things and we shall never ancestors) by running like the wind. His
know what his ideas or his conscious ex- behavior was graded in relation to a varia-
periences were. But we do know this. He tion of his retinal images.
discriminated among the variations of his Moreover his behavior was specific to
retinal images and could therefore react the contour, shape, size, color, and motion
differentially to the objects of his en- of objects. He did not confuse zebras
vironment. with tigers for, we may conjecture, he pur-
197
sued the one and fled from the other. The we must turn our attention to the question
differences in visual stimulation went with how he could know the tiger.
differences in his behavior, and we can
therefore be sure that he could identify
things. The animals that he could eat or Meaningful Perception
that could eat him may not have been Our own experience of the visual world
named but he could react appropriately to can be described as extended in distance
such light-reflecting objects, and they must and modelled in depth; as upright, motion-
have aroused at least a primitive kind of less as a whole, and unbounded; as color-
meaning. It was, in fact, on account of ed, textured, shadowed, and illuminated;
their meaning that he needed to discrimi- as filled with surfaces, edges, shapes, and
nate them. The color, shape, motion, and interspaces. But this description leaves
distance of things were of no interest to out the fact that the surfaces are familiar
him in themselves. These abstractions and the shapes are useful. No less than
were merely the identifying features, often our primitive ancestor, we apprehend their
slight and subtle, of objects which in- uses and dangers, their satisfying or annoy-
vited or compelled action. ing possibilities, and the consequences of
Judging from his probable behavior, action centering on them. Surfaces and
primitive man discriminated the solidity, shapes are in actuality perceived as ice,
separateness, and spacing of things with apples, fur, fences, clouds, shoes, people,
great accuracy. In his place, we would say and so on. Furthermore, our world is en-
that we saw a visual world. But he also larged and complicated as compared with
behaved toward things with circumspection, that of our ancestor, by the inclusion of
for he saw a world of meanings.. Specula- certain forms, lines, and man-made pat-
tive as all such accounts must be, it is terns which we know as pictures, symbols,
reasonably certain that our primitive an- and printed words. The visual world, in
cestor got about in his environment, and short, is meaningful as well as concrete:
knew one object from another. His be- it is significant as well as literal.
havior was based on locomotion and re- The psychology of meaning is a large
cognition: it was adjusted to space, and subject. One difficulty is that there are
at the same time it consisted of reactions so many levels or kinds of meaning. For
to objects. Vision provided him both example, there is first of all the possi-
guidance for his actions and cues for bility of a sort of primitive concrete mean-
his actions. Presumably, then, an object ing which either results from the infant's
like the sabre-toothed tiger was both active exploration of his physical environ-
localized and meaningful in his experience ment or is evidenced by such action. He
since he reacted to both its distance and fingers and manipulates things, and later
its significance. Heretofore we have been he gets about among obstacles and goes
mainly concerned with the question how he for moveable objects, and pulls and pushes
could localize the tiger. In this chapter and upsets things. Examples of such
ME ANI NG 199
concrete meanings for the adult would be and discovery. The world of symbolic
the way things look as if they were capable meanings stands at a far extreme from the
of being grasped or pushed or walked on. world of surfaces, edges, and shapes with
Second, there are all the simple use-mean- which this book is primarily concerned.
ings or meanings for the satisfaction of The former induces thinking, the latter
needs such as are embodied in food- only sitting, standing, walking, and grasp-
objects, play-objects, tool-objects, dan- ing. Nevertheless, they are both the same
gerous objects, and what Freud called world. Things must be substantial before
love-objects, the parents being the first they can be significant or symbolic. A
instances of the latter. For example, food man must find a place to sit before he can
looks eatable, shoes look wearable, and sit down to think.
fire looks hot . Third, there are the mean- The kinds of meaning listed are not ex-
ings of instruments, devices, constructions, haustive. There are also the unspoken
and machines. Fourth, there are the values meanings which go with visual motion such
or emotional meanings of things which make as the expansion, deformation, and trans-
the shapes of the world attractive or re- position described in Chapter 7, and these
pulsive in a vast variety of ways. Fifth, need special consideration. There are the
there is the kind of meaning exemplified in meanings of perceived events and se-
signs, by virtue of which one object or quences. There are meanings when one
event suggests another not physically surface touches another, or collides with
present. Clouds are said to be a sign of another, or when one object produces an
rain. The red light means stop. Signifi- action in another. There is also the whole
cance may also be defined broadly, if one range of social meanings, facial expres-
wishes to do so, to include the simpler sions, gestures, persons and actions be-
kinds of perceptual meaning referred to tween persons (92, 52). The visual world
above. Sixth, there is the particularly hu- is saturated with many kinds of meaning,
man kind of meaning embodied in symbols. and it seems to get fuller with meaning as
Such meanings are said to be abstract. we live from year to year.
Names mean things or persons, but their
principal advantage is that they can also
stand for classes, variations, and proper- How Much of Perception is Learned?
ties of these things. Carriers of symbolic We can be sure that the meaningful world
meaning like,money, flags, words the is fully achieved only by means of learning.
latter above all are in common use among The unsettled question is, how much of it
people who interact with one another; they can be achieved without learning? Are the
are completely determined by culture. primitive concrete meanings unlearned?
Symbolic meanings are the most complex Are the fundamental spatial impressions
and the most momentous of the list. They unlearned? Is there an embryonic un-
mediate knowledge, as distinguished from learned meaning for every perception? The
perception, and they are the basis for issue has remained alive ever since John
reasoning, creative imagination, invention, Locke asserted in 1690 that all knowledge
of the world is learned, the mind at birth are no unlearned meanings. And a fourth
being a complete blank. The parallel is that when meaning is added to things it
question, of course, is how much behavior does not substantially modify their con-
is achieved by learning and how much crete spatial qualities color, size,
arises spontaneously with the growth of shape, motion, and all the rest remaining
the organism. unaffected.
Having discarded the doctrine that men The evidence regarding learning and
do not learn their sensations of color but meaning in visual perception needs to be
learn their perceptions of everything else, gone over if we are to judge the adequacy
what shall we say to this question? The of these four generalizations. We may
simplest theory to fit all that has gone be- consider them in order.
fore might be to suppose that the visual
The Possibility of Spatial Meanings
world is an unlearned experience, that it is
meaningless when seen for the first time, The introspective observations of Chap-
and that what one learns is to see the ter 3 suggested that the characteristics of
meanings of things. the visual world could be described with-
As a general formula, this is consistent out reference to the practical meanings of
with a good part of the evidence. It is at the particular environment observed. But
least an improvement over the sensation- very possibly the practical meanings which
perception theory, and as an approximate make things look ordinary and useful are
or working hypothesis it is worth adopting. not the only kind. The world may have a
But, as will be evident, it makes a number residual and important sort of meaning
of simplifications which are at best im- even when viewed with the purest attitude
perfect. Moreover, it fails to fit with of contemplation. Even what was termed
another general formula which seems to the visual field the array of colors di-
be valid for all the studies that have been vorced from objective character may not
made of instincts, habits, and capacities be wholly meaningless if painters are to be
in men and animals, namely that no acti- believed. Colors, they say, have their own
vity is ever either wholly learned or wholly meanings, whatever the doctrine of sensa-
unmodified by learning. tion claimed, and an array of colors has a
The first oversimplification is the as- meaning appropriate to the array, even
sumption that the constituents of the visual when it does not compose a recognizable
world such as colors, surfaces, shapes, scene. There are distinguished painters
edges, and interspaces are in themselves who are thoroughly convinced that an array
meaningless. The second oversimplifica- which does not yield familiar objects on a
tion is the assumption that the meaning of level ground can have a more interesting
anything is detachable from its concrete meaning than one which does! Pictures
spatial qualities; that one can separate of this sort are not projections of existing
things and events from their meanings by or physically possible environments and
introspection. A third oversimplification are therefore non-representative. Abstract
is that all meaning is learned that there or non-objective paintings do not have use-
0,:rk4A9
c4'r
e k,'%',441'110/4,, '"
54.1 4,141, s,
4, 5'
i:r.44,4fk A? ,
n4"V4,,`",4,
FIGURE 77
Examples of
Abstract Paintin
Courtesy of The
Museum of Mode,
Art, New York
VIM
meaning, and their forms are not symbols as they have been differentiated from one
(as defined); but it is profitless to deny another; memorizing depends on the forma-
that they have any meaning, for the argu- tion of a unique or identifying response to
ment reduces to a conventional definition each nonsense item, and this is the next
of the word "meaning." The definition of 'thing to the formation of a meaning for each
this term has never been agreed upon. item (33, 34). The experiments on drawing
For linguistic meaning, it is true, nonsense forms from memory, summarized
systematic definitions of signs and sym- by Woodworth (121, Chapter 4), point in
bols can be worked out on the basis of the same direction: a senseless form must
logical and psychological theory. This gain sense in order to be recalled. Re-
has recently been done by Morris (84). search on learning to recognize aircraft
But visual meaning has so far defied (39, Chapter 7) strongly suggests that
systematic analysis and the whole subject, when a nonsense form becomes identifiable
including art-criticism, is notoriously it also becomes meaningful. The common-
speculative. sense or toughminded view, therefore, is
The common-sense opinion and the view not strictly in accordance with the facts.
of tough-minded psychologists is that Non-objective forms lack namable or social-
colors and shapes which do not produce in ly agreed-upon meaning, but it is not true
perception remembered objects or things to say that they lack all meaning.
which the onlooker knows what to do about A visual scene which has modelling,
should be termed meaningless. For the color, texture, surfaces, shapes, and inter-
psychologist, meaning originates in adap- spaces is ordinarily caused by a physical
tive response. Drawings not resembling environment, and constitutes what we have
anything familiar are called nonsense called a visual world. Almost invariably
forms and, along with nonsense syllables, it has a fundamental surface with the
are employed in experiments on memorizing. meaning of a terrain or a floor, since. human
They are also shown to observers under animals live as they do. On rare occasions,
conditions which impoverish the retinal however, men peer into microscopes or
image, for example a very brief exposure, look under the sea through diver's hel-
in experiments on the accuracy of percep- mets, they stare at flames and clouds, and
tion. Nonsense forms are much more dif- they sometimes gaze at abstract paintings.
ficult to reproduce or discriminate than No terrain or floor is induced in these per-
familiar or conventional forms. The fact ceptions. Here, the modelling, color, tex-
is, however, that the results of these ex- ture, and the rest of the variables of the
periments suggest that nonsense forms are visual world do not produce a visual world
only relatively meaningless. Meanings of the everyday sort. Take the abstract
which are peculiar to the individual ob- paintings, for instance. The surfaces and
server are always repolted and are, in fact, shapes induced by them do not, of course,
what make the forms memorable. According correspond to any physically existing sur-
to E. Gibson, nonsense syllables or non- faces and shapes. The personal meanings
.sense forins can be memorized only insofar they arouse could never be verified, and,
MEANING 203
consequently, they are not practical. The Detachable Significance

surfaces and shapes seen by the diver and Titchener's context theory implied that
the biologist, in contrast, are real, but the meaning accrues to impressions, that is,
environment is not a human environment. is added to them. Meaning is attached by
The world in the microscope is novel; the association. A bell acquires the meaning
space of the cloud-patterns is useless; the of dinner, or someone at the door, or a
things in the painting are imaginary; but streetcar at one's back, depending on the
they are composed of the same stimulus context. Stimuli come to arouse specific
variables which enable us to get around in expectations, in more recent, terminology,
the environment we inhabit. A contem- and the perceptions thereby acquire signi-
porary artist and teacher, Kepes, has ficance.
called these variables "the language of Qualities of touch, temperature, and
vision," referring to the painter's use of muscular feeling can unquestionably be
them as abstract constituents of percep- added to a purely visual perception by ex-
tion (63). They have meaning, and the experience in manipulating and using the
perimental combination of them creates object in question. Presumably this is
new meanings, but they are essentially why fur looks soft, ice looks cold, and a
spatial and it is pretentious to suppose book looks openable instead of solid and
that they enable the painter to represent box-like. Manipulative meanings are often
some realm of non-physical reality. The added to visual perceptions during adult
claims of some abstract painters that by life. The change can then be noticed in-
the use of these plastic devices they
. can trospectively as, for instance, in the case
represent high and hidden truth, however, of a new kind of hand-operated tool en-
should not force us to the opposite ex- countered for the first time. After its use
treme of declaring that surfaces, edges, is understood the object looks different.
and shapes with color are without meaning, The perception now has properties it did
for this would be false. They have at not have before and here is the point
least the meaning of surfaces, edges, and the properties are not directly aroused by
shapes. present retinal stimulation. The associa-
tion theory ascribed them to memory-images
The implication seems to be that spatial clustering around the visual shape, and the
impressions can be largely but not wholly latter is said to redintegrate one's past ex-
meaningless. If spatial meanings are in- perience with the object. It is only logical
separable from visual objects, the first to suppose that the coldness of a piece of
generalization of the formula being tested ice or the squeezableness of a pair of
is too simple. Perhaps these meanings pliers is not given by stimulation and must
are to some extent unlearned. Conceivably, be given by memory. But these qualities
then, only the learned meanings are sepa- are not like the memory images of recollec-
rate from spatial qualities and are what tion. Meaning is not literally recall.
get attached to the visual objects. What is Squeezableness is something which seems
the evidence for this possibility? to be located in the object, not in the hand,
and it refers to the present, not to the taneously, and even when it is gone the
past. Visual objects appear to have observer usually has a feeling that he could
soaked up such qualities and to be fairly get it back if necessary.
saturated with them, the use of the object My own observation indicates and it
and the shape of the object being almost is confirmed by experiments performed by
indistinguishable. Nevertheless they are my students that this recession of mean-
distinguishable to introspection and they ing can be noticed with prolonged attention
are separable when the use is learned. to objects other than words. If you stare
An even clearer instance of attachable at a paper match long enough, its familiar-
meaning is furnished by symbols. Printed ity disappears, although it does not break
letters and words are merely shapes to the up into sub-units as a word does. The
illiterate, to foreigners, and to the very same thing happens when a piece of sand-
young. (The sound patterns and the vocal paper is rubbed monotonously with the
patterns of words were also meaningless in fingers. Perhaps any scene begins to ap-
the beginning.) To most of us, on the con- pear strange when the eyes are fixated long
trary, the words we read are merely mean- enough. The attempt to observe one's
ings and we are scarcely aware of their visual field leads in this direction as does,
shapes. If the reader will stare fixedly at in a certain sense, the painter's intent
a word like "abyss" however, he will no- view of something which interests him.
There is a third kind of evidence for the
separability of meaning and spatial im-
abyss evil pressions detachable meanings and
this is derived from studying the percep-
tions of Patients with brain injuries. With
tice almost immediately that it begins to injury to the occipital lobe of one of the
look like a mere shape. A familiar word hemispheres there often results a peculiar
like "evil" usually keeps its meaning visual incapacity without apparent in-
longer when fixated, but after an interval capacity of other kinds. This defect is
the visual appearance becomes prominent, sometimes called psychic blindness. The
the meaning becomes separated of some- retinal surfaces are projected by neurones
where, and the word disintegrates into on the surfaces of the brain in such a way
e - vil or e- VI-1. Eventually even the let- that the right half of the visual field (the
ters may begin to look unfamiliar and the left half of each retina) is represented on
word can become completely geometrized. the cortex of the left occipital lobe.
If you repeat the word to yourself rapidly, Damage to the nerve-cells in this area pro-
the sound also becomes meaningless and duces a sort of a half-blindness called
the pronunciation may tend to disintegrate. hemianopsia. The patient is unable to
The phenomenon has been called loss or make any visual discriminations to the
lapse of meaning (7, 27, 96), but it is bet- right of his line of sight, as is proved by
ter to say that the meaning recedes from the mapping his visual field with the eyes
word, since it frequently comes back spon- fixated on the midpoint of a screen. The
MEANING 205
visual field has been constricted to half of mean. The forms as such are discriminable
the ordinary oval.1 but they are no longer phonetic symbols.
Simple right hemianopsia in right-handed Along with word blindness there usually
persons is often accompanied by word goes some degree of what has been called
blindness or alexia. The patient can, by psychic blindness or visual agnosia. This
ordinary tests, see things fairly well, but is a failure to recognize or know the names
print has become wholly or partly meaning- of objects, persons, pictures, and places.
less. The letters are visible enough, but The evidence suggests that the spatial
they are no longer words. In one test, for aspects of things distance, size, shape,
instance, he may be able to take a set of location, movement are relatively un-
children's alphabetical blocks and turn disturbed, but that their meaning is re-
each block to make the letter right side up duced. Verbal and conceptual meanings
without being able to name a single letter are affected most, concrete everyday
or arrange the blocks in alphabetical order, meanings least, and the most primitive
let alone read print (117, page 288). Or meanings seem to be retained.
such a patient may be able to trace letters On the whole, the conclusion is that
with his fingers without knowing what they meaning becomes less intrinsic to and
more detachable from spatial impressions
the more it approaches high order concepts.
1
A significant fact, incidentally, is that his At least this seems to be true as meaning
visual world is not halved how could it be? becomes verbalized. Spatial meanings are
The patient usually complains only that he
cannot see things as definitely as he used to. tied into their perceptions, relatively
The visual world is not a point-to-point pro- speaking. Verbal meanings and civilized
jection of the retinas. The world can still
be integrated out of successive fixations of use-meanings are relatively detachable.
half-fields. The old theory would say that
perception fills in the blind area. The point Are There Unlearned Meanings?
is, however, that a localized area of injury
in the visual brain produces a localized area The fact that men differ in the meaning
of blindness in the visual field but not a which things have for them is a truism. Men
disappearance of objects in a localized area
of the phenomenal environment. The brain of different training, interests, and convic-
maps the momentary retinal images of the tions do not, as we say, see the same
world but it does not map the world we see.
The brain and the retina are in spatial cor- world. A human body is perceived by an
respondence with one another, whereas the anatomist differently from the way in which
brain and the phenomenal world can only be in it is by the rest of us. A nickel is not the
a spatio-temporal correspondence that is, a
correspondence other than one of simple point- same thing to an adult as it is to a child.
to-point projection. This means that one can- An industrial machine is not the same thing
not expect to find any simple localization in
the brain of what is ordinarily called space to its operator as it is to the plant owner.
perception. Consequently it is only natural, The specialized perceptions of the con-
not paradoxical as neurologists assume (8),
that a localized brain lesion should cause the noisseur, the photographer, the doctor, the
loss of a piece of the field of view (a blind woodsman, and the engineer are all differ-
spot) and at the same time the loss of a
feature of the perception of the world (for ent. Moreover the apprehension of the en-
example, distorted size or defective meaning). vironment differs in a systematic way
among peoples as well as among indivis that exceptions to it should be carefully

duals. The traditions and culture of an scrutinized.
uncivilized tribe consist not merely in Nevertheless there is reason to believe
strange ways of behaving, such as used to that some meanings, or some components
be described, but also in strange ways of of meaning, are not learned. When
perceiving and apprehending things. Ac- McDougall applied instincts to human
counts by anthropologists of these foreign psychology at the beginning of this cen-
modes of apprehension, in fact, begin to tury he included in his definition of an in-
make the customs intelligible to us. The stinct "an innate disposition to perceive
Trobrianders, it has been reported, re- and pay attention to objects of a certain
cognize that a child resembles its father class" (80). He meant that there were
but never its mother. The former kind of intrinsically fearful objects, intrinsically
resemblance is proper and to be expected; good-to-eat objects, objects which in-
the latter is inadmissible. The percep- trinsically invited acquisition, mating,
tions of similarity are part and parcel of a curiosity, self-assertion, and all the rest
set of customs regarding kinship. Kline- of the list. The emotional excitement, the
berg suggests that the effect is a failure impulse to action, and the purposive striv-
to note any resemblance rather than an ing of an instinct were all contingent upon
actual interference with the sensory per- an innate perceptual inlet. The instinct
ception (65), but this is hard to be certain theory led to absurdities in social psychol-
about. In any event, it is a perceptual ogy and encouraged a tendency to name
custom which is foreign to our ways of various aspects of human behavior instead
seeing people. of describing them. Instincts fell into
disrepute among those interested in the
All these facts seem to point to the growing science of human learning. Sti-
general conclusion that each man learns muli, reflexes, and chained reflexes were
the meaning of the world for himself, with- more exact terms with which the experi-
in the framework of his upbringing and the menter could work. But the neutral term
society in which he lives. Significance and stimulus is not adequate to explain why
value are formed partly by the cultural behavior is a function of objects, and the
background and partly by the individual's theory of patterned stimulation or Gestalten
unique experience, but in any event are arose to reintroduce the notion of intrinsic
learned. This is the conclusion of what meaning in a new fashion. McDougall was
has been called empiricism. It is consis- surely wrong about intrinsically fearful or
tent with the emphasis on learning and eatable objects in perception, but perhaps
education and with the scientific approach his successors have overstated the case
to human nature and culture in the history for learning.
of modern thought. The conviction that Among animals, certainly, there is good
all meanings are learned implies that human evidence for innately meaningful percep-
beings are plahtic rather than cast in rigid tions. Lashley has stated the case con-
molds. It is such a valuable hypothesis vincingly in the following way (71). If
MEANING 207
you put the question, "What is an egg to a field when they first smile while looking at
nesting bird?" a perfectly sensible answer a doting adult, it would suggest an innate
can be given. The answer is, "any object meaning for the pattern of a human face
which causes the bird to retrieve it, tc an instinctive response nicely adapted to
clean its surface, and to sit on it." Such induce or maintain the doting attitude of
an object is discriminated or recognized the adult. To be sure of this one would
as an egg, or we might say, has the mean- have to prove that the supposedly in-
ing of an egg. The experimental fact is stinctive smile was not a random but a
that such behavior in nesting birds is visual response; that it always occurred
aroused only by rounded objects within to a pattern similar to a face; and that it
definite limits of roundedness and certain did not occur to such patterns as a milk
limits of size, and with a texture approach- bottle, a lampshade, or a square of card-
ing a specific sort. Objects with these board. It is not asserted here that babies
qualities are eggs; objects without them, instinctively recognize their mothers; it
when put in the nest, are not eggs in the would be enough to show that they have an
sense that they are pushed out of the nest. unlearned response which would make their
This sort of behavior appears during the mothers think they do.
first nesting season without opportunity to Just this seems to have been demon-
learn. Fine discriminations are not made strated recently by Spitz and Wolfe (99).
between eggs and non-eggs but gross dis- Babies between 2 and 6 months of age,
criminations are. Lashley points out that they report, almost universally fixate and
discriminative behavior of this sort is as smile at any face-like object if it moves.
frequent as are instincts, for the two go Any face with any expression will serve,
together. or a crude mask, or a dummy, as illus-
The human animal matures slowly and trated in Figure 78. A vicious leer is as
learns more than the other species. He good as a benevolent smile; the baby is
can eventually make very fine discrimina- not imitating, or even discriminating,
tions among birds' eggs for example facial expressions. It must be the pattern
and these are learned. In the use of lan- of a full face; a profile will not do. Too
guage he has acquired a special method for small a face will not do, such as that of a
learning and retaining fine discriminations. doll. A completely motionless face will
But apart from language and apart from not do, nor will a milk bottle, or a toy, or
differentiated adult meanings is it possible other objects. These are fixated but not
that infants have wordless and crude smiled at. After 6 months the baby's smile
primitive meanings for some of their becomes more discriminating. He no long-
earliest visual impressions? er smiles at a stranger or a dummy; he
The facts are hard to interpret and not recognizes his parents; he reacts to an ex-
conclusive. Most observations of babies pression of disapproval; he begins to smile
are biased. The smiling response is a when smiled at. All these responses can
good instance. If we could be sure that be learned. But the original smile is ap-
babies identify a human face in the visual parently not. The effective stimulus for
s!:!,!!! !9;;!!4;4,0,,s, ,r,0!,!!!!4!?1!'' ',,!! , N!!!,0
Pet //4%
omplew, rfi
fY
00.11.'0
FIGURE 78. Patterns which Evoke the Infant's Smile

Courtesy of Dr. Rene A. Spitz. From "The
Smiling Response", by Spitz and IT olfe
(Generic Psych. lionog., 6 °I. 34, 1946)
it is an indeterminate pattern with only the ception of space is unlearned, meaningless,

elementary characteristics of a human face. and acquires its meaning in the course of
The wordless meaning of this crude pat- experience makes an additional simpli-
tern is equally primitive, but it neverthe- fication which is far from perfect. It sup-
less is a discriminable pattern and it has poses that when a certain shape, let us
the meaning of something-to-be-smiled-at. say, gains a new meaning, the shape re-
These facts, if they are correct, imply mains just what it was before. Fhe tex-
that the human infant does not begin to ture, slant, color, contour, and other con-
learn meanings at a zero level. They show stituents of a thing are supposed to be
the falsity of the notion that we are born unaltered by the gain in meaning. It may
with a set of meanings ready-made or a be apprehended differently but it is sense(
set of innate ideas, but at the same time the same as before, to use the older ter-
they contradict the notion that all meaning minology.
is acquired. That generalization is too There are few generalizations in psycholb
easy. There is probably an embryonic ogy which have been refuted as often a5
meaning which goes with an embryonic this one. It can easily be disproved; wha
visual perception. is harder to understand is why neverthe.
less it must have some amended validity,
The Alteration of Spatial Perceptions by Consider first the evidence for its untruth
Meanings The color of anything, it might seem
The formula being tested that the per- should be unaffected by the mere fact tha
MEANING 209
it resembles something of another color. This latter result is the dramatic feature
Duncker, however, has proved that under of the experiment the suggestion that
certain conditions a piece of cloth in the need of money and infrequent possession of
shape of a leaf is judged as noticeably it do more than make the child stand in
greener than an equivalent piece of cloth awe of a coin; that indeed they modify the
in the shape of a donkey, when the light very spatial structure of 'the child's world,
reflected from both was only faintly green- and that the substance and dimensions of
ish. There is an effect of what is called the environment for a poor child are dif-
memory-color, that is, the color associated ferent from what they are for a rich child.
with leaves as contrasted with the color Economic class may affect even the sensa-
associated with donkeys (29). He also tions of things. It is worth noting, however,
found that a piece of brown chocolate had that when Carter and Schooler repeated the
a stronger chocolate taste than a piece of experiment (20) only remembered or im-
white chocolate when the taster could see agined coins were exaggerated in size;
them, but not when he was blindfolded. actual coins were perceived by children,
This result failed, however, if the subject rich or poor, in close relationship to their
came to suspect the purpose of the ex- real size.
periment. The outline of a thing ought not to suffer
The size of a thing, likewise, ought not alteration just because one knows what it
to be affected by its meaning. Bruner and is good for, nor should a geometrical form
Goodman, however, found that ten-year- change because it looks like a familiar
olds perceived coins as larger than equi- object. Nevertheless, outline and form
valent cardboard disks by about 25 per are modified by meaning, and the fact has
cent (14). Presumably the value of the been demonstrated over and over again.
coin influenced its size. There was a The writer once found that the nonsense
striking tendency for poor children to see form on the left in Figure 79 might be
the coins larger than did rich children. interpreted by one observer as a woman's
THIS MAY BE OR THIS

THIS OR THIS
FIGURE DRAWN AS
FIGURE 79. The Effect of Meaning on Visual Form
torso, by another as a dumbbell, and by a things, of valued things, of attractive or

third as a violin (35). The drawings of the repulsive things, and of customary things
form made by each observer are shown on or things that other people also see. Per-
the right. It might be guessed that the ception, in Bartlett's term, tends to be
three observers had somewhat different schematic.
interests in life. It should be noted that The study of this tendency in percep-
the nonsense form had been shown along is valuable, for it illuminates a set of
with others and was reproduced after an problems of the very greatest importance.
interval of time. How do social stereotypes arise those
The same kind of result has been obtain- schematic and usually distorted percep-
ed by many experimenters (121, Chapter 4). tions of race, nation, religion, and class?
The errors made by observers in reproduc- Such perceptions tend to be caricatures
ing visual forms constitute a fascinating with respect to physical qualities as well
subject. Among many hypothetical ways as being simplified with respect to mean-
of accounting for these errors (habits, ing. Why are persons judged as types
conventions, dynamical organizations, rather than as persons? Why do rumors and
norms, standards, concepts, the law of myths persist? How do words exercise
Pragnanz, or simply the association of their tyranny over thinking?
ideas), perhaps the most comprehensive is
that of Bartlett, who suggests that a pat-
tern tends to be schematized during per- The Literal Visual World and the Schematic
ceiving and remembering (5). Bartlett him- Visual World
self had investigated story-telling from It is easy, however, to misinterpret all
memory as well as form-perception, and he the evidence for a schematic trend in per-
discovered, as had earlier students of the ception. Itis tempting to conclude that
psychology of testimony, that a story gets all apprehension is selective and dis-
retold so as to express the meaning it torted; that perception is inevitably a
has for the teller. It is altered in ac- constructive process which creates the
cordance with a schema which is charac- world to suit the perceiver; that we see
teristic of the individual, his interests, and things not as they are but as we are. Any
his culture. such general conclusion is unwarranted,
Color, size, form, sequence, and still for it neglects the existence of what we
other qualities of perception may un- shall call in this book literal perception.
questionably be affected by the past ex- Perception can be studied scientifically
perience and attitudes of the observer. by either of two general methods. The first
William James once remarked that per- is the psychophysical method of the labora-
ception was of probable things. The ex- tory in which it is expected that the sub-
periments suggest that perception is also ject will make the best observations of
of familiar things, of expected things, of which he is capable, and conditions ari
known, typical, average, or normal things, arranged to facilitate them. This methoc
of namable things, of specific or precise yields literal perception. The average
MEANING 211
errors for color, size, and form are of to color, size, shape, and sequence than is
known small amounts. The second is the necessary, since literal perception takes
method of impoverished, ambiguous, or time and effort. The percept is reduced to
equivocal stimulation in which the experi- a cue for action. But perception can be-
menter arranges conditions so that errors come literal whenever the observer needs
will occur (40). The experimenter uses to discriminate. Under favorable condi-
dim light, or a flash exposure, or presents tions it can be surprisingly exact, as the
more items than are perceptible at once, or experiments of the laboratory demonstrate.
makes his test after an interval of time. One can always look at a thing carefully
He may present undifferentiated or non- if there is reason to do so.
sense configurations, such as ink blots, Perception is not always or necessarily
which have unusual shapes and suggest distorted by needs or affected by purposes
many different interpretations. Or he (18). It is not fated to be stereotyped or
presents reversible figures or patterns with assimilated to social norms (97). Mis-
equivocal contours which have more than perception is not a consequence of sensory
one meaning. This method yields schema- organization but of the inattention of the
tic perception. Impoverished, ambiguous, perceiver or the weakness of physical
or equivocal stimulation is necessary be- stimulation. It is perfectly true that per-
cause the observer in a psychological ception can be fluid, subjective, creative,
experiment, if he is an adult, is usually and inexact, but it can also be literal. It
wary. He will look with attention and per- can be literal with respect to fine dif-
ceive literally when conditions permit. ferences and complex qualities, as the
The fact is that the evidence for the wine-taster, the artist and the scientific
schematic trend in perception has all been observer prove. The student of human na-
obtained either by this method of presenta- ture and society needs to remember this
tion or with relaxed attention on the part of when he is in danger of assuming that men
the observer. The alterations or distor- are the passive victims of their stereo-
tions might have been eliminated if the types and perceptual customs. The detec-
conditions for observation had been dif- tion of witches by the citizens of Salem,
feren t. Massachusetts, is a case of gross mis-
It is true, to be sure, that the perceiving perception, but it does not always happen.
of everyday life is often a matter of glances The world of visual experience which
and faint or ill-remembered impressions, this book is about might now be defined as
and the results of impoverished or am- the literal visual world the world of
biguous presentations are therefore truer qualities as they appear to attentive ob-
to life than the results of optimal presenta- servation. It is a world of color and space
tions. The perception of everyday life is in combination; it is not the world of color-
very often schematic. In common speech, sensations, since that is a myth of seven-
a man tends to "see things in his own teenth century psychology. It resembles
way." In the course of practical behavior, the world of everyday experience more than
perception is no more literal with respect traditional sensations ever did. But, on
the other hand, it is not simply the world mulation contributes to knowledge before
of everyday behavior either, since that is they can discover where error creeps in.
a schematic world of cues and signs from In the study of social psychology (and the
which many of the qualities of color and other social sciences which rest upon it),
space have dropped away. Only in an un- there is an enthusiastic interest in per-
familiar environment or a problem situation ceptual customs, the social norms which
do we become fully aware of the literal make a man's world what it seems. Study
visual world. One has to pause and look the values of a culture, the formula has it,
in order to see it. and you will understand why its members
The world of visual experience with behave as they do. These enthusiasts
which this single chapter deals is the need to be reminded that all human beings,
schematic visual world. Its richness and everywhere, probably see the ground and
complexity cannot be described in one the sky in the same way. They need to
chapter or even in one volume. In a know the basic perceptual capacities of
sense the real study of perception begins the human species before they can hope to
with it. Schematic perception, however, is describe the private worlds of persons,
an even more intricate human act than classes, races, or nations. It is important
literal perception, and we cannot hope to to understand that a poor child remembers
discuss it sensibly without coming first to a fifty-cent piece- as much bigger than it
an agreement about fundamentals.2 really is. But this fact can never be corn,
In the thinking of many artists, for ex- prehended without understanding how size
ample, there exists a confusion between is judged as a feature of literal space.
the seeing of space and the seeing of
symbols. They are not clear where the
difference lies between representing and Conclusions
symbolizing. They need to agree upon the 1. The classical formula of empiricism,
first problem before they can undertake to that two-dimensional color-sensations are
deal with the second. Philosophers and innate while all other perception depends
logicians have been concerned for cen- on learning, fails completely to meet the
turies over how to define true knowledge. facts.
They need to know what part sensory sti- 2. The formula that space is innate but
that meaning is learned meets more of the
facts but it, too, is inadequate.
2For 3. Meanings and spatial properties are
a summary of the evidence about
schematic perception and a consistent presen- not entirely separable from one another;
tation of the theory that perception depends
mainly on constructive processes within the meaning is not wholly detachable from
organism rather than on stimulation, the color, form, and texture. Symbolic mean-
reader is referred to Vernon's Visual Percep- ings, however, seem to be detachable from
tion (115). Vernon's entire emphasis is on
the degree to which percepts are not in cor- their objects and are presumably learned.
respondence with stimulation and are not 4. There is some evidence that, in
referable to physical objects outside the ob-
server. animals and infants, embryonic meanings
ME ANI NG 213
do not have to be learned and consequently optimal, approach an exact correspondence

that all kinds of meaning are not learned. with variables of physical stimulation.
5. There is overwhelming evidence to 6. The correspondence of perception to
show that meanings react upon their per- stimulation does not have to be a wholly
ceptions to select or modify the spatial innate correspondence. A psychophysical
properties (color, size, outline) and that relation may be shown to exist without our
these properties therefore depend upon the having to decide how it came to exist.
personality and the culture of the perceiver. Conceivably every such relation is partly
This evidence applies, however, to innate and partly acquired. Even the rela-
schematic perception, not to literal perception of color to wave length may be refined
tion. The properties of the literal visual and made more exact by past experience
visual world, insofar as conditions are with colors.
Learning
In What Sense is Behavior Mediated by Percep-

tion? .. .. In What Sense Do We Learn to See?
A generation ago psychologists were true that one learns to perceive? It might
debating whether their principal concern be the case that perception is a pre-
was consciousness or behavior, and they requisite to learning in one sense and a
called themselves introspectionists or result of learning in quite another sense.
behaviorists according to which side they The first question asks what the rela-
favored. Today that dispute does not seem tion is between perception and perfor-
as important as it once did. Of much mance. The investigators of learning in
greater importance is an understanding of animals and children have tended to lo-
the process of learning in animals and men. cate the learning process in the perfor-
A similar issue in a new form arises, how- mance of their subjects. But the Gestalt
ever, which might be put in this way: is theorists and others haves tended to locate
learned behavior mediated by perception or the learning process in perception. The
is perception only an incidental accompani- former emphasize that learning is an altera-
ment of learned behavior? In other words tion of behavior, the latter that learning is
do we adjust to the world because we see comprehension, cognition, expectation, or
it or is our seeing of the world the result insight (55). The argument of the former
of our adjusting to it? Is learning a matter is that behavior is what counts, and they
of insight or does insight follow upon are content merely to specify the "stimuli"
learning? This issue is not merely a ver- or the "cues" which evoke behavior. The
bal dispute, for differing opinions yield argument of the latter is that the perceived
quite different experiments. Neither is it environment of an organism, the "field,"
trivial, for it involves a choice of the determines its behavior, and that learning
direction in which a science shall move. is best understood as a "reorganization"
How is perception related to learning? of the field. Learning is conceived to be
Two separate questions are implicit the understanding of the values of things,
here which a theory of perception should of "what leads to what," of where things
try to answer. The first is, in what sense are, and of what to expect from a given
is learned behavior mediated by percep- event. Learning is thought to be percep-
tion? The second is, in what sense is it tual and spatial or, in Tolman's words, to
214
LEARNING 215
consist of "cognitive maps" (110). The ing and behavior. There is a great differ-
real issue between these theorists reduces ence between these two. The stimulus in
to whether it is enough to say that stimuli a psychophysical experiment is only one
evoke behavior or whether it is necessary instance of a systematically varied set of
to suppose that perception evokes be- objects or events an instance selected
havior. Perhaps the issue can be re- from a series of experimental variations of
solved. their physical properties made by an ex-
The second question asks what the perimenter. The stimulus in a learning
relation is between perception and learn- experiment is often taken to be simply an
ing. Granting that our perception of the object or event a junction of routes in a
world is not simply constructed out of un- maze, a bell, a flash of light, a black box,
learned sensations plus images, in what or a printed word. Strictly speaking, these
sense do we learn to perceive space and latter are not stimuli at all. They are
meaning? The simple formula discussed usually called stimulus-objects or sti-
in the previous chapter proved inadequate, mulus-situations. It was maintained in
and the question remains unanswered. Chapter 5 that these terms cloak our ig-
norance. Behavior is, of course, a speci-
In What Sense is Behavior Mediated by
fic function of objects and situations, but
Perception?
we need to know why this is so. The sti-
The term "stimulus" is used very mulus-energies delivered to the receptors
loosely in the biological sciences, and have been assumed in the past to be poor
the absence of agreement upon a definition representations of the objects and situa-
has been at the root of much fruitless dis- tions to which they correspond.
cussion.1 In psychology the term is cur- The classical theory of perception as a
rently used in two different ways. It may special process of inner learning was in-
mean a variation of energy to which there vented to account for the discrepancy be-
corresponds a variation of either an ex- tween the stimulus-energies and the per-
perience or a response, or it may mean the ceived objects. So was the more recent
occasion for a namable experience or the theory of sensory organization propounded
occasion for an identifiable act. The first by the Gestalt psychologists. When a
usage, the more precise one, is employed behaviorist, therefore, maintains that he
in psychophysical experiments; the second need not concern himself with a rat's per-
usage is employed in experiments on learn- ceptions, when he denies that behavior is
necessarily mediated by perception, and
assumes that his rat responds to the alley
1
The definition given in Chapter 5 is worth of a maze as a "stimulus-object," he is
repeating here: a type of variable physical glossing over a difficulty and concealing a
energy, falling within certain limiting thresh-
olds, which excites a receptor-cell differential- problem. He may use the term "cue" in-
ly; also the differential excitation of different stead, as Miller and Dollard have recently
receptors with respect to the adjacent and done in explaining Hull's theory of learn-
successive order of differences over the array
of cells. ing (83), but this expedient begs the ques-
tion. How do stimulus-energies become process is mistaken if it implies that the

cues? only way to be sure an animal has learned
The situation is different if one assumes is to intuit the animal's perception. It is
that the stimulus. energies delivered to the useful to suppose that a rat has insights
receptors are precise mathematical trans- and hypotheses and expectations, that he
formations of the objectsand situations to makes simple inferences, and that he re-
which they correspond. Inner perceptual members a map of a maze, but there is no
inference and sensory organization become compulsion to suppose this. Sign-learning
gratuitous. Behavior is a specific function may be described as the accrual of meaning
of ordinal stimulation. The definition of to a percept, and it may also be described
"stimulus" must be revised, it is true, as an altered tendency to react to the sti-
and the convenient notion that a plurality mulus-correlate of an object. Which for-
of discharging receptor-cells represents a mula is more revealing we do not yet know.
plurality of stimuli must be wholly rejected.
But it becomes possible to refer to the
stimulus-correlate of an object and to In What Sense Do We Learn to See?
understand how a response can be a con- The 19th century controversy over
stant function of an object. nativism and empiricism in perception,
The kind of perception with which this discussed in Chapter 2, proved to be in-
book deals spatial perception or literal decisive. The Gestalt theory of sensory
perception is not taken to be a special organization did not clarify the issue, for
mental process. It is not something in- its critics called it a kind of nativism, but
termediate between stimulation and re- its adherents denied that it was. The
sponse. Both perception and behavior may formula advanced in the preceding chapter
be activities of the organism specific to was inadequate, inasmuch as the percep-
ordinal stimulation. Behavior, then, is tion of literal space is not wholly unlearned
evoked or mediated by perception only in and the perception of schematic meaning is
the special sense in which perception is not wholly learned. The simple solutions
the study of ordinal stimulation. The so- fail. Nevertheless there must be some
called stimulus-objects for behavior are respects in which perceiving develops
the stimulus-correlates of the literal spontaneously and other respects in which
visual world. The so-called cue's for be- it depends upon practice, experience, or
havior are certain invariants of stimulation training. The difficulty is to state in what
which yield objects with color-constancy, sense we learn to see.
shape - constancy, and size-con stancy.
Physical objects must be specific in The First Perceptions of Cataract Pa-
stimulation if they are to be specifically tients Blind from Birth. The method of
responded to, but it does not follow that determining whether or not a function is
they must necessarily be known if they are learned is to deprive a young individual
to be specifically responded to. of all opportunity to exercise the function,
The theory that learning is a perceptual at the same time providing him with all the
LEARNING 217
conditions of life and growth which might descriptions of what they could do, and
enable the function to mature. The func- their responses to various simple tests.
tion of vision is much too valuable in the The foremost fact is that the patients were
human subject to allow such experimenta- bewildered and confused by the new visual
tion, but the circumstances required for impressions, especially by the continuous,
the experiment sometimes occur naturally. unrelenting flow of these experiences and
A few individuals develop cataracts in by their enormous variety. It must be re-
both eyes at, or soon after, birth. In this membered that, in common-sense terms,
disease the lens becomes semi-opaque, so they did not know what anything looked
that the individual is, for all practical pur- like. They knew only what things felt
poses, blind. Some diffused light can like and only those things which could
enter the eye, and such patients can be touched. A subject would complain
usually distinguish night from day, but that there were "so many new things he
their impression of the world is at best could not comprehend." On a city street,
no more than a vague gray fog, brighter at "so many different things, and the rapid
some times and darker at others, or movement of the mass, confused his sight
possibly brighter in one region and darker to such an extent that finally he could no
in another when facing strongly contrasting longer see anything, the latest thing hav-
surfaces. When the patient is a child or ing not yet faded when the next impressed
an adult one or both cataracts can be itself" (94, page 148). The patients all
surgically removed, and after this happens had the use of language but they found it
the eye can form a differentiated image on difficult or impossible to describe what
the retina (although glasses must be worn they saw or to apply words to it. The
to bring the image into perfect focus). question, "Are things projected in space?"
After the bandages are removed, the simply did not mean anything to them.
operated patient is stimulated for the The use of visual perception seemed to
first time by a projection of his environ- require an extended period of development
ment. The question is, what does he see? weeks, months, or even longer. Too
The study of these first perceptions has much seems to have been expected of the
interested scientists and physicians for patients by relatives and friends, who
many years. Sixty-six cases have been could not understand why they were not
collected by von Senden and the facts now normal persons, and the task of learn-
brought together in one volume (94). Al- ing to see sometimes overwhelmed them
though the tests of perception were im- with its difficulty. Some, discouraged and
perfect, and although the reports were ob- depressed, quit trying to see. These un-
scure and sometimes contradictory, the fortunates were treated by establishing for
snatches of evidence they yield are il- them a rigid program of exercises and
luminating. drill in perceiving. The conclusion of
The evidence consists of what the pa- one investigator was that "to give sight
tients said when they opened their eyes, to a blind-born person is more the task of
their answers to the doctor's questions, an educator than a surgeon" (page 145).
Nevertheless there is evidence of spon- no words for the features of his environ-
taneous unlearned visual capacity in these ment. The patient could not assign to his
patients. Fixation of objects and the impressions terms like black and white,
scanning of the environment was possible moving or still, far or near. He could not
from the outset. The patients could fol- immediately say whether there were two or
low a moving object with their eyes and three black spots on a piece of paper. He
head. Grasping an object, in which the could not even point on request to an edge
hand is guided by the visual images of the or a corner. He could not, in fact, say
hand and the object (as contrasted with anything about his visual impressions. He
the groping of the blind), seemed to be had these terms in his vocabulary but they
possible, although the act was slow and referred to tactual and muscular impres-
unsure. Watching the movements of the sions only. Although this inability to
hands in front of the eyes was reported describe anything made it almost impossi-
to be an action of great apparent interest ble to discover what the patient saw, his
to them. Looking at brightly colored sur- behavior indicated that he did have visual
faces was reported to be pleasurable, al- impressions and that some of them differed
though the colors could not, of course, be from others.
named. There are perfectly clear cases in which
Most of the investigators had precon- the patient could use the words "same" or
ceived ideas of what the patients might "different" (94, p. 155-157). For example,
see. Believing as they did in sensations two strips of cardboard, 10 cm. and 20 cm.
of color and perceptions of space, they long, were seen as different but the patient
assumed that there were three alternatives: could not say that one was "longer." That
the observer might see things at the eye, word meant something quite different from
that is, touching it, or he might see things what it does to us. A silver pencil and
in a flat plane in front of the eye, or else a large key appeared different as they lay
he must see space with things in it. To side by side on the table, but they could
the eager questions about such percep- not be identified or named. When they were
tions the patient could not respond. He put in the patient's hand, however, they
did not understand them, although he could be named at once. This happened
might fall in with the unconscious coach- with other pairs of objects such as knives
ing of the questioner. On reading the and forks, or cubes and spheres. They
answers now, the obvious conclusion is could be told apart but not recognized
that the patient saw none of these possi- by sight, even though they were things
bilities. There are indications that he which could be recognized when felt.
saw more nearly a visual world than a flat At the outset, the patient could not
visual field, but something less determinate even say longer or shorter, curved or
and less specific than the literal world straight, square or round, thick or thin,
which we see. wide or narrow, much less words like
The reports make it obvious, again and cardboard, key, or fork. He could react to
again, that the newly seeing person had the difference, apparently, before he was
LEARNING 219
able either to specify the difference or to therefore manifest a minimum of stimulus-

name the objects. generalization. There is nothing to show
The histories suggest that some com- in any of the cases described that the two
parative terms were applied to the visual modes of sensory impressions for the same
world much more rapidly than others. object were in the least connected at the
Colors were soon named; visual motion was beginning. There was no evidence that
quickly called what it is; large and small an object intrinsically looks the way it
together with far and near seemed to be feels. One might suppose, for instance,
used appropriately without much delay. that the visual and the kinesthetic quality
The chief difficulty seemed to come in of "vertical" are originally alike. A case
learning to use the innumerable terms for is described by von Senden (page 158) in
comparing and naming visual shapes. Per- which the patient was asked, some weeks
haps this is not surprising in view of the after the operation, to identify a horizon-
subtlety and variety of those spatial fea- tal and a vertical line drawn on a card.
tures by which we distinguish objects from He could do so only after tracing them
one another. According to the accounts with his finger. Presumably he had learned
it took many weeks or months to be able by this time to trace, that is, to correlate
to name common shapes, that is, to the sight and the feeling of his moving
identify the objects, places, events, people, finger. This phenomenon will be further
and animals of the environment, and still considered in the next chapter.
longer to learn the signs and symbols. Another task faced by the patient is the
The patient had to undertake the task of naming of new visual impressions which,
revising radically the meanings of the unlike manipulable objects, never had any
words he used. tactile-muscular counterparts, for instance,
Kinds of Visual Learning. Since a per- the many kinds of forms, places, events,
son who sees for the first time literally signs and symbols. Such impressions are
does not know what anything looks like, very numerous and many of them are quite
although he knows what anything he could similar to one another. An extreme ex-
touch feels like, he must learn. A con- ample would be learning faces or, more
siderable part of his task is to name the precisely, learning to identify people
new visual impressions using the words by the shape of their faces. With this
previously aroused by the old tactile and kind of learning all the operated patients
muscular impressions. This kind of had extreme difficulty and made slow pro-
naming is analogous to what is called gress. There are analogous experiments
paired-associates learning in the psycho- on identifying nonsense forms (34), learning
logical laboratory when new cue-items are a code (62), and recognizing aircraft (39).
substituted for old ones. Such experi- It could be predicted from these experi-
ments would imply that transfer of learn- ments, and from E. Gibson's theory based
ing should occur (47). It should be noted, on "intra-list generalization" (33), that
however, that visual stimuli and tactile - the more items there are to be identified,
muscular stimuli are dissimilar and should and the more similar they are to one
another, the more difficult will be the like objects at the first nesting season
learning. It would also be predicted that inasmuch as they retrieve and care for
there should occur mis-identification of such objects only (page 207). Geese show
similar items,
tems, and that the learning of new fear at the sight of a moving hawk-like
items should cause forgetting of similar silhouette in the sky but not at an other-
old items. All these phenomena are re- wise identical goose-like silhouette (107).
ported in the learning of the cataract If Spitz is correct, human infants retain a
patients. trace of instinctive behavior in their
It is obvious that these individuals had tendency to smile at face-like objects and
to learn to see the world. Putting words at these only (page 208). It would appear,
to their impressions had to be learned, then, that identifying responses are not
and perhaps this is no small part of seeing. necessarily learned in their entirety,
The accounts do not suggest, however, possibly not even by men. In the terminol-
that surfaces, edges, slants, and shapes ogy of the previous chapter, there may be
that is to say space were at first invisible embryonic meanings in the first visual im-
and later became visible. They suggest, pressions.
instead, that these variables were not at The question of whether or not space-
first, or not completely, identifiable. The perception in animals is learned is a poor
impressions were, however, discriminable, question. What can be asked sensibly is
and they might have been demonstrated to whether animals with no opportunity for
be such had the investigators used more practice will react appropriately to spatial
ingenuity in testing. situations at the first occasion. One of
Unlearned Visual Identifications in the best of such experiments is Lashley
Animals. Things, places, and events may and Russell's study of the jumping be-
be identified without necessarily being havior of rats reared in complete darkness
named. Animals other than man are (72). Thirteen rats were allowed to grow
limited to this kind of identification. When to maturity without seeing even the walls
an organism responds in some unique way of their box except for a few seconds
to a fact of the physical environment it every other day when food was inserted.
may be said to have made an identifying At 100 days of age each rat was brought
response. When this reaction depends on into the light and placed on a high platform
a retinal image it is a visual identification. with a gap between this and another plat-
Naming things is only one type of visual form containing food. The rat was first
identification, although an extremely im- allowed to step across a 5 cm. gap five
portant one for human learning. Our hypo- times, and then was confronted with a 20
thetical pre-human ancestor who could cm. gap. Twelve jumped the distance
pursue zebras but fled from sabre-toothed successfully on the first trial, and all
tigers was identifying things. could do so very consistently over ten
Instinctive behavior involves not only trials.
unlearned motor activities but unlearned Subsequently each rat was given three
identifications. Birds can identify egg- trials at 40 cm., three at 20, and three
LEARNING 221
more at 40. Most of them failed at 40 cm. The Study of Visual Identification. The
or refused to jump, but the significant fact nature of identifying responses, including
is that they increased the force of their naming, is not very well understood. It
jumps with the increased distance. The is becoming clear, however, that the dis-
force of the jump was recorded automatical- criminating of stimulus-variables is an
ly by the jumping platform. Finally they essential component of all learning. An
were allowed nine jumps with the landing organism cannot learn a reaction without
platform at varying distances. The force identifying the cue for it. The object must
proved to be graded in proportion to the be reacted to as the same object on dif-
distance. These rats could not jump as ferent days. Similar objects and events
successfully as normal rats reared in the are said to be equivalent for behavior un-
light (they fell more often) but the grada- less the organism discriminates among
tions of force were nearly as accurate. them; a stimulus to which a response has
The implication is that they could dis- been conditioned is said to show general-
criminate the distance nearly as well. ization in that the response will occur to
In the terminology of this book the rats a whole class of stimuli.
were probably responding to depth-at-an-
edge, which is the essential feature of a The ordinary discrimination experiment
gap in the floor. Not only were they re- isolates a pair of objects or situations
acting differentially to different abrupt differing in only one respect. The psycho-
changes in optical stimulus gradients: physical experiment in its simplest form
they were with some success identifying a does the same thing. The former elicits
fact of the physical environment by an differential reactions while the latter
appropriate act. The identifying act seems elicits a discriminative judgment of "more"
to have been innate. or "less," but there is a basic similarity
Other experiments with animals reared between them. Objects and events in
in darkness are being continued in Lash- daily life, however, do not come in neatly
ley's laboratory using chimpanzees, whose controlled pairs. In order to understand
perceptions are probably more like man's. the nature of cues and of learning a dif-
The first tests for visual perception in- ferent kind of experiment is necessary. It
dicated that these animals were blind (90), would be an experiment which requires
or at least that they failed to use vision in the subject to react in a unique way to
guiding their behavior, but no conclu- each of a whole set of objects or events,
sions can be drawn since it is possible more or less similar. These items should
that the chimpanzee's retina does not differ in many dimensions of variation and
mature normally in darkness.2 should fall into classes and sub-classes.
2 Attention is centered on the errors of iden-
Since this passage was written, the writer
has been informed by Dr. A. Riesen that this tification and how they are eliminated. As
explanation is indeed the most probable one. a type it might be called the identification
The visual defect seems to be due to abnormal experiment. Research of this type is be-
maturation, and it cannot be ascribed wholly to
the absence of learning. ginning to appear (34, 62, 39).
Learning to See. Considering the evi- one another as they do and innumerable as
dence, in what sense do we learn to see they are.
and in what sense do we not? Clearly, we As things become identifiable, and as
do not learn to accommodate, converge, we learn to notice the differences between
fixate, and move the eyes, although it is them, our perceptions of the world become
very possible that practice improves these differentiated. Formerly indefinite quali-
functions. We do not, in other words, learn ties become definite. Shapes and textures
to produce optimal visual images of and surfaces and colors become specific.
speckled light having borders, cycles, Indeterminate movements, locations, sizes,
shades, and gradients of intensity and and distances become determinable.
frequency-mixtures. We do not learn the Properties like inside and outside, con-
space-values of separate retinal points. gruency, symmetry, opposition, and con-
We do not learn to associate retinal points tinuity are elaborated. Objects, events,
so as to see form. We do not learn to in- and situations are recognized. In the
terpret color and form sensations so as to case of human beings, things are named.
see the third dimension. What we do learn The qualities of, or differences between,
preeminently is to identify the features of objects are also named. This enables us
visual stimulation which correspond to the to name classes of objects. Once this
important features of the physical environ- process is started it builds upon itself;
ment. new differences emerge, new similarities
Words facilitate but are not essential to become visible, and more general classes
this process of identification. Whatever are named. At the same time more and
it may prove to be in detail, it is sure to more objects are identified. The tradi-
involve the discriminating of complex tional way of putting it is to say that things
variables discoverable in retinal images. have meaning and that we have abstract
There must occur, in other words, a dif- ideas about them. But the progress of
ference in response along with a difference learning is from indefinite to definite, not
in stimulation either a discriminative from sensation to perception. We do not
reaction or a discriminative judgment. The learn to have percepts but to differentiate
discriminating of variables is necessary them. It is this sense in which we learn
for the identifying of things, grading into to see.
1M
Spatial Perception and Spatial Behavior
The Motor Theory of Space-Perception ....

Vis-
ual Kinesthesis . The Co-variation of Visual
and Muscular Motion ... . The Ego in Percep-
tion.... The Impression of Distance from
Here to There . ... Orientation
Ever since Berkeley's V ew Theory of does not state one. Why does it seem to
Vision in 1709, it has appeared plausible be consistent with the observations of so
that the seeing of space depends, in some many people? Introspectively, it is a fact
fundamental way, on exploring and mani- that seeing is almost inseparable from
pulating the environment. Seeing things, acting. Spatial behavior is intimately con-
Berkeley argued, could always be verified nected with spatial perception. Things
by touching things, and hence it was pos- do look capable of being grasped, or
sible that the solidity and depth of the pushed, or touched on all sides, or of
visual world were originally not visible resisting these actions. The floor does
but only tangible. Vision might get its look capable of being walked on whereas
spatial character from the tactile and the walls do not, and neither does a gap
muscular impressions which always ac- or a hole in the floor. An edge looks
company it. We learn to trust our vision capable of being traced with the finger. A
of the table as being there, for instance, slant looks as if you could angle the palm
because we can always go over and touch of your hand to it, or climb it. Thus, the
i t. visual world has the appearance of in-
viting many kinds of behavior.
The Motor Theory of Space Perception It is one thing to say that the visual
We know that the infant and young child world has motor meanings, however, and
ceaselessly explores his environment as quite another to maintain that it gets its
his vision develops. Is it not likely that spatial qualities from these meanings. It
his visual impressions get their solidity is possible to agree that we can almost
and depth from their association with feel the visual scene without concluding
these movements? This is the argument that we do not really see it but only re-
for a motor theory of perception. member how it feels. Perhaps what is
This theory has such great persuasive- wrong with a motor theory of perception
ness that it must embody a truth even if it is its one-sidedness. The infant explores
223
his environment and feels the things he stimuli which are precisely co-ordinated
sees, true enough, but he also sees the with both locomotion and manipulation.
things he feels. If visual impressions ac- In cases of what is called locomotor
quire motor meanings, is it not just as in- ataxia, the patient has no kinesthetic
evitable that motor impressions should sense in the lower part of his body. Dam-
acquire visual meanings? Perhaps the age to nerve centers in the spinal cord
trouble with a motor theory of space per- has blocked the transmission of impulses
ception is that it needs to be supplemented from the receptors in the muscles and
with a visual theory of muscular impres- joints and from the soles of the feet. He
sions. For every seeing individual there cannot feel the position or motion of his
is a co-variati:_ of retinal stimulation and legs. He walks with a very peculiar gait,
muscular-tactile stimulation during be- as if he had to throw his legs forward, and
havior. Neither kind has to be taken as when he is blindfolded he cannot walk at
primary. all, or even stand up. The thing to note,
however, is that he can stand and walk,
after a fashion, if he looks at the ground
Visual Kinesthesis
and his feet while he does so. Without
The term kinesthesis, meaning sensiti- bodily sensitivity the use of the legs is
vity to motion, is usually applied to the impossible unless visual sensitivity ful-
muscle-sense. Actually, there are recep- fills the necessary functions. It would
tors in the muscles, tendons, and joints seem that there are two forms of the kin-
which yield impressions of both the motion esthetic sense; one based on the known
and position of our limbs. There are also list of proprioceptors, and one based on
receptors in the inner ear which yield im- vision.
pressions of the motion and position of The term sense is almost as misleading
our head and body. Moreover, the touch as the concept of sensation. Kinesthesis
receptors anywhere on the skin can yield is not one of the senses; there are not just
impressions of motion, and those in the so and so many departments of sense.
hands and feet notably do so when we Kinesthesis is mediated by a number of
manipulate things and walk about. Im- types of receptors. It probably should be
pressions of our own movements are known conceived very broadly, but its basic
to depend on all these sources, but there function is to adjust and set the pace for
is another and usually unrecognized muscular action. It guides manipulation
source in the retina. The reasons for and the using of tools. It also guides
using the term visual kinesthesis have locomotion of the ordinary sort which is
already been given in Chapter 7. A driven by muscular action. In a more
tabulation of the kinds of retinal motion complex form it guides locomotion of the
and their correspondence to the move- sort driven by gasoline engines, propel-
ments of external objects and of the body lers, and jets, and in the case of the
is given on page 132. Thc point to be em- automobile and the airplane, locomotion
phasized is that there are normally retinal requires complex manipulation. In such
SPATIAL PERCEPTION AND SPATIAL BEHAVIOR 225
advanced types of spatial behavior, visual The diagram helps to explain why visual
kinesthesis plays an essential role. impressions and motor impressions imply
each other when we introspect on our ex-
The Co-variation of Visual and Muscular perience. Retinal motion is automatically
Motion linked to bodily action from birth onward,
so long as the eyes are open and there is
In as apparently simple an act as pick- light to see by. Bishop Berkeley was cor-
ing up a pencil both visual and muscular rect in asserting that to walk over and
impressions are involved in the control of touch the table is to confirm visual space;
the performance. The contracting image of what he did not understand is that the ex-
the hand projected on the retina as it panding visual field also confirms muscular-
coincides with the image of the pencil is tactile space.
paralleled by the feeling of the hand being Spatial behavior and spatial perception
extended and touching the pencil. This is are coordinate with one another. We have
called an eye-hand co-ordination. The neither to see space before we can behave
retinal motion and the muscular impression nor to make spatial responses before we
vary together, both being controlling can see. A man who gets about his en-
stimuli for the pace or flow of the act. vironment without collisions and mani-
Animals stimulate themselves as they pulates objects to the satisfaction of his
act and this stimulation affects the action. wants is exercising a function in which
The process is circular; it has recently his behavior is spatial and his space is
been compared to the feedback mechanism behavioral.
of electronics (120). In the case of spatial
behavior a visual component must be The Ego in Perception
added to the circle, so that the complete Perceiving the world has an obverse
process is like that diagrammed in Figure aspect, perceiving oneself. Observers
80. have often pointed out that one's own
SPATIAL BEHAVIOR MUSCULAR TACTILE RETINAL MOTION

OF THE BODY STIMULATION STIMULATION
FIGURE 80. The Coordination of Spa-

tial Impressions and Spatial Behavior
body is represented in the visual field, also by acceleration of the head affecting
and it has been argued that the ego is the inner ear.
therefore an object in the field of ex- 4. The movement stimulation aroused
perience like any other object (67, page by changing contact of the skin with sup-
319-331). A more satisfactory statement, porting surfaces and resisting surfaces.
however, is that perceiving the environ- 5. The so-called "boundary of the
ment includes the .go as part of the total visual field," i.e., the peripheral retinal
process. In order to localize any object images of the nose and other parts of the
there must be a point of reference. An body. This includes images of the hands
impression of "there" implies an impres- and feet which protrude into the visual
sion of "here", and neither could exist field from its lower margin.
without the other. 6. The retinal displacement of these
The definition of the ego is a problem bounding images. They shift in a specific
with which psychologists and philosophers way when the eyes turn; they also shift in
have struggled without much success. a different way when the head turns. The
The concept of a self, by whatever term shift is "concomitant and reciprocal" with
it is called, is necessary for any scientific the turn.
theory of personality, of social behavior, 7. The deformation of the whole retinal
of abnormal behavior, or of ethical be- image. The visual field expands when one
havior. A clarification of the social ego goes forward and contracts when one goes
(98) might be possible if a few solid facts backward. The maximum velocity of this
could be stablished about the biologi- motion is reached at the boundary at
cal ego, that is, the ego manifested in the images corresponding to the parts of
maintaining equilibrium and posture, in the body.
locomotion and manipulation, and in literal 8. Finally, several types of more or less
visual perception. continuous organic stimulation known gen-
What are the stimulus-correlates for the erally as sornaeithesia, the self-stimula-
perception of oneself? There must exist tion involved in breathing for instance, and
multiple correlates, or what used to be also in eating, drinking, and sexual acti-
called "co-operation of the senses," vity, together with the stimulation under-
rather than a single correlate. Tenta- lying bodily discomfort or comfort.
tively, the following can be listed: These are at least some of the forms of
1. The tensions of the skeletal muscles stimulation which might yield a primitive
which maintain equilibrium and regulate ego in perception. The notable thing about
the posture of the legs, trunk, head, and them is that they all co-vary with action.
eyes. The process is circular. An overt perfor-
2. The contact stimulation of the skin mance is controlled by various types of
against the surfaces on which the body kinesthetic stimulation, including the
rests. visual. A posture, on which any overt
3. The movement stimulation aroused performance is based, is controlled by
by action of the muscles and joints, and stimuli of the first seven types listed
4'0
40,-, "- "'
/°°:Pol/,*44:.6.,
,6 ..
0,ex .
. --
""f- _, ,!. ", //
,. -ter- Llllrr
4,
Pr
.. do 44011-404,. *AO
or
414
, ..
40
-.4611111*LageAlkorafigh.
'.0
,
,
"'''';',/'19-r/e/,
,4 :///
,
A.,
-
V.:,n,t.l."!::::.'Ye7 iv , .. :
' I
/ 44 4ep;././:/C.:
/
--. -
de 01
.7rY e 4 //JP/ ..t.'411PiO
:
..
. , i
%
[ ; - .' ; ..---41.1.1'; '
;:"§,#,,,,,.,,al fi
AII #4,014,.
0,1;e,./ZO": 2/"../4,..4.2.".4./...;ii,:.=.. .'
, ...,0,...,
1 ,.t.,0:24,
. -,
,,, : .
,.,. .7:
/,.. .-
,
.
/#0,,V!,:::;,!:-,; : ,. /
, ,5/
_ _
I *VV.*
. /... .
,41 .,S, ' " ..-,.
v-,0,s`
'/ .../..../.
j.;'
" 0 ,//, 1; in
.../..,,,1 , i ',/ ,,,y , t,
i ; .',' :,,,/
0.('....:;; ,# i .: i ..,-/- ",(4"7/2,"
_:. -. ,,, / ,. i
..,,...././,,,,,,- ,;.5. 0,'/././//,41.4Z4V,V %/7
1 , 11%/90
de f /7.::,-.
,- e , ;:.- -. 4:- 1111111;'
"::::,
A 40.--
/.:,,:-.-
,,...,......,.. /
e:.;.-
,,,,
..--,.:
fre
7, -
°
i
,,.........,....4,, ,.,,,,.
5
',pm, 4 .10. %%1114 1,,,,,,
;
5
.1,--
4.--
-4 'Ill/MAI,/
, /,.
./ .,
1
7 :--
,... 1 ..:?#;.0
1.:41' : ; ..., ,1'1/
,,,,
, ,, ,.....,.2 ,......., ,.... ..,,,,
,,4.'k
-.
........-
,. ,
,
"-.47 / ... 1 4 \ A17
'-
.,,
::".,,
00::::.:- . rv :::;. -,,,,, , -,-. ........,"-.rt, '',4,00.740," / -,.. OW, ': 11111.1111111M. V4 /,
re .:.;' " . , =.?
40:!,:::: 411
Vfifill
"Ak-r
. ;
,?
1
II'
.1,,,.
"(
...k-
1.111111 ., ,
1:
il111111111/4\k
%.; ,
ifili
Ar : ::
9414t::: :': : :
Al
.. ,..
.
..//' /,
-1
, "'",':?-/55.5/,
. lz. s'4 4 7/.., re/ ,
' Ie 1
7:.: ,.. _
J4
/
/
A,A
7
/Sk
.
,'/..',: : . ... . .., 4 /
"17,1 41
'144. :-
4944':`,.
... ..,.......... ....... ,% // .:
44, ,, A
/4:::,'::f
//....,..- .owl ,,,
/// / ''-/,./
./
1 ...' ' tO 4
I ,< . ':.. ::16::.,(.."74/
A.71. '''^_- ''.*- ;///,'.--, ,. ., ...... ,, v..
/ .---
. ...:,,,..
.40
A.;.-_-
-
,..
.,...,,,,. -,4. <, ,,:.;
..
,
y
,1
4.--$0',...A0 -
,/,...".-/...
- .7 '.:,,
,,, - 1 dr .::::: :4::iir: ... AC' 91/
:,?;w/./
.,.-,". 'i:.../.."::;'111:::.r.
a ./"... -:-.- oe /
fr., , / ,;.*
, .._ 4 .6
' - vire
%-\.
,010:`
00
-.- ' ro
420
sir
4/ /
4.0
/1.'4/ 47,1447.;,-
/ :" .;1:1°1
9*,*,..
4//-?11
' fr -;;. 1-',0 w .//,
"41,////4;
/
/ /;..44///
./0 b4Y/ ihy" ., 4w ir
////
above. A motive or need, from which any the rate of deformation of texture, and a
performance takes its origin, is controlled step in the binocular disparity of texture.
by stimulation of the eighth type. What- The steps are proportional in amount to
ever else the ego may be, it includes the physical difference in depth between
elements of performance, posture, and the two surfaces in question.
need. Perhaps these are the fundamental
The same kind of reasoning may be ex-
components of the biological ego the
sense of one's own body and the impres- tended to the impression of the distance
from here to there the space between
sion of something that is uniquely and
continually "here." oneself and the ground or the distance
to the wall. The boundary of the visual
field, particularly the image of the nose,
incorporates all three of these abrupt
The Impression of Distance from Here to steps in ordinal stimulation. The nose is
There given as a very large area in the visual
The central hypothesis of this book as- field. (In the writer's case, the line from
serts that space is constituted of the the tip to the nostril coincides with a foot
same variables as things. It holds that rule held at arm's length, but he may be
surfaces and margins are what we see, not especially well endowed.) It is also a
air. Space must be filled to be visible; crossed double-image. Moreover the edge
empty space is an abstraction. Against of the nose has a marked visual velocity
this hypothesis the objection can be made when the head moves or turns. The nose,
that we do have visual impressions of in fact, projects the maximum possible
empty space and we have them all the degree of crossed retinal disparity and
time. We see the .distance from here to should, therefore, according to the present
there; we "look out" upon the world, and theory, represent the greatest possible
the space between one's eyes and the degree of visual nearness (see page 106).
nearest object is plainly empty. The ob- Likewise, at the margin formed between
jection is convincing, but it can be over- the nose and external reflecting surfaces,
come. the greatest possible velocity of parallac-
One kind of empty space, the distance tic motion is reached in the retinal image,
between an object and the surface behind and this also should represent the
it, has already been accounted for in maximum impression of nearness. These
terms of depth-at-a-contour. The visual facts establish an absolute zero of dis-
superposition or overlapping of surfaces; tance in stimulation and this, in turn,
it was argued, is an important type of makes possible an absolute scale of dis-
depth-perception, not a cue for depth- tance in experience.
perception. The explanation was found There is nothing mysterious, then, in
in steps, as contrasted with gradients, of the impression that a person "looks out"
the main stimulus-variables for distance upon the world, that he sees distance or
a, step in the density of texture, a step in space continuously extending from "here"
SP ATI AL PERCEPTION AND SPATIAL BEHAVIOR 229
to "there" and that he, himself, is "here." houses, streets, cities, and countries,
This impression is in correspondence with contains millions of things to which a
retinal stimulation. The perception of the man can find his way a cigarette, a cup
world and the perception of oneself are of tea, a tax receipt, a racetrack, or a
both, figuratively speaking, cut from the sweetheart in Texas.
same cloth) The study of locomotion is in its in-
There is no necessity for the assumption fancy. Although a good deal is known
that a person projects his sensations out- about the movements of walking, swim-
ward from the eyes. All the ingenuity that ming, and flying, surprisingly little is
has been devoted to explaining such an known about how animals and men find
event has been wasted. The steps of their way about (88). The explanation is
distance at the edges of things and the probably that we do not fully understand
great first step of distance at the windows the process of orientation. There is one
of the eyes themselves are all to be found special form of locomotion, it is true,
within the eyes. which has been very thoroughly investi-
gated the behavior of white rats in run-
Orientation
ning through mazes. But maze-learning
All organisms from the lowest to the seems to be a very complicated sort of
highest can orient themselves to certain behavior, and, it does not represent a
forces of the physical environment. Plants simple form of locomotion. How the rat
can orient to the direction of gravity and comes to be oriented to the goal-box of a
the direction of light. Animals possessing maze is a disputed issue, as the last chap-
vision and locomotion react to finer varia- ter indicated (58, 109).
tions and can orient themselves, not only How, exactly, is oriented locomotion
to the sun and the earth, but to things. to be defined? What would be a fundamen-
The human species has a still more dif- tal experiment on the process of getting
ferentiated environment and can orient to about? Lewin and his students (74) made
very complex and unique features of it. a start on these questions fifteen years
The human habitat, consisting of rooms, ago, but they became so interested in
using locomotion as an analogy for higher
forms of behavior that they never got down
to studying the literal process. The sim-
plest kind to study would probably be the
1So act of going to a visible object or place,
important is the impression of being
"here" to having a sense of self, and so im- the goal. Locomotion of this sort is
portant is the visual image of the nose to the
impression of being "here," that nose-per- oriented directly toward the goal. The
ception must be a prominent factor in the body movement is a function of optical
awareness of the ego. There have been a
few students of nose-perception such as stimulation which yields the perception of
Cyrano and Durante, and they have clearly a visual world with the goal - object in it.
understood the relation between a man and
his nose, but the subject deserves more in- Body movement is modified only by the
vestigation. necessity of avoiding obstacles, or direct-
ing the movement into the field of safe more inclusive spaces of which the most
travel (41). general is the astronomical universe.
A more advanced form of locomotion The conception of an objective world,
would be the act of going to an object or independent of the standpoint of any ob-
place beyond the range of vision. This server, rests upon this type of orientation.
might be called the destination. "Being The ability to take the position of
oriented," in the popular sense of the another person, to see from his point of
term, refers to this ability. In common view, depends on being oriented in space.
sense terms, one must know both where he Orientation is inseparable from locomotion,
is going and where he is now. It requires, for, only because an observer gets a dif-
over and above the visual world, a frame ferent visual field at every different stand-
of reference (93) or a topographical point, does he perceive a single integrated
schema (46). The individual must per- world (Chapter 8). That is, because the
ceive the space which surrounds him on visual field changes systematically with
all sides, as described in Chapter 3, and change in the position of the head, an
must also apprehend the world beyond the ordinal network is established among the
visible scene the layout of the building, diverse fields yielding a visual world in-
of the city and its streets, of the region, dependent of any standpoint. The facts
and of the country with its highways and of spatial perception and of spatial be-
cities. He is then said to be oriented in havior are united in the fact of the visual
space actually, in a series of more and ego.
References
Ames, A. 'tBinocular Vision as

(1946), Bender, M.D., and Teuber, H.L. (1949),
Affected by Relations Between Uniocular "Disturbances in Visual Perception
Stimulus-P atterns in Commonplace En- Following Cerebral Lesions, " Jour.
vironments," Amer. J. Psychol., 59, pp. Physiol., 28, pp. 223-233.
333-357.
Berkeley, G. (1922), 4 New Theory of
Asch, S.E., and Wtkin, HA. (1948), Vision and Other Writings. (Everymans
"Studies in Space-orientation. I Percep- Library ed.) New York: Dutton.
tion of the Upright With Displaced Visual
Fields," J. Exper. Psychol., 38,pp. 325-337. Boring, E.G. (1933), The Physical Dimen-
sions of Consciousness. New York: Cen-
tury.
AscIi, S.E., and Witkin, H.A. (1948),
"Studies in Space Orientation. II Percep- Boring, E.G. (1935), t1The Relation of the
tion of the Upright with Displaced Visual Attributes of Sensation to the Dimensions
Fields and with Body Tilted," J. Exper. of the Stimulus," Philos. Sci., 2, pp. 236-
Psychol., 38, pp. 45 5-477. 245.
Asch, S.E., and Witkin, H.A. (1948), Boring, E.G. (1942), Sensation and Percep-
"Studies in Space Orientation. III Percep- tion in the History of Experimental Psychol-
tion of the Upright in the Absence of a ogy. New York: Appleton-Century.
Visual Field," J. Exper. Psychol., 38,
pp. 603-614.
Bourdon, B. (1902), La Perception Visuelle
de l'Espace. Paris: Schleicher Freres.
Bruner, J.S., and Goodman, C.C. (1947),
"Value and Need As Organizing Factors in
Perception," J. Abri. Soc. Psychol., 42,
pp. 33-44.
Bartlett, F.C. (1932), Remembering, A
Study in Experimental and Social Psychol- Brunswik, E. (1944), "Distal Focusing of
ogy. Cambridge: Cambridge Univ. Press. Perception: Size-Constancy in a Represen-
tative Sample of Situations," Psychol.
Bartley, S.H. (1941), Vision, A Study of Monog., 56, No. 254, pp. 49.
its Basis. New York: Van Nostrand.
Buhler, K. (1922), Handbuch der Psychol-
Bassett, M.F., and Warne, C.J. (1919), ogie. i Teil. Die Struktur der Wahrnehmun-
'tOn Lapse of Verbal Meaning With Repeti- gen. J Heft. Die Erscheinungsweisen der
tion," Amer. J. Pschol., 30, pp. 415-418. Farben. Jena.
231
232 REFERENCES
Buswell, G.T. (1935), I/ow People Look at Eisenstein, S. (1942), The Film Sense
Pictures. Chicago: Univ. Chicago Press. (Tr. by J. Leyda) New York: Harcourt
Brace.
Cantril, H. (1948), 'The Nature of Social
Perception," Trans. \'.Y. Acad. ScI., lo, Ellis, W.D. (1938), 4 Source Book of
Gestalt Psychology. New York: Harcourt
pp. 142-153. Brace.
19 Corr, H.A. (1935), An Introductwn to
Space Perception. New York: Longmans Gibson, E.J. (1949), 'A Systematic Ap-
Green. plication of the Concepts of Generaliza-
tion and Differentiation to Verbal Learn-
Carter, L.F. and Schooler, K. (1949), ing," Psychol. Rev., 47, pp. 196-229.
"Value, Need, and Other Factors in Per-
ception," Psychol. Rev., S6, pp. 200-207. Gibson, E.J. (1942), tIntra-List General-
ization As A Factor in Verbal Learning,"
Cassirer, E. (1944), tThe Concept of J. Exper. Psychol., 30, pp. 185-200.
Group and the Theory of Perception,
Philos. arid Phenom. Res., 5, pp. 1-35. Gibson, J.J. (1929), 'The Reproduction
of Visually Perceived Forms," J. Exper.
Child, C.M. (1941), Patterns and Prob- Psychol., 12, pp. 1-39.
lems of Development. Chicago: Univ.
Chicago Press. Gibson, J.J. (1933), tAdaptation, After-
effect, and Contrast in the Perception of
Collins, F0T,, (1922), Arboreal Life and Curved Lines, ' ' J . E xper. P sychol. ,
the Evolution of the Human Eye. Phila- pp. 1-3 1.
deiphia: Lea and Febiger.
Gibson, J.J. (1937), "Adaptation, After-
Conant, J.B. (1947), 'tOn Understanding effect, and Contrast in the Perception of
Science," 4mer. Scientist, 35, pp. 33-55. Tilted Lines. Il Simultaneous Contrast
and the Areal Restriction of the After-
Courant, Robbins, H. (1941),
R., and effect, J. Eper. Pcychol.," 20, pp. SS 3-
What is Mathematics? New York: Oxford S69.
Univ. Press.
Gibson, J.J,.(1937), 'Adaptation With
N egative After- Effect., ' ' Psy chol. Rev.,
Dodge, R. (1902), 'Five Types of Lye 44, pp. 222-244.
Movement," 4mer. J. Physiol., 8, pp. 307-
329. Gibson, J.J. Ed. (1947), .1otion Picture
Testing and Research. (Report No. 7,
Don, V.J., Weld, H.P. (1924),
and A AF' Avi ation P sychology Research
"Lapse of Meaning With Visual Fixa- Reports). Washington: U.S. Gov't. Print-
tion," Amer. J. Psychol., 35, pp. 446- ing Office.
4 50.
Gibson, j.J.
(1948), Studying Perceptual
Duncker, K. (1929), uUber .lnduzierte Phenomena". In T. Andrews, Ed.,
Bewegung. Fin Beitrag zur Theorie 1ethods of Psychology, New York:
Optischer Wahrgenommener Rewegung," Wiley, pp. 158-188.
Psycho!. Forsch., 12, pp. 180-2S9.
41. Gibson, J.J., and Crooks, L.E. (1938),
29. Duncker, K. (1939), "The Influence of tA Theoretical Field-Analysis of Auto-
Past Experience Upon Perceptual mobile-Driving,'' Amer. J. Psyc/zol., 51,
Properties," Amer. J. Psychol., 52, pp. 453-471.
pp. 255-265.
42. Gibson, J.J., and Mowrer, O.H. (1938),
30. Duke-Elder, WS. (1938), Textbook of "1)eterminants of the Perceived Vertical
Ophthalmology, Vol. II. London: Henry and Horizontal," Psychol. Rev., 45,
K i mp ton. pp. 300-323.
REFERENCES 233
Gibson, J.J., and Radner, M. (1937), Holt, E.B. (1903), "Eye-Movement and
t
'Adaptation, After-Effect, and Contrast Central Anaesthesia," Psychol. Monog.,
in the Perception of Tilted Lines. I 4, 17, pp. 3-46.
Quantitative Studies, J. Exper. Psychol,
20, pp. 453-467. Holway, A.H., and Boring, E.G. (194fl,
"Determinants of Apparent Visual Síze
Grant, V.W. (1938), Psychological Optics With Distance Variant," Amer. J.
Chicago: Prafessional Press. Psychol., 54, pp. 2 1-37.
Graybie A., Clark, B., MacCorquodale, Hull, C.L. (1943), Principles of Be-
K., and Hupp, D.I. (1946), 'tRole of havior. New York: Appleton-Century-
Vestibular Nystagmus in the Visual Per- Crofts.
ception of a Moving Target in the I)ark,"
Amer. J. Psychol., 59, pp. 259-266.
Griffin, D.R. (1948), "Topographical Ivins, W.M. (1946), Art and Geometry.
Orientation,'' (Ch. 17 Boring,
in E.cI;. Cambridge: Harvard Univ. Press.
ILS. Langfeld, and h.P. Weld, Pounda-
lions of Psychology) New York: Wiley.
James, W. (1890), The Principles of
Psychology, Vol. II. New York: holt.
Hamilton, R.J. (1943), t R etroacti ve

Facilitation As a Function of Degree of 61 . Katz, D. 1935), The l'orid of Colour.
(
Generali zation Between Task s, ' ' J. (Translated by R.B. MacLeod and C.W.
/tper. lychol., 32 pp. 363-376. Fox) London: Kegan Paul, Trench,
Trubner and Co.
Harper, RS., and Boring, E.G. (1948),
":ues,'' Arrier. J. PsycIlc)1., 61, pp. 1)- 1 Keller, F.S., and Taubman, R.E. (1943),
123. "Studies in International Morse Code.
II Errors Made in Code Reception,"
Hebb, DO. (1937), "The Innate Organiza- J. 4pp1. Psychol., 27, pp. 504-S09.
tio)fl of Visual Activity. I Perception of
Figures by Rats Reared in Total Dark- Kepes, G. (1944), The Language of
ness," J. (;enet. Psychol., 51, pp. 101- iSiOfl, Chicago: George Theobald.
i 26.
Klemm, O. (1914), 1 I!istory of Psychol-
Hebb, D.O. (1949), The Organization of ogy. (Tr. by E.G. WUm and R. Pintner)
Behavior. New York: Wiley. New York: Scribner.
Heider, F. (1927), "I)ing und Medium," Klineberg, O. (1940), Social Psychology.

Syînposton, 1, pp. I()9-1S7. New York: holt.
Heider, F. (1944),"Social Perception Kluver, H. ed. (1942), Visual Mechan-

and Phenomenal Causality, ' ' P sychol. isms. Lancaster, Pa.: Cattell Press.
Rev., 5 1, pp. 358-374. (Vol. VII of Biological Symposia).
Helmholtz, H. (1925), Physiological Koffka, K. (1935), Principles of Gestalt
Optics, Vol0 3 (Ed., J.P.C. Southall). Psychology. New York: Harcourt-Brace.
Optical Soc. Amer.
Köhler, W. (1940), Dynamics in Psycho!-
Henneman, R.H. (1935), "A Photometric ogy. New York: Liveright.
Study of the Perception of Object-Color,"
rch. Psychol., 179, pp. 88. Khler, W. and Wallach, H. (1944),
"Figurai After-Effects. An Investigation
Hilgard, E.R. (1948), Theories of Learn- of Visual Processes," Proc. Amer.
ing. New York: Appleton-Century-Crolts. Philos. Soc., 88, pp. 269-357.
234 REFERENCES
Langeweisch., W. (1944), Stick and Rud- Miller, N.E., and Dollard, J. (1941),
der New York: McGraw-Hill. Social Learning and Imitation. New
Haven: Yale Univ. Press.
Loshley, K.S. (1938), "Experimental
Analysis of Instinctive Behavior," Morris, C. (1946), Signs, Language and
Psychol. Rev., 45, pp. 445-471. B ehavior. New York: Prentice-Hall.
Lashley, K.S. and Russell, J.T. (1934),

"The Mechanism of Vision. XI' A Pre- Nicolson, M.l1. (1946), Newton Demands
liminary Test of Innate Organization," the Muse. Princeton: Princeton Univ.
J. Genet. Psychol., 45, pp. 136-144. Press.
Le Conte, J. (1897), Sight. An Exposition
of the Principles of Monocular and
Binocular Vision. New York: D. Apple- Ogle, K.N. (1938), "induced Size Effect.
ton and Co. I. A New Phenomenon in Binocular
Space-Perception Associ ated With the
Lewin, K. (1936), Principles of Topol- Relative Sizes of the Images of the Two
ogical Psychology. New York: McGraw- Eyes," .4rch. Ophthal., 20, pp. 604-623.
H i i I.
Lewin, K. (1931), "The Conflict Between

Aristotelian And Galileian Modes of Pfaffmann, C. (1948), "Aircraft Land-
Thought in Contemporary Psychology," ings Without Binocular Cues: A Study
J. Gen. Psychol. 5, pp. 141-177. Based on Observations Made in Flight,"
Amer. J. Psychol., 61, pp. 323-335.
Luneberg, R.K. Mathematical
(1947),
Analysis of Binocular Vision. Prince- Raboud, E. (1928), hou' Animals Find
ton Univ. Press. Their ay About. (Tr. by 1.H. Myers)
New York: 1-larcourt Brace.
89. Ratoosh, P. (1949), "On interposition As

A Cue For the Perception of Distance,"
Proc. Nat. Acad. Sci., 35, pp. 257-259.
Mach, E. Contribiaions to the
(1897), 90, Riesen, A. H. (1947), "The Development
Analysis of the Sensations (Tr. by G.M. of Visual Perception in Man and Chini-
Williams) Chicago: Open Court. panzee," Science, 106, pp. 107- 108.
MocLeod, R.B. (1947), "The Effects of Rubin, E. (1921), Visuell Wahrgenom-
Artificial Penumbrae on the Brightness of mene Figuren. Copenhagen: Gyldendaiske
Included Areas, ' ' Miscellanea Psychol- Boghandel.
ogica Albert Michotte, Louvain: Institut
Suprieur de Philosophie, pp. 138-154. Ryan, TA. (1938), "Dynamic, Physio-
gnomic, and Other Neglected Properties
MacLeod, R.B. (1932), "An Experimental of Perceived Objects: A New Approach to
Investigation of Brightness Constancy," Comprehending," Amer. J. Psychol., 51,
Arch. Psychol., 135; pp. 102. pp. 629-650.
McDougall, (1921), An Introduction to

W. Ryan, T.A., and Ryan, MS. (1940),
Social Psychology. Boston: Luce and Co. "Geographic Orientation," Amer. J.
Psychol., 53, pp. 204-215.
8$. Metzer, (1930), ItOptische Untetsu-
W.
chungen am Ganzfeld. II Zur Phdnomen
ologie des Homogenen Ganzfelds,"
Psychol. Forsch., 13, pp. 6-29. Senden, M. von (1932), Raum-und Gestalt-
auffas sung bei operierten Jitindgeborenen
82. Metzger, W. (1936), Gesetze des Sehens. Ior und nach der Operation. Leipzig:
Frankfurt am Main: Kramer. Barth.
REFERENCES 235
Senders, V.L. (1948), "The Physiological Tinbergen, N. (1948), "Social Releases

Basis of Visual Acuity," Psychol. and the Experimental Method Required
Bull0, 45, pp. 465-490. For Their Study," gilson Bull., 60, pp.
6-5 1.
Severance,E., and Washburn, MF. (1907),
"The Loss Associative Power in
of Titchener, (1910), 1'etbook
E.B. of
Words After Long Fixation," Amer. J. Psychology. New York: Macmillan.
Psychol., IR, pp. 182-186.
Tolman, E.C. (1932), Purposive Behavior
Sherif, (1936), The Psychology of
M. in Animals and Ien. New York: Appleton-
Social Aorrns. New York: Harper. Century.
Sherif, M., and Cantri, H. (1947), The Tolman, E.C. (1948), "Cognitive Maps
P sychology of Lgo-lnvo1vemenL' . New in Rats and Man," Psychol. Rev., 55,
York: Wiley. pp. 189-208.
Spitz, R.A., and Wolfe, K.M. (1946), "The Troland, L.T. (1929), The Principles of
Smiling Response: A Contribution to the Psychophysiology, I The Problems of
Ontogensis of Social Relations," Genet. Psychology and Perception. New York:
Psychol. Monog., 34, pp. 57-12. Van Nostrand.
Stavrianos, B.K. (1945), "Fhe Relation Troland, L.T., (1930), The Principles
of Shape-Perception to Explicit Judg- of Psychophysiology, II Sensation. New
ments of Inclination," Arch. Psychol., York: Van Nostrand.
296, pp. 94.
Troland, L.T. (1932), The Principles of
Stevens, S.S. (1946), "On the Theory of P sychophys iology, III C erebration and
Scales of Measurement, Science, 103, Action. New York: Van Nostrand.
pp. 677-680.
Stevens, S.S. (1935), "The Operational Vernon, M.D. (1937), "The Perception of
I)efinition of Psychological Concepts," Distance," Bru. J. Psychol., 28, pp.
Psychol. Rev., 42, pp. S 17-527. 1-11, 115-149.
Studies in Visual Acuity (1948), Staff, Vernon, M.D. (1937), Visual Perception.
Personnel Research Section, the Adjutant Cambridge: Cambridge Univ. Press.
General's Office (PRS Report No. 742)
Washington: U.S. Gov't. Printing Office.
'116. Ware, W.R. (1900), Modern Perspective.
New York: MacMillan.
i 17. Weisenberg, T., and McBride, K.E.

(1935), Aphasia. New York: The Corn-
monwealth Fund.
Ternus, J. "Experimentelle
(1926), i 18. Wertheimer, M. (1912), "Experimentelle
Untersuchungen Uber Phnomena1e Identi- Studien liber das Sehen von Bewegung,"
tät," Psychol. Forsch, 7, pp. 81-136. Zsch.f. Psychol. 61, pp. 161-265.
Thompson, D.W. (1942), Growth and Forni. Wertheimer, M. (1923), "Untersuchungen
New York: Macmillan. zur Lehre von der Gestalt II," Psychol.
Forsch., 4, pp. 30 1-350.
Thouless, R.H. (1931), "Phenomenal
Regression to the Real Object I," 1?rit. Wiener, N. (1949), Cybernetics. New
J. Psychol., 21, pp. 339-359. Also, York: Wiley.
"Phenomenal Regression to the Real
Object Il," lint. J. PsychoL, 22, pp. Woodworth, (1938), Experimental
R.S.
1-30. Psychology. New York: Henry Holt.
Index
Abstract paintings, 196, 200, 202, 203 Bullis, G., 18

Accommodation, sensations of, 20; 21, degree Buswell, 155
of, 72; as cue for distance, 111; reflex of,
111; 137, 141, 157, 177, 222 Carmichael, Dr. Leonard, 65
Acuity, 9, 108, 112, relation of to texture, 110; Carr, 100
dependence on illumination, 113; 175
Carter, 209
Adjacent Order, of retinal image 9; 134, 154 Cartesian, coordinates, 149; geometry, 19
Aerial perspective, 72, 114, 116, 137, 14F Cassirer, 153, 193
After images, 31, 32, 58, of color, 133; 147, Cast shadow, 94
i 59 Cataract patients, 216, 220
Aircraft, cues for landing, 128, 129 Child, C. M., 73
Aircraft recognition, 91, 202, 219 Circle of confusion, 112
Air light, 5 Color, sensation, 15, 17; extensity, 17; con-
Air theory, of space, 6; 7 stancy, 166; film, 166; surface, 166; memory,
Alexia, 205 209
Alley experiments, 190 Conant, 7
Ames, 104 Concurrent stimuli, 108
Aniseikonia, 106 Conscious present, 158, 159
Aniseikonic glasses, 104, 105 Constancy, hypothesis, 62; of perceived
Aperture colors, 5 objects, 169; index of, 171; of size at dis-
Asch, 150 tance, 183; index of, 186
Astigmatism, 111 Continuity, of outline, 143
Attached shadow, 94 Contour, depth of, 106
Autokinctic phenomenon, 127 Convergence, 20, 21, 30, degree of, 72; as cue
Automobile driving, 128, 1i3, 135 for distance, ill; 137, 157, 177, 222
Aviation, night, 150 Correspondence, projective, 47, 48, 53; psycho-
Aviation psychology, 59, 107, 127, 131, 180 physical, shift in, 195, 196; 153, psycho-
physical, 57; 61, 62, 75, 104, 165
Bartlett, 210 Courant, 153
Bartley, 65, 73, 113 Crossed disparity, 104
Berkeley, Bishop, 19, 20, 145, 223, 225 Cues, theory of, 19; for depth, 21, 22, 59; 69,
Binocular, cues for depth, 21, 22; disparity, 71, secondary, 72; primary, 72; 90, 103, 111,
20, 21, 72, 100, 106, 107, 177; fusion, 103; for aircraft landing, 128, 129; for depth,
parallax, 117, 118; perspective, 139; 137, 138; conflicting, 149; reciprocal, 149;
vision, importance of, 107, 108 for depth, 150; reliability of for space per-
Blindness, psychic, 204, 205 ception, 150; shading, 168; for depth, 171;
Blind spot, 48, 49 177,for behavior, 214, 215; for behavior, 221
Blur, gradient of, 112; perspective, 141
Boring, 14, 19, 75 Dartmouth Eye Institute, 104, 106
Brain localization, 205 Da Vinci, Leonardo, 114
Brightness, as a cue for distance, 137; dis- l)epth, in vision, 15, of field, 112; test for
crimination, i 10 perception of, 130
Bruner, 209 Depth at a contour, 93, 106, 107, 137, 138,
Brunswik, 29, 150 141, 165, 177, 178, 221, 228
Buhier, 5 Depth cues, 21, 22, 137, 138, 150, 171
Depth perception, 6, 7, test for 107, 108 Graybiel, 136
Depth shape, 34, 54, 91,94; Ground theory of space, 6, 7, 59, 60
Disorientation, of man in visual space, 5, 6 Haunted swing, 150
Discrimination experiment, 221 heider, 63
Discrirnirative judgment, 221 Helmholtz, Hermann Ludwig von, 19, 21, 44,
Distance, in space, 15 59, 118, 119, 145, 147, 148, 157
Distant stimulus, 63 Hemianopsia, 205
Dodge, 155 Henneman, R. H., 168, 183
Dollard, 215 Homogeneous stimulation, 4, 5, 6, 53, 64, 108
Double images, 72, 73 Horizontal and vertical axes, 148, 150
1)uke Elder, 116 Horopter, 60, lOO
Duncker, 209 Hull, 215
Edge, impressions on visual world, 8 Identification experiment, 221
Ego, phenomenal as explained by Mach 27, 28 Identifying response, 202
Eidetic image, 159 Illusory perceptions, 14
Eisenstein, 160 Image, of imagination, 158; eidetic, 159; pri-
Elementary sensations, 22, 107, 108 mary memory, 158, 159, 160; of the nose, 228
Emmert's law, 32 Indentation, 95, 104, of modelling, 143
Empiricism, 14, 15, 16, 17, 24, 206, 212, 216 Ink blots, and stimulation, 211
Entoptic phenomena, 31, 32 Inhomogeneity, 9
Epistemology, 13, 24, 212 Insihts, and learning, 216
Equilibrium, and gravity, 33 Instinct, theory of, 206; and discriminative
Euclid, 36, 42, 54, 8, (Euclidean geometry), behavior, 207, 220
122, 188, 190; (non-Euclidian geometry), Invariant properties, 153-154
122, 189; 188, 189 Ives grating, 109
Euclidean space, 14, 166
Extensity, of space, 15, 16; in coior, 17; 169 James, William, 15, 210
Eye, defects as an optical instrument, 116 J ohnson, Samuel, 145
Eye hand co-ordination, 225 Katz, 5, 169
Eye movements, pursuit, 32; 57, 58, effect on Kepes, 196, 203
retinal gradient of motion, 124; compensary, Klineberg, 206
136; and pictorial composition, 155, 156; Koffka, 9, 23, 25, 28, 29, 40, 63, 64, 127,
saccadic, inhibition of vision during, 146, 148, 150, 169, 171, 176, 191, 196
147; 155, 156. 159, 160
Landolt ring, 109
Facial expressions, 207 Langewiesche, 129
Farsightedness, 111 Lashley, 206, 207, 220, 221
Feedback mechanism, and stimulation, 225 Law, of Pragnanz, 210
Field theory, 23, 29, 74 Law of the Visual Angle, 83, 175
Figure ground phenomenon, 8, 196 Laws of sensory organization, 149, 191, 192
Film, color, 5, 75, 65, 166; editing, 159; mo n- Lewin, 229
tage, 160 Linear perspective, 35, 36, 69, 71, 82, 85, 86,
Foreshortening, 83, 172, 183 138, 139, 175
Form, or shape in vision, 15; or shape in two Linear projections, 190
dimensions, 16, 17; transposability of, 18, Linguistic meaning, 202
19; 99, (Form perception), past experience Literal perception, 9, 10, 210, 211
in, 192; (Form quality), 19, 22; 151 Local sign theory, 16, 64, 146, 147
Frame of reference, geographical, 230 Localization, of points in space, 15, theory
Freud, 199 of, 133
Frontal surfaces, 66, 70, 76, 84 Location, of points, 15; of after-images, 31, 32
Locke, John, 12, 199
Ganzfeld, 5 Locomotion, point of aim for, 123, 124; 127,
Geometrical space, 15, 188
Gesell, A, 18 128, 131, perception of, 135; passive, 135,
Gestalt Theory, 2, 9, 10, 19, 22, 25, 39, 57, 136; 153, passive, 154; and retinal stimuli,
62, 74, 100, 151, 158, 191, 196, 216 224; and perception, 226; study of, 229
Gibson, E., 202, 219 Locomotor ataxia, 224
Longitudinal surfaces, 66, 70, 76, 84
Goodman, 209 Luneberg, Rudolf, 189, 190
Gradient, definition of, 73; of blur, 112; Mach, Ernst, 27, 113
physiological, 73; of density of texture, 78, Mach rings, 114
80, 172 MacLeod, 65, 73, 168, 169
Gravity, influence on perceptions, 6; 33
Macrogradient, 73 constancy in, 33, 34; 43, 44, test for depth,
Material world, perception of, 9 107, 108; depth, test for, 130; of accelera-
Maze learning, 229 tion, 134, 135; and sensation, 138; form,
McDougall, 206 past experience in, 192; meaningful, 198,
Meaning, psychology of, 198, 199, 200, 202; 199; literaI, 210, 211; schematic, 211, 212;
emotional, 199; social, 199; (use-meaning) space, motor theory of, 223, 224.
200, 202; linguistic, 202; context theory of, Perceptual constancy, 23
203, lapse of, 204; innate, 205, 206. Perspective, aerial, 72; 114, 116, 137, 141
Meanings, spatial, 200, 202 Perspective, linear, 70, 82, 85, 86, 17, 18,
Memory, immediate, 57; primary, 158, 159; 165 139, 175
Memory color, 209 Perspective, binocular, 139
Memory trace, 159 Perspective geometry, 82
Metzger, 5, 52 Perspective illusion, 181
Michotte, A., 135 Perspective, motion, 124, 140, 141
Microgradients, 73, 113, 114 Perspective, of blur, 141
Microstructure, 5, 65, 113, 172 Perspective, size, 138
Miller, 215 Perspective, texture, 138
Modelling, perception of, 92; and grades of Pfaffmann, 129
illumination, 9/4; 143 Phenomenal ego, 225, 226, stimuli for, 226
Monocular cues for depth, 21 Phi phenomenon, 134, 161
Monocular field, 27, 28 Physiological gradients, 73
\lorris, 202 Pictorial vision, 181
Motion, types of retinal, lU Picture plane, 54, 55, relation to retinal pro-
Motion parallax, 71, 124, 117, ilS jettiOfl, 79, 104, 122, 153, 218
Motion perspective, I 19, 120, 124, 140, 141 Pinhole camera, experiment of, 47
Motion pictures, 119, 130, ß4, sequence and Plato, 152
scene, 159 Point of aim, 127. 128
Motion, theory of relativity of, 127 Point sensations, 15, 22, 59, 1 14, 133, 191
Movement, pursuit, 132; saccadic, 132; scro- Post, Wiley, 108
boscopic, 134 Posture, relation to vision, 32, 33; and per-
Movements, eye, 136, 146, 147 Ception, 226; 228
Mowrcr, 149, 150 Postural equilibrium, 150
Muscle sense, 134, I5, 224 Postural stimulation, 149, 150
Narrow angle lens, 157 Pragnanz, IcLW of, 210
Nativism, 14, lS, 24, 216 Primary cues for depth, 21, 72
Nearsightedness, ill Primary memory, 158, 159, image, 160, 165
Negatìve after image of motion, 13. Primitive man, vision of, 198
Negative after effect of tilt and curvature, Projected form, definition of, iô; 188, 190,
195, 196
shape, 34, 35; and transformation, 169; 171,
Newton, Sir Isaac, 14, 15, 189 174
Nonsense forms, 202, 210 Projcctive correspondence, 47, 48, 534, 153,
Nose, image of, 106 geometry, 153
Nystagmus, 136 Projective transformation, I 53
Protuberance, 95, 104
Oculogyral illusion, 136 Protuberances, of modelling, 143
Ogle, 105 Proximal stimulus, 63
Ophthalmology, 147, 148 Psychic blindness, 204, 205
Optic nerve, anatomy of, 49, 50 Psychomotor behavior, 135
Optical range-finder, 177 psychophysical experimentation 9, corres-
Order, definition of, 64; successive, 134; pondence, to visual perception, 8, 25, 51,
adjacent, 134, 151, 154; successive, 154; 57, 61, 62, 5; shift in, 75; 104, 165, 195,
161 196; shift in, 195, 196
Ordinal stimulation, 56, 63, 64, 152, 168, 216 Pursuit movements, 126, 127, 132
Organization, phys iological, 25 ; sensory,
215 Ratoosh, 143
Orientation, to physical environment, 229 Reaction, identifying, 220
Outlines, 86 Reference - axes, 149
Relativity of motion, theory of, 126, 165, 189
Painting, abstract, 196, 200, 20 Relief, perception of, 92
Panoramic vision, 108 Resolution, definition of, I 12
Peephole vision, 178 Response, identifying, 202; smiling in the in-
Perception, relation to sensation, 12, 13; of fant, 207; smiling in infants, 220; identi-
external world, 13, first visual, 17, 18; fying, learned, 221
Retina, description of, 46, 56 Stimulus variable, relation to visual space,
Retinal image, differentiation of, 8, 9; 33, 46, 8, 51. 52.
50, 52, 57, 67, 61, 62, 79, 117, 118, motion, Stravrianos, 172
types of, 131 Stroboscopic movement, 134, 161
Retinal receptors, density of, 48, 49; distribu- Successive order, 134, 161
tion of, 55 Surface - color, 6, 166
Reversal of relief, 99 Surface, impressions on visual world, 8
Reversible figures, and stimulation, 211 Superposition, 71, 137, 142, 177; type of
Riesen, Dr. A., 221 depth-perception, 228
Robbins, 153 Symbols, 164, 204, 205, 212, in meaning, 199
Rubin, 39
Russell, 220
Tachistoscope, 9, 211
Saccadic eye movements, 29, 30, 126, 132, Television camera, likeness to retina, 48; 56
146, 147, 155, 159, 160 Ternus, 57
Scanning in visual world, 29, 30, 31, 126 Test for depth perception, 130
Schematic perception, 9, 211, 212 Texture, compression of, 173; gradient of
Schooler, 209 density of, 66, 67; relation of to acuity, 110;
Secondary cues for depth, 21, 72 visual, elements and gaps of, 80, visual, 81
Selective perception, 9 Texture - perspective, 138
Senders, 113 Theory, context (of meanìng), 203; instinct,
Sensation, relation to perception, 12, 13; 43, 206; motor, of space perception, 223, 224
44; and perception, 128; color, 15, 16; 15, Thompson, D'Arcy, 153
16; 16, 17 Thouless, 168
Sensations, elementary, 19, 22, 107, 108; in- Titchener, 203
nate, 15, 163, 171, 218; point, 15, 22, 133, Topology, 153
191; of accommodations, 19 Transfer of learning, 219
Sensory organization, 22, 23, 215, 187; laws Transformation, 104, of images, 132; of retinal
of, 191, 192; external forces of, 196 pattern, 152; projective, 153, 169; groups,
Shading, 72, 94, 137, 143, 144; as a cue for 193; projective, 153; 191
depth, 168 Transposability, of form, 18, 19 , 55, 56, 151;
Shape,or form in vision, 15; or form in two di- of retinal image, 153
mensions, 16, 17; constancy of perceived Triangulation, 19
objects, 169; apparent, 171, 172, projected, Troland, 100
174. Unconscious interence, 19, 119, 145, 146
Sìgns,ìn meanìng, 199; local, 222
Uncrossed disparity, 104
Size, constancy of, 174 Use - meanings, 199, 200, 202
Size - perspective, 138
Snellen, 109 Vernier acuity, 109
Social meanings, ¡99 Vernon, 212
Social norms, 212; relation to perception, 211 Vertical and horizontal axes, 148, 149, 150
Social stereotypes, 210 Visual acuity, 108, 109, 110, 111, 112, 175
Somaesthesia, 226, 227 Visual agnosia, 205
Space, geometrical, 183, 188; sensation of, 8; Visual field, boundaries of, 27, deformation o
perception, analysis of, 4; perception, motor 40; boundaries of, 45, 46, 155, 226, 228
theQry of, 223, 224 Visual kinesthesis, 124, 224
Spatial meanings, 200, 202 Visual motion, stimulus for, 160, 161
Spatial world, perception of, 9 Visual perception, evolution of, 60
Spectroscope, 166 Visual texture, 52, 65; elements and gaps o
Spitz, 207, 220 80, 81
Stereoscope, purpose of, 6, 20, 104, 107 Visual world, properties of, 3
Stereoscopic acuity, 110 Volkmann, John, 181
Stereoscopic impressions, 4, photographs, 107 Von Senden, 217, 219
Stereotypes, social, 210; 211
Stevens, 75
Stimulation, homogeneous, 53; ordinal, 63, 64; Wertheimer, 149, 150, 160, 161, 196
homogeneous, 64, 108, impoverished 151, Witkin, 150
211, ordinal, 216 Wolfe, 207
Stimuli, concurrent, 108 Woodworth, 202
Stimulus, method of impoverishing, 9, 10; Word blindness, 205
definition of, 63; distant, 63; proximal, 63; Wundt, 15
definition of, 151, 152, 215; for a psycho- Wheatstone, 20, 107, 111
physical experiment, 215 Wide angle lens, 157
Stimulus generalization, 219 Zeno, 134
Stimulus - objects, 215, 216

The Perception of The Visual World-GibsonJJ PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Perception of The Visual World-GibsonJJ PDF

Uploaded by

Copyright:

Available Formats

THE I'ERCEPTION OF THE

KOUGHTON MIFFLIN COMPANY BOSTON

All rights reserved including the right to re-

THE RIVERSrDE PRESS

Artists and philosophers as well as physicists, physiologists, and

scientists who are interested in learning about the nature of man's

The principal subject of this book is the visual perception of

grateful likewise to a number of other friends and colleagues who have

Many of the pictures are intended to give

George Berkeley, Bishop of

The Theoretical Approach . . . The Initial Hy.

FIGURE 1. A Scene for Analysis

There is another problem with which we objects. accepted terminology, our

FIGURE 2. The Look of the World from the Air

There is a naive theory of perception to ferentiated visual image is variously

The Distinction between Sensation and Perception

j The traditional explanation of vision is

Whenthis doctrine of perception is ap British philosophers in the eighteenth

Fgeometry.4 Next there

(ßy Life photographer 1/erben Gehr. Copyright Time, Inc.)

FIGURE 3. An Attempt to Represent the Vision of an Infant

consisted of white on black or black on line represented on the retina as a single

He added co it the idea that sensations of

respectable unreality which clung to the Summary

The Bounded Visual Field . . . . The Gradient of

FiGURE 5. The Monocular Visual Field of Ernst Mach

Compensatory movement of the eyes ¡n the

(By Life photographer herbert Gehr. Copyright Time, ¡nc.

FIGURE 8. Convergence of Parallel Lines to o Vanishing Point

(From W. R. Ware, Modern Perspective (IV. Y., Macmillan, 1900)

FIGURE 9. Two Scenes in Perspective

as a scene in perspective. He saw it as not attend to the spaces between the

V/'FIGURE lo. Inattention to Interspaces in Ordinary Perception

FIGURE 1. One Object in Front of Another

projected shapes like geometrical forms of locomotion he is at least moving his

FIGURE U. Objects Seen Behind Others

FIGURE 13. Surfaces in Three Dimensions

The Sequence of Events in Vision . . . . The

If we are to understand visual percep- nature of the optical process or at least

SINGLE N 'N SINGLE

FIGURE 14. The Theoretical Reflecting

FIGURE 15. The Optical Projection of a Room

CHEEK bONE SCLERA

FIGURE 17. A Diagram of the Human Eye

FIGURE 18. The Sequence of Transforma-

two parts of the image. The image, there-

image is unquestionably a projection rather dimensional surface.

perception in psychological laboratories imae is an event in the light-flux of the

ence to the world, are pure geometrical I'*"*"

; have boundaries and it is more nearly movement. The stimulus complex to

Abstract Space and the World of the Flier ....

'The theories of space-perception which

minate distance. the stimulus correlate of a visual surface.

............................................... ., %60104%. slime

FIGURE 22. Grades of Texture

FIGURE 23. Gradients of Natural Texture and

o430 Ato "I.Mb

will be described in subsequent chapters.

FIGURE 26. A Gradient of Density in Four Differ-

visual acuity, and the elements of a

over the width of a number of cells.

microscopic shadow-edge as it falls gradient of the density of texture would be one

border within the retinal image. He is thinking considerable of the retina. A

ence to the world, are pure geometrical I'""