You are on page 1of 164


Ro^Jtrb fl



Bradford Books

Edward C. T. Walker, Editor. Explorations in THE BIOLOGY OF LANGUAGE.

Daniel C. Dennett.BRAINSTORMS. ry79.
MIND. r98o.
John Haugeland,Editor. MIND DESIGN. r98r.
Ned Block, Editor. IMAGERY. r98r.
Roger N. Shepardand Lynn A. Cooper. MENTAL IMAGES AND THEIR
Owen D. Flanagan.THE SCIENCE OF MIND. 1984.
Robert N. Brandon and Richard M. Burian, Editors. GENES, ORGANISMS,
Elliot Sober. THE NATURE OF SELECTION. 1984.


Valentino Braitenberg

The MIT Press


Sixth printing,1998
First MIT Presspaperback edition, 1986
@ 1984 by The MassachusettsInstitute of Technology
All rights reserved.No part of this book may be reproduced
in any form by any electronic or mechanicalmeans(including
photocopying, recording or information storageand retrieval)
without permissionin writing from the publisher.
Jacket artwork and portfolio of drawings
by Maciek Albrecht:
Copyright @ 1984 by Maciek Albrecht
All rights reserved.This arnvork may not be
reproducedin any form or by any means,electronic
or mechanical,including photocopying or by any
other means,without permission in writing from
Maciek Albrecht.
Technicaldrawings by Ladina Ribi and Claudia
This book was set in Linotron eoz Sabonby Achorn
Graphic Servicesand printed and bound
in the United Statesof America
Library of CongressCataloging in Publication Data
Vehicles,experimentsin syntheticpsychology.
"Bradford books"-Series t.p.
Bibliography: p.
r. Neurophysiology. z. Psychology,Physiological.
r. Tide.

IDNLM: r. Psychophysiology.WL ro3 B8I48VJ

QP356.B7a rg84 rSz 84-9727
ISBN 0-262{2208-7 (hard)
0 - 2 6 2 - 5 2 1 l 2 - l( p a p e r )

Alle mie creature


Foreword by Michael A. Arbib


Introduction: Let the Problem of the Mind Dissolve

in Your Mind r
Vehicle r

GemingAround 3

Vehicle z

Fear and Aggression 6

Vehicle 3

Love ro

Vehicle 4

Values and SpecialTastes r5

Vehicle 5


Vehicle 6

Selection,the ImpersonalEngineer z6

Vehicle 7

Concepts 29

Vehicle 8

Space,Things, and Movements jj

Vehicle 9

Shapes 4j

Vehicle ro

Getting Ideas to

Vehicle r r

Rules and Regularities t j

Vehicle rz

Trains of Thought 6z


Vehicle 13 Foresight 70
Vehicle 14 Egotism and Optimism 8o


Biological Notes on the Vehicles gt

The uirtues of crossedconnections gJ
The compound eye of the fly: reconstructionof
continuity in the uisual representation roo
Olfactory orientation: control of behauiorby
sytnrnetricalreins ro2
Orientation and obiect fixation in flies rot
McCulloch-Pins neuronsand real neurons ro8
Euolution rr3
Memory rr4
In segrchof an engram: the anatoffiy of me?nory 116
Maps and their use rzo
Shapes.The motphemesof uisualperception r2j
An inborn categoryof acousticform perception rjj
Structure of the cerebral cortex r3j
Cell assemblies:embodimentsof ideas r4o
Thresbold control and the pump of
thoughts r4j
References r45


Valentino Braitenbergis a cybernetician,a neuroanatomist,and a

musician.He seeksto understandhow the beautiful structuresof
the brain constitutea machinethat can enableus to exhibit such
skilled behavioras that involved in playing music. Sincethe early
1.950s,I haveturned to Valentino for detailedneuroanatomyand
for lively essaysthat cut aw^y the technicaldetailsto illuminate
the key issuesof what we may call cyberneticsor artificial intelligenceor cognitivescience.
One of the most exciting of theseessayshad the most formidable of titles: "Taxis, Kinesisand Decussation,"publishedin
1965. Taxis is the reflex-orientedmovementof a freely moving
organismin relation to a sourceof stimulation; kinesis,by contrast, is movementthat lacks orientation but dependson the intensityof stimulation; and a decussationis a band of nerve fibers
that connectsone half of the body to the oppositehalf of the
brain. The title was forbidding, the essaywas delightful. By designinglittle vehiclesthat moved around in responseto smell and
vision, Braitenberggavehis readersvivid insightsinto how the
brain might have evolvedso that olfactory input goesto the


sameside of the brain while vision, touch, and hearingsendtheir

input to the oppositeside of the brain.
Having sharedthis paper with friendsand studentsover the
years, I was delightedto hear from Valentino)at a workshop in
1983, that it had provided the nucleusfor this book. Vehicles:
Experiments in Synthetic Psychologyis fun to read, and this fun
is heightenedby the incredibleillustrarionsof Maciek Albrecht.
But it is seriousfun and will help many people,specialistand
layman alike, gain broad insightsinto the ways in which intelligenceevolvedto guide interaction with a complex world.
Michael A. Arbib
Amh erst, Massachusetts



Let tbe Problemof the Mind

Dissoluein Your Mind

This is an exercisein fictional science,or sciencefiction, if

you like that better. Not for amusement:sciencefiction in the service of science.Or iust science,if you agreethat fiction is part of it,
always was, and always will be as long as our brains are only
minusculefragmentsof the universe,much too small to hold all the
facts of the world but not too idle to speculateabout them.
I havebeendealingfor many yearswith certain structureswithin
animal brainsthat seemedto be interpretableas piecesof computing machinerybecauseof their simplicity and/or regularity. Much
of this work is only interestingif you are yourselfinvolved in it. At
times, though, in the back of my mind, while I was counting fibers
in the visual gangliaof the fly or synapses
in the cerebralcortex of
the mouse,I felt knots untie, distinctionsdissolve,difficultiesdisappear,difficultiesI had experiencedmuchearlierwhen I still held my
first naivephilosophicalapproachto the problem of the mind. This
processof purificationhas been,over the years,a delightful experience.The text I want you to read is designedto conveysomeof this

to you, if you arepreparedto follow me not througha world of real
brains but through a toy world that we will createtogether.
We will talk only about machineswith very simpleinternal structures, too simplein fact to be interestingfrom the point of view of
mechanicalor electricalengineering.Interestarises,rather, when
we look at thesemachinesor vehiclesas if they were animalsin a
natural environment.We will be tempted,then, to usepsychological languagein describingtheir behavior.And yet we know very
well that there is nothing in thesevehiclesthat we havenor put in
ourselves.This will be an interestingeducationalgame.
Our vehiclesmay move in water by jet propulsion.Or you may
prefer to imaginethem moving somewherebeirveengalaxies,with
negligible gravitational pull. Remember,however, that their jets
must expel matter in order to function at all, and this implies replenishmentof the food storeswithin the vehicles,which might be a
problem berweengalaxies.This suggestsvehiclesmoving on the
surfaceof the earth through an agriculturallandscapewhere they
have good support and can easily find the food or fuel they need.
(Indeed the first few chaptershere conjure up imagesof vehicles
swimming around in the water, while later what comesto mind are
little carts moving on hard surfaces.This is no accident,if the
evolution of vehiclesr to 14 in any way reflectsthe evolution of
animal species.)
It does not matter. Get used to a way of thinking in which the
hardware of the realizationof an idea is much lessimportant than
the idea itself. Norbert Wiener was emphaticabout this when he
formulated the title of his famous book: Cybernetics,or Control
and Communication in Animals and Machines.

Vehide 1

Getting Around

Vehicle r is equipp.d with one sensorand one motor

(figurer). The connectionis a very simpleone. The morethereis of
the quality to which the sensoris tuned, the fasterthe motor goes.
Let the quality be temperature and let the force exerted by the
motor be exactly proportionate to the absolutetemperature(the
temperatureabove zero degreesKelvin) measuredby the sensor.
The vehiclewill move, whereverit is (the absolutetemperatureis
nowhereequal to zero), in the direction in which it happensto be
pointing. It will slow down in cold regionsand speedup whereit is
Here we have introduceda bit of Aristotelianphysics.Aristotle,
like everybodyelsebetweenthis ancientGreekphilosopherand the
lessancient Italian physicist Galileo, thought that the speedof a
movingbody is proportionateto the forcethat drivesit. This is true
in most instances,namely when there is friction to slow down the
vehicle.Normally friction will seeto it that the velocity becomes
zero in the absenceof any force,that it will stay at a certainsmall
valuefor a certain small force, at a higher value for a higherforce,
and so forth.
Of course,as you all know, this is not true for heavenlybodies

Figure r
Vehicle r, the simplest vehicle.The speedof the motor (rectangularbox at
the tail end) is controlled by a sensor (half circle on a stalk, at the front
end). Motion is always forward, in the direction of the arrow, except for

I Getting Around

(especiallyif you don't investastronomicaltime in observingthem).

Their velocity is a complicatedresult of all the'forcesthat everhit
them. This is another reasonfor letting our vehiclesmove in water
or on the surfaceof the earth rather than in outer space.
In this Aristotelian world our vehiclenumber r may evencome
to rest. This will happen when it entersa cold region where the
forceexertedby its motor, beingproportionateto the temperature'
becomessmallerthan the frictional force.
Once you let friction come into the picture, other amazingthings
may happen.In outer spaceVehicle r would move on a straight
coursewith varying speed (the gravitationalpull of neighboring
galaxiesaveragesout to nothing). Not so on earth. The friction,
which is nothing but the sum of all the microscopicforcesthat arise
in a situation too messyto be analyzedin detail, may not be quite
symmetrical. As the vehicle pushes forward against frictional
forces,it will deviatefrom its course.In the long run it will be seen
to move in a complicatedtraiectory,curving one way or the other
without apparentgood reason.If it is very small, its motion will be
quite erratic, similar to "Brownian motionr" only with a certain
drive added.
Imagine,now, what you would think if you saw such a vehicle
you would say,and does
swimmingaround in a pond. It is restless,
not like warm water. But it is quite stupid,sinceit is not ableto turn
back to the nice cold spot it overshotin its restlessness.
you would say,it is euve, sinceyou havenever seena particle of
deadmatter move around quite like that.


Fear and Aggression

Vehicle z is generallysimilar to Vehicle r except that it

has two sensors,one on each side, and rwo motors, right and left
(figure z). You may think of it as beinga descendantof Vehicle r
through someincompleteprocessof biologicalreduplication:two
of the earlier brand stuck togethersideby side.Again, the more the
sensorsare excited,the faster the motors run.
Of courseyou noticeright awaythat we can make threekinds of
such vehicles,dependingon whether we connect(a) eachsensorto
the motor on the sameside, (b) each sensorto the motor on the
opposite side, or (c) both sensorsto both motors. We can immediatelydismisscase(c), for this is nothing but a somewhatmore
luxurious versionof Vehicle r. The differencebetween(a) and (b),
however, is very interesting.
Consider (a) first. This vehiclewill spendmore time in the places
where there is lessof the stuff that excitesits sensorsand will speed
up when it is exposedto higher concentrations.If the sourceof the
stuff (say, light in the caseof light sensors)is directly ahead,the
vehicle may hit the sourceunlessit is deflectedfrom its course.If
the sourceis to one side(figure3), one of the sensors,the one nearer
to the source, is excited more than the other. The corresponding

Figure z
otherwiselike Vehicler. The
Vehiclez, with two motorsandnvo sensors;
differ in a, b, and c.

Figure 3
Vehiclesza andzb in the vicinity of a source(circlewith raysemanating
from it). Vehiclezb orientstowardthe source,za awayfrom it.

9 | Fearany' Aggression
the vehiclewill turn
motor will work harder.And as a consequence
away from the source.
Now let us try the other schemeof sensory-motorconnections,
(b) in figure 3. No changeif the sourceis straight ahead.If it is to
one side,however,we notice a differencewith resPectto Vehicle za.
Vehicle zb will turn toward the sourceand eventuallyhit it. There
is no escaping:as long as zb staysin the vicinity of the source'no
matter how it stumblesand hesitates,it will hit the sourcefrontally
in the end. Only in the unlikely casethat a strongperturbationin its
coursemakesit turn exactly away from the source,and no further
perturbation occurs,can it escapeits fate.
Let Vehiclesza and zb move around in their world for a while
and watch them. Their charactersare quite opposite.Both DIsLIKE
sources.But za becomesrestlessin their vicinity and tendsto avoid
them, escapinguntil it safelyreachesa placewhere the influenceof
the sourceis scarcelyfelt. Vehicleza is a cowARD' you would say.
Not so Vehiclezb. It, too, is excitedby the presenceof sources,but
resolutelyturns toward them and hits them with high velocity,asif
it wanted to destroythem. Vehicle zb is AccREsslvn,obviously.



The violenceof Vehicle zb, no lessthan the cowardiceof

its companion za, are traits that call for improvement.There is
something very crude about a vehicle that can only be excited by
the things it smells(or seesor feelsor hears)and knows no soothing
or relaxing stimuli. \UUhat
comesto mind is to introduce someinhibition in the connectionsbenveen the sensorsand the motors,
switching the sign of the influence from positive to negative.This
will let the motor slow down when the correspondingsensoris
activated.Again we can make two variants,one with straight and
one with crossedconnections(figure4). Both will slow down in the
presenceof a strong stimulus and racewherethe stimulusis weak.
They will therefore spend more time in the vicinity of the source
than away from it. They will actually cometo rest in the immediate
vicinity of the source.
But here we notice a difference betweenthe vehiclewith straight
connectionsand the one with crossedconnections.Approaching
the source, the first (figure 4a) will orient toward iq sinceon an
oblique coursethe sensornearerto the sourcewill slow down the
motor on the same side, producing a turn toward that side.
The vehicle with straight connections will come to rest facing the


Figurc 4
Vehicle3, with inhibitoryinlluenceof the sensors
on the motorc.

source.The vehiclewith crossedconnecrions(figure4b) for analogous reasonswill come to rest facing away from the sourceand
may not staytherevery long, sincea slightperturbationcould cause
it to drift away from the source.This would lessenthe source's
inhibitory influence,causingthe vehicleto speedup moreand more
as it gets away.
You will haveno difficulty giving namesto this sort of behavior.
ThesevehiclesLIKEthe source,you will say, but in differentways.
Vehicle 3a LovEs it in a permanentway, stayingcloseby in quiet
admiration from the time it spots the sourceto a[ future time.
Vehicle3b, on the other hand, is an ExpLoRER.
It likes the nearby
sourceall right, but keepsan eyeopen for other, perhapsstronger
sources,which it will sail to, givena chance,in order to find a more
permanentand gratifying appeasement.
But this is not yet the full developmenrof Vehiclet. Ifle are now
readyto make a more completemodelusingall the behavioraltraits
at our disposal.Call it Vehicle 3c. \UUe
give it not just one pair of
sensorsbut four pairs, tuned to different qualitiesof the environment, say light, temperature,oxygenconcentration,and amount of
organic matter (figure 5). Now we connectthe first pair to the
motors with uncrossedexcitatoryconnections,asin Vehicleza, the
secondpair with crossedexcitatoryconnections,as in Vehiclezb,
and the third and the fourth pairs with inhibitory connections,
crossedand uncrossed,as in Vehicles3b and 3a.
This is now'a vehiclewith really interestingbehavior.It dislikes
high temperature,turns awayfrom hot places,and at the sametime
seemsto dislike light bulbs with evengreaterpassion,sinceit turns
toward them and destroysthem. on the other hand it definitely
seemsto prefer a well-oxygenatedenvironmentand one containing
many organic molecules,sinceit spendsmuch of its time in such
places.But it is in the habit of moving elsewherewhenthe supplyof
eitherorganicmatter or (especially)
oxygenis low. You cannothelp
admitting that Vehicle 3c has a systemof veruns, and, come to

Figure 5
A multisensorial
vehicleof brand3c.

think of it, KNovLEDcn,since some of the habits it has, like deasif the vehicle
stroying light bulbs,may look quite knowledgeable,
knows that light bulbs tend to heat up the environmentand consequently make it uncomfortableto live in. It alsolooks asif it knows
about the possibility of making energyout of oxygen and organic
matter becauseit prefersplaceswhere thesetwo commoditiesare
But, you will say,this is ridiculous: knowledgeimplies a flow of
information from the environment into a living being or at least
into somethinglike a living being.Therewas no suchtransmission
of information here.We were iust playingwith sensors,motors, and
connections:the propertiesthat happenedto emergemay look like
knowledgebut really are not. lU?eshouldbe carefulwith suchwords.
You are right. I[e will explain in a later chapter (on Vehicle 5)
how knowledge may enter a systemof connections.And we will
introduce an alternativeway of incorporatingknowledgeinto the
systemin our chapteron Vehicle7. In any case,onceknowledgeis
incorporated,the resultingvehiclemay look and behavequite like
our Vehicle 7c.
Meanwhile I invite you to considerthe enormouswealth of different propertiesthat we may give Vehicle3c by choosingvarious
sensorsand various combinationsof crossedand uncrossed,excitatory and inhibitory, connections.
If you considerthe possibilityof strongandweak influencesfrom
the sensorsto the motors, you realizethat the variety becomeseven
greater.The vehiclemay not care much about light but care very
much about temperature.Its senseof smellmay be much keenerfor
organic matter than it is for oxygenor viceversa.And theremay be
many more than just four pairs of sensorsand four sensoryqualities: the vehiclesmay be equipped with all sorts of shrewd detectors of energy and of chemicals.But this is best discussedin
connectionwith a new idea incorporatedin the vehiclesof the next


Valuesand SpecialTastes

We are now in a position to createa new brand of vehicle,

starting from all the varieties of Vehicl, j, by working on the
connectionsbetweensensorsand motors. They were,up to now, of
two very simplekinds: the more the sensorwas excited,the faster
the correspondingmotor ran, or, alternatively,the more the sensor
was excited, the slower the moto[ ran. !7e did not carg"whatthe
rules of the dependencewere, as long as they were of the nature
"the more, the more" or "the more, thg less." The vast classof
mathematicalfunctionsdescribingsuch dependences
is sometimes
called monotonic. Obviously, there is something very simpleminded about creaturesgovernedby such unconditionedlikes or
dislikes,and we can easilyseehow such the-more-the-merrierbehavior could lead to disaster.Think what happensin the caseof a
tendencyto follow downhill slopes!
Let us considerthe following improvement.The activation of a
certain sensorwill make the correspondingmotor run faster, but
only up to a point, where the speedof the motor reachesa maximum. Beyond this point, if the sensor is activated even more
strongly,the speedwill decreaseagain (figure6). The samesort of
with a maximum efficiencyat a certainlevel of sensor

16 | V E HI CL E +

Figure 5
A nonlinear dependence of the speed of the motor V on the intensity of
stimulation I, with a maximum for a certain intensity.

activation, can be engineeredfor the inhibitory connectionsbe'We

tweensensorand motor.
may setthe maximum efficiencyof the
varioussensorsat any levelwe choose,and we may evenplay with
having more than one maximum. Any vehicleconstructed according to this prescription we will assignto a new
brand, labeled 4a.Of course,if you like, you can keepsomeof the
connectionsof the old monotonic type and mix them with the
nonmonotoniconesin everypossiblecombination.
You will have a hard time imagining the variety of behavior
displayedby the vehiclesof brand 4a.A 4a vehiclemight navigate
toward a source(as Vehicle zb would) and then turn away when
the stimulus becomesstrong, circle back and then turn away over
and over again,perhaps describinga traiectory in the form of a
figure eight. Or it might orbit around the sourceat a fixed distance,
like a satellitearound the earth, its coursebeing correctedtoward
the sourceby a weaker stimulus and away from the sourceby a
stronger stimulus, dependingon whether the stimulus intensityis

17 | Valuesand SpecialTastes
on one side or the other of the maximum describingthe sensorymotor dependence(figure Z). Vehicle 4a might like one sort of
stimuluswhen it is weak but not when it is too strong; it might like
anotherstimulusbetter the strongerit becomes.It might turn away
from a weak smell and destroythe sourceof a strong one. It might
visit in alternationa sourceof smelland a sourceof sound,turning
away from both with a changeof temperature.
Watchingvehiclesof brand 4ain a landscapeof sources,you will
be delightedby their complicatedtraiectories.And I am sure you
will feel that their motives and tastesare much too varied and
intricateto be understoodby the observer.Thesevehicles,you will
say, are governedby rNsrtNcrs of various sorts and, alas,we iust
don't know how Nature managesto embody instinctsinto a piece
of brain.
You forget, of course, that we have ourselvesdesignedthese
But instincts are a lowly sort of behavior anyway. We can do
better.Let us improve on type 4aby adding a new sort of connection benrveensensorsand motors. This time the influenceof the
sensoron the motor is no longersmooth; there are definitebreaks.
There might be a range of intensitiesof sensorystimulation for
which the motor is not activatedat all and then, under stronger(o,^lil|,1
stimuli,the motors are running at full speed.Or else,theremight be .rF ,^1.,*ft
smoothchangesof motor activationfor certain ranges,with abrupt
changesin between.A very lifelike paffern would be: no activation
up to a threshold value of the stimulus, and increasingactivation
beyondthe threshold,startingwith a certainfixed minimum (figure
8). You are by now experiencedin the art of creativeinventionand
will haveno difficulty dreamingup more schemesof this sort.
In a way thesenew vehicles,which we call 4b, are alreadycontainedin the vast classof vehicles4?, sinceabruptnessof behavior
can of coursebe simulatedwith any degreeof approximation by
that are in reality, mathematicallyspeakfunctional dependences




-\r( s . ) - \








l :Ai



/ -a.


\\_/ ----_


Figure 7
Traiectoriesof vehiclesof brand4a aroundor benveen


19 | Valuesand SpecialTastes

Figure 8
Variousbizarrekindsof dependence
of the speedof the motor (ordinate)
on the intensityof stimulation(abscissa)
in Vehicle4b.
ing, continuous.Moreover, if friction plays a role, as we have al'
readydecidedit should,thresholdsin motor activationwould ensue
naturally: the vehiclewill staft moving only when the force exerted
by.the motor exceedsa certain value, sufficientto overcomethe
initial friction.
Whatever their origin, thresholds in some behavior patterns
make a lot of differencein the eyeof the observer.Thesecreatures,
the observerwould say, ponder over their DEcIsIoNs.
come close to them with a lure, it takes them some time to get
going. Yet once they havedecided,they can act quite quickly. Th.y
do indeed seemto act in a spontaneousway: none of this passive
being attracted one way or the other that was so obvious in the
vehiclesof the more lowly types.You would almostbe temptedto
say:where decisionsare beingmade,theremust be a wIn to make
them. Why not? For all we know, this is not the worst criterion for
establishingthe existenceof free will.



\ ,/

At this point we are ready.tomakea fundamentaldiscovery. We have gatheredevidencefor what I would like to call the
"law of uphill analysisand downhill invention." What I mean is
this. It is pleasurableand easy to createlittle machinesthat do
certain tricks. It is also quite easyto observethe full repertoireof
behavior of thesemachines-even if it goesbeyond what we had
originally planned,as it often does.But it is much more difficult to
start from the outside and to try to guessinternal structureiust
from the observationof behavior.It is actuallyimpossiblein theory
to determineexactly what the hidden mechanismis without openwith
ing the box, sincethere are alwaysmany differentmechanisms
identical behavior.Quite apart from this, analysisis more difficult
than invention in the sensein which, generally,induction takes
more time to perform than deduction: in induction one has to
searchfor the wxy, whereasin deductionone follows a straightforward path.
A psychologicalconsequenceof this is the following: when we
analyzea mechanism,we tend to overestimateits complexity. In the
uphill processof analysis,a givendegreeof complexityoffersmore
resistanceto the workings of our mind than it would if we encoun-

zr I Logic
teredit downhill, in the processof invention.We havealreadyseen
this happenwhen the observerof Vehicle4b conjecturedthat the
vehicledoessome thinking beforeit reachesa decision,suggesting
complicatedinternal processeswhere in reality there was nothing
but a thresholddevicewaiting for sufficientactivation.The patterns
of behaviordescribedin the vehiclesof type 4aundoubtedlysuggest
much more complicatedmachinerythan that which was actually
usedin designingthem.
We may now take pleasurein this and createsimple"brains" for
our vehicles,which will indeed(asexperienceshows)tax the mind
of eventhe most playful analyst.All we have to do is introduce
specialelements,calledthresholddevices,which will be eitherinterposedberweensensorsand motors or connectedto eachother in
complexesthat receivesomeinput from the sensorsand give some
output to the motors.
The individual thresholddeviceis of the simplestsort: it givesno
output if its input line carriesa signalbelow the threshold,and it
givesfull output beyond the threshold.We will also use anorher
variety giving output all the time unlessthe input carriesa signal
above the threshold. Each of thesedevicesis fitted with a knob
which may be turned to set the threshold,so that the input would
becomeeffectivewith one, two, or any specifiednumber of input
activationunits. (The word thresholdof courseimplies that, for a
given threshold value, any input strongerthan the one specified
would also be effective.)
We arenot limited to the typesof connectionsthrough which the
thresholddevicesactivateeachother.we can alsouseanotherkind,
call them "inhibitoryr" which counteractthe activationthat comes
from other sources(figure 9).
In order to make a brain out of thresholddevices,we may connectthem togetherone to one, or many to one, or one to many, or
manyto one and one to many, in whicheverway we like.'Sfhenyou
are designingbrains, it is important for you to know that in one of

z z I V E HI CL E 5

Figure 9
How threshold devices act on each other. Explanation of symbols: The
circles stand for threshold devices. The L-shaped fiber berween B and C
stands for inhibition; the penetrating fiber from A to C means activation.
Each active element contributes one unit of activation to the element
(threshold device) to which

it sends an activating connection. The

threshold device becomes active when the activation reaches at least the
threshold value indicated within the circle. An inhibitory connection from
an active element subtracts r from the sum of all the units of activation
reaching the same target element. A negative threshold (or threshold o)
implies activity in the absence of external activation. Such an element can

be silencedby a correspondingamount of inhibition.


Q, n,i -)



+" '.q,hh ,-,6afr'i"srartrs

these threshold devicesthe output does not appear immediately

upon activation of the input, but only after a short delay, say one
tenth of a second.During this time the gadgetperforms its liale
calculation,which consistsof comparingthe quantity of its activation with its threshold.
You can abeady guesssome of the things that a vehicle fitted
with this sort of brain can do, but you will still be surprisedwhen
you seeit in action. The vehicle may sit there for hours and then
suddenlystir when it sightsan olive greenvehiclethatbuzzesat a
certain frequencyand never movesfasterthan 5cm/sec.Sinceour
brand 5 vehicleis not interestedin any othervehicles,you might say
that the olive greenvehicle is its specialfriend. You will have to
conclude that Vehicle 5 has somethinglike proper nouns in his
mind, NAMEsthat refer to very particular obiects,like James,Calcutta, or Jupiter.


I Logic

Figure roa
A network that gives a signal when a burst of 3 pulsespresentsitself,
precededand followpd by a pause.


*"dtr l-n


Figure rob
A nenvorkof thresholddevicesthat emitsa pulsefor everythird pulsein a
row in the input.
But Vehicle 5 can do much more than that. It can count (figure
ro). It may associateonly with groups of four vehicles,not more
and not less,to make a party of five. Or it may visit every tenth
sourceit encounterson its way. Or it may turn awayfrom a vehicle
whosenumberof sensorsis a multiple of seven,implying that such
vehiclesbring bad luck. In some way, it 'tseemsto operate with
tztlugr q" hf(/p
Lan f re Iz /a\1*",
If you fit such a vehicle with a very large roilb., oI shrewdly
connectedthresholddevices,you may get it to play a passablegame
of chess.Or you may make it solve puzzlesin rocrc or prove
theoremsin euclideangeometry.You realizewhat I am driving at:
with enoughthresholddevicesit can do anything a computer can
do, and computerscan be made to do almost everything.
But where is the memory, some of you will ask, realizing that


24 | VEHICLE 5
most of the activitiesof a digital computerconsistof putting data
into memory, taking the data out again to perform somecalculation, putting the resultsback into the memory, and so forth. The
answer: there is room for memory in a network of thresholddevices,if it is largeenough.Imaginea thresholddeviceconnectedto
a sensor for red light.
it is activated by the red light, it
activatesanotherthreshold devicewhich in turn is connectedback
to the first device.Once a red light is sighted,the nvo deviceswill
activateone anotherforever.Take a wire from the output of one of
the two thresholddevicesand connectit to a bell: the ringing of the
bell then signalsthe fact that at sometime in the past this particular
vehicle sailedin the vicinity of a sourceof red light.
This is an elementarysort of unuonv. It is not difficult to understand how out of such elementarymemory stores (consistingof
synthesized,with the possibility of storing extremely complex
events.But there is a limit to the quaatity of facts the vehiclecan
store this way. For instance,when storing numbers,if the vehicle
has a bank of ten elementarymemory devices,it cannot fit any
number that has more than ten digits (in binary notation), since
qlch elementarydevicecan at most rememberone digit by being
active or inactive ("one bit of information").
There is a trick that can be used by our brand 5 vehiclesto
overcomethe intrinsic limitation of their storagecapacity.Imagine
a vehicleinvolvedin a calculationin which numbersoccur that are
much larger than the number of parts in the vehicle'sown interior.
You might think that such a task would be forever beyond the
comprehensionof that particular vehicle.Not so if we employ the
following strategy. Let's transfer our vehicle to a large, sandy
beach. The vehicle can crawl on the beach,leaving marks in the
sand indicating the successionof digits in the large numbersthat
emerge from its calculations.Then it can crawl back, following

zS I Logic

its own track, to rgad off the digits and put them back into the

calculation. A


)a g t,

The vehicleis never able to comptehendtheselarge numbersat

any one moment. But using itself as an instrument in a larger
schemeinvolving the environment,and partly directedby it, it ends
up with the correct result. (Of course,to be on the safe side, we
must supposethat the sandy surfacehas no limits.) If you want a
concreteexample,think of the-vehiclecalculating the difference
(small enoughfor it to comprehend)benveennvo large numbers,
which it can produce but not comprehend.It will produce one
number by leavingmarks on its way along the beach.It will produce the other number on its way back. And then it will measure
the differenceby counting the numberof marks that afe in default
or in excessof the first number.
Later on, we will learn how to incorporateinto a vehiclesomething quite analogousto the sandoutside,and almost as boundless
in its capacity.


Selection,the Impersonal

In this chapter things get slightly out of hand. You may

regret this, but you will soon notice that it is a good +deato give
chancea chancein the further creationof new bran{s of vehicles.
This will make availablea sourceof intelligence
is much more
powerful than any engineeringmind.
Out of the collection of vehiclesthat wgdave producedfor the
purposesof our experimentation,we wilfchoose someof the more
complicatedspecimensand put therrp"6ntoalarge table. Of course
there will also be somesourceso
sound,smell,and so forth
on the table, some of them
and someof them moving. And
there will be various shapes."drlandmarks,including the cliff that
signalsthe end of the
Now you and I will
a plentiful supply of materials (tin,
plastic, threshold deylces,wheels, motors, sensors,wires, screws
and bolts) and proceed to build vehicles,taking as our models
vehiclesthat we,,pickfrom the ones circulatingon the table. Each
time we copy a vehicle,we will put both the model and its copy
back on the iable, pick up another vehicle,copy it, and so on. Of

27 | Selection
course we will not pick up vehicles that have fallen on the floor
becausethey have proved their own inability to cope with the environment. We will be careful to produce vehicles at a pace that
roughly matches the rate at which vehicles fall off the table, to
prevent the race from dying out, on one hand, and to prevent the
table from becoming unduly crowded, on the other.

Note that while we are playing this game,we won't havetime to

testthe behavioror to study the wiring,let aloneto understandthe
logic of the vehiclesthat we pick up,as modelsfor copying. Nor
should we. All we are askedto do is to slavishlyconnectthe parts
according to the pattern in the mgdel.
Note also that when we do thic in a hurry, we are bound to make
occasionalmistakes.It may bl our fault when our copy of a perfectly well-testedvehicle
off the table as soon as we put it
down. But it is also
that we will unwittingly introduce a
particularly shrewd
into the pattern of connections,so
that our copy will
while the original may turn out to
It doesseemsurpfisingthat errors arisingin the sloppyexecution
of a task should/ct as germsfor improvement.What is lessastonishing is the cryitive power of a specialsort of error consistingof
new combinay'onsof partial mechanisms,each of which is not
disrupted iny'ts own well-testedstructure.This can easily happen
when we
up one vehicleas a model for one part of the brain
and thenfy mistakepick up anothervehicleas a model for another
part of phebrain. Sucherrors have a much greaterchanceof transcendiy'gthe intelligenceof the original plan.
i6 is an important point. If the lucky accidentslive on forever,
they/will alsohavea multitude of descendants,
for they will stay on
table all the time while the lesslucky onescomeand go. Therethey have a much greaterchanceof being picked up by the
ists as models for the next generation.Thus very good ideas

unwittingly introduced into the wiring, though improbable,do becomequite widespreadin the long run.
. .'
This story is quite old and goesby the,nameof Darwinian evolution. Many peopledon't like the idea tliat everythingbeautifuland
marvelousin organic nature shoul{,,6edue to the simplecooperation of reproduction, errors, and/election. This is no problem for
us. We have convinced oursefvi:sthat beautiful, marvelous,and
shrewdmachinescan be mailCout of inorganicmamerby this simple trick. Moreover, we alfiady know that analysisis much more
difficult than synthesis.,.&hetethere has been no consciousengineeringat all, as i
caseof our type 6 vehicles,analysiswill
necessarilyproducldthe feeling of a mysteriou.ssupernaturalhand
guiding the
. We can imagine that in most casesour analysisof brains,/rypr 6 vehicleswould fail altogether:the wiring that
behavior may be so complicatedand involved that
we will7,{everbe able to isolatea simplescheme.And yet it works.



have already usedthe word knowledge,evenif in a
somewhatfacetiousway, when we discussed
the propertiesof Vehicle 3. And we haveiust observedhow a processakin to Darwinian
evolution may incorporate knowledgeinto machinesin a mysterious way, though it is not immediatelyobviousthrough what channel the knowledge (about the dangersconnectedwith a cliff)
enteredthe "brain" or in what form it is containedthere. In both
caseswe are referring to fixed, inborn knowledge that, whether
right or wrong, belongsto the individual vehiclefor better or for
worse.This is fine for a set environmentbut may be catastrophic
when the conditionschange.Therefore,in a preciousvehiclethat
we love, we should build in mechanisrns
of adaptationto make it
more flexible.Not only will our vehiclethen be preparedro meet
catastrophiceventsbut it will also be readyto cope with a greater
variety of situations and thus be less confined ro a particular
We proceedas follows. First, we buy a roll of a specialwire,
calledMnemotrix, which has the following interestingproperry: its
resistanceis at first very high and stayshigh unlessthe two componentsthat it connectsare at the sametime traversedby an electric

current. When this happens,the resistance
of Mnemotrix decreases
and remains low for a while, little by little returning to its initial
value.* Now let's put a piece of Mnemotrix between any two
threshold devicesof a fairly complicatedvehicleof type 5. This is a
lot of wiring, but the effect is not great at firsr, due to the high
resistanceof Mnemotrix. Very little current will spread from an
active component to all the other componentsto which it is
As the vehicle(which is now type 7) movesaround and experiencesvarious situationsin its environment,someof its Mnemotrix
connectionswill changetheir strength.Supposeaggressive
in that particular environmentare often paintedred. Then the sensor for red in our type 7 vehiclewill often be activatedtogetherwith
the threshold devicethat respondsto aggressive
behavior,and the
Mnemotrix wire connectingthe two will have its resistancedecreasedso often that it will not have time to return to its initial
value. The consequenceis obvious: every time the vehicle senses
red, the whole setof movementswith which it normally respondsto
aggressivebehaviorwill be activated.Soour vehiclewill turn away
from its dangerousfellow. The enhancedconnection between the
components representswhat philosopherscall AssocrATron,the
associationof the color red with aggression.More generally,we
may say a new coNcEpr has arisen in the vehicle: wheneveran
aggressivevehicleis around, evenif it is blue or green,our type 7
vehicle will "see red." As far as we are concerned,this can mean
* I don't care if the electriciansshudder.They know very well
that evenif Mnemotrix
is not available commerciallyas a wire, it can be simulatedby a simple circuit. And
they also know that suchthings exist in animals' brains.If you want a fairly realistic
explanation of Mnemotrix wire, think of a material that changesits conductanceas
a function of temperature: the current heats the two components connectedby
Mnemotrix, and the temperaturechangeat the two endsof the connectioninduces
the change in resistance.


I Concepts

only: the vehicle does some of the things it did previously only \-\
when it was confrontedwith the color red.
This processof translating things that happen together in the rl.
environmentinto "complexes" of activity within the vehiclesis of \ 5 - <i
suchgreatimportancethat we ought to familiarizeourselveswith it
some more. One consequence,
we have already seen,is concept
formation. when it happensbetweendifferent categoriesof things
(suchas red color and aggression),
we prefer to call it association.
But it may happen within a single category, say smell, when a
number of chemicalsdissolvedin the air are frequently perceived
together,suchas burned plastic,lubricatingfluid, and battery acid,
which are set free when a vehicleis wrecked.So it is iustified for
survivingvehiclesto storethe "smell of death" in order to be able,
later on, to identify dangerousregionsof their environment.This is
done by the formation of a new olfactory concept.
Visual concepts may be formed in a similar manner. Tlie
straightnessof a line in differenr parts of the visual field, for example, may come to signify the dangerouscliff at rhe side of the.
table. And the movementof many obiectsin different directions
may cometo representthe concept"region crowded with vehicles."
But visualconceptscan be treatedmoreefficientlylater on when we
provide our vehicleswith the a priori categoryof space.For now,
we should explore some of the philosophicalimplications of the
processof conceptformation.
Let philosopherswatch a breedof rype 7 vehiclesand let
about the vehicles'behavior.One philosophersaysi4his
is all very well, but learningto recognizesituationsthat are of some
importanceis a fairly trivial performance,especiallyif it is donethe
hard way, by reward and punishment.It would be a different matter altogether if these vehicles could form their own conceptsin
quiet meditation, without an external tutor telling them what is
important. But they never will, becauseabstraction is one of the
powersthat is unique to the human mind. \\

But look, saysanother philosopher,I iust watchedan ABsrRAcTIoN beingmadeby one of thesecreatures.It wasmovingaround in
a crowd of peaceful,unpainted gray vehicleswhen it met a vehicle
painted red that proved to be aggressive;
then it met a greenvehicle
that also proved to be aggressive.When my vehiclemet another
painted fellow, this one painted blue, it immediatelythought that
too. And it turned awayin a hurry. This is a
this one was aggressive
true abstraction,the conc'ept'ofcolor replacingthe individual colors red and greenof the original experience.Or if you wish, we can
has taken placefrom particular colors
say that a cENERALIZATIoN
indicating dangerto the generaldangersignal"color."
Sureenough,saysthe third philosopher,but that is not difficult
to explain either. It has somethingto do with the way colors are
representedby the activity of the electronicparts in the gadget.
Undoubtedly in all the messof wires there will be one wire that
signifies"gray" as the evenmixture of all colors.Then theremight
well be one that signifies"not gt?yr" and that one was activewhen
the red vehicleappeared.So the "not gray" wire had the strongest
and this was learned.No wonder
correlation with aggressiveness,
this "not gray" wire functioned as a dangersignalwhen the blue
All right, saysthe fourth philosopher,but nobody in his right
mind ever suspectedanything more mysteriousbehind the "faculty
of generalization."
Fine, says I, as long as you admit it.


Space,Tbings, and Mouements

We take the next step in the improvement of our vehicles

primarily as a favor to ourselves, to keep things tidy and to make
the wiring less cumbersome. But we will find that the introduction
of internal maps of the environment is of inestimable value for the
vehicles too, making it much easier for them to discover the truth
about their environment.

What I mean by a map is this: take a set of photocells,say one

hundredof them, but insteadof distributing them messilyover the
surfaceof the whole vehicle,arrangethem in a neatsquareof ten by
ten photocellson the front surfaceof the casing(figurer r). Now fit
a lenson top of the array,making it into a camera.You know that
if everythingis setcorrectly,the invertedimageof things in front of
the vehiclewill be projectedonto the array. Of course,you cannot
pick up a perfectTV picture with just one hundredphotocells,but
you will get a picture. It will not be scrambledinformation about
the outsideworld; it will be a representationof the order of things,
of their neighborhoodrelationsand, roughly, of the distancesbetween them.
It is easy to make good use of this orderliness.
may build
networks of thresholddevicesthat can distinguishamong random

Figure rr
Vehicle8 with a lenseye.

35 | Space,Things,and Mouernents
environmentsand environmentsthat contain lumps of matter,
things that move and orderedstructures.
Build yourselfan array of thresholddevices,eachconnecredro a
group of neighboring photocells,say four of them arrangedin a
square(figurerz). Now aslong asrhevehicleis surroundedby little
insignificantobiectsor by obiectsquite far away,all of the photocellsmight seeiust a few of thesethings,all in more or lessthe same
numbers. Consequently,the photocells will all become active
roughlyto the samedegree.Evenif somephotocellaccidentallysees
a few more things than its neighborsand consequentlygivesa little
more output, the effect will probably be averagedout by the
thresholddevices,which alwaysadd the ourput of four neighboring
photocells.But when a largerobiectappearsin the neighborhoodof
our vehicle,it will be seenby one or more groupsof photocellsthat
are all connectedto the samethreshold device.This devicewill be
activatedmuch more strongly than the others and thus will function as an obiect detector,of inestimablevalue for the vehicle.
It might be even more useful to construct a set of mouernent
detectorsconnectedwith the array of photocells(figurer3). Put the
output of eachphotocellinto a delay,a devicethat givesoff a signal
a little while after it has receivedone. Nothing's easierthan that.
A sluggishthreshold device will do. Now make a new array of
thresholddevices.Eachis connectedwith one photocell via a delay
device,and with anotherneighboringphotocelllocatedto the right
directly, without a delay device.Thesethreshold devicesbecome
activeonly when they receivea signal from both channels.Every
time a bright object movesby from left to right, it will elicit a signal
in onephotocell,which will be storedfor a short while in the delay.
By the time the obiect elicits a signalin the neighboringphotocell,
the delaywill give off its signalaswell so the rwo signalswill hit the
movementdetector-thresholddeviceat the sametime, making it
active.obviously, a spot moving in the oppositedirectionwill not
havethe sameeffect becauseit will hit the fast thresholddevicefirst

Figure rz
An obiectdetector.Eachof the thresholddevices
on the right responds
only when four neighboring sensors arranged in a square are active

Figure 13
A set of movement detectors (C) for movement from right to left. The
threshold devices C become active when they receive input directly from
the sensor F to the left, and at the same time receive input indirectly, via a
delay element D, from the neighboring sensor to the righr

17 | Space,Things,and Mouernents
and the sluggishone afterward-so their output will not coincideat
the next level.Thus our movementdetectorsare directional.
We can of coursemake different setsof movementdetectorsfor
different directionsso that no movementwill escapethe attention
can also make them for various velocities,or
of our vehicle.
evenfor objectsof various sizes.In order to do this, we first make
an array of obiect detectors,as in figure rz, and connect their
outputs in pairs to the movement detectors.Only movementof
obfectsof a certain size, defined by the wiring of the individual
obfect detectors,will elicit activity in the movementdetector.We
may also proceedthe other way around.First we make an anay of
movgmentdetectors,all tlned to movementof the samevelocityin
the samedirection.Then we take the output of setsof neighboring
movementdetectorsand connect each set to a threshold device,
which then actsas an obiect detector.But this obiect detectorsees
an object only as a set of points, all moving in the samedirection.
This, by the way, is how we humansseecertainobjectstoo-such
as a cuttlefishmoving on the sandy ocean floor, no maffer how
good the mimicry of the beast.
Another well-known way to make gooduseof an arrayof photocellsis what is often called lateral inhibition (figure r4). Make an
arrayof thresholddevicesbehind the array of photocells.Connect
them one-to-oneto the photocells,so that eachwill be activatedby
light in the correspondingposition. Now introduce lateral inhibition: let eachactivethresholddeviceput a brake on the activity of
its neighbors,so that the more it is activated,the more its neighbors
are inhibited.You can easilyseethat therewill be an unevenmatch
benveenneighboringthresholddevicesreceivingdifferent amounts
of excitation:the one more strongly excitedwill put the other one
completelyout of business.Thus, insteadof getting a continuous
distribution of activity reflectingall the shadesof the environment
seenby the photocells,you will get a representationof isolated
bright spots.Only in the caseof an entirely uniform illumination


Figure 14
excitedby thatmanysensors,
to its
by inhibitoryconnections.
Uniformexcitationof thewholeset
will besubdued
by theinhibitoryinteractions,
whileisolatedspotsof excitition will standout.
will all the thresholddevicesstay at the samelevel (althoughthere
are difficultiesat the borders of the array).But in the caseof uniform illumination, the thresholddeviceswill also inhibit eachother
by the sameamount. Thus uniformiry will be weakly represented,
which is all right, for uniformity is uninteresting.
It is quite clearthat thesetricks, and a numberof other tricks that
you might invent, are only possiblewhen there is an orderly representation of the "sensory space" somewherein the body of the
vehicle.This neednot be z-dimensionalvisual space,as in the examplesjust discussed.
It may be 3-dimensionaltactile space;we can
representinternally,in a 3-dimensionalarray,all the points that the
vehicletouchesby meansof a jointed arm carrying a tactile sensor.
can also represent3-dimensionalvisual space,if we passrhe
signals from two eyesthrough a devicethat performs the sort of
computationknown as "stereoscopicvision" in human psychology.
can invent all sorts of bizarre internal spaceswhich we might
useto file in a convenientway the information reachingthe vehicle.
Two-dimensionalvisual spacecombinedwith one temporaldimension may lead to a representationof all the images,past and pres-

3g I Space,Things,and Mouements
ent, in a 3-dimensionalspatial arraywithin the vehicle.Inspiredby
someof the things that are known about animal brains,we could
also invent a 3-dimensionalarrayfor the filing of acousticinformation, with one dimensionrepresenting
the frequency,the secondthe
intensity,and the third the phaseof the acousticsignals.
Curiously,when we constructinternal spacesfor vehicles,we are
not even confined by the 3-dimensionalityof familiar spacethat
seemsto limit our immediateintuitive understanding.It is difficult
to imaginesolids of more than 3 dimensions,say a 4-dimensional
cubeor a 5-dimensionalsphere.In fact, when we think of an ordinary 3-dimensionalcube,we tend to imaginesomethinglike a box
with g squaresides.If we want to imaginea 4-dimensionalcube,we
noticethat the sideswould haveto intersect.But we cannotpicture
this, so we give up.
On the other hand, it is quite easy to imagine or to draw
networks of more than 3 dimensions(figure r5). The drawing
shows spheres connected by wires. The nenryork is truly 4-dimensional,sincein order to specifythe coordinatesof one of the
balls (or the path that leadsfrom one ball to a certainother ball),
you haveto indicatehow many stepsto move in directionsx, y, z,
and w. If you disregarddistanceand angleson the drawing (you
can't keep them equal on a projection even in the caseof a 3dimensionalnet), and if you imaginethe net continuedad infinitum
in all 4 directions,the network will look the sameno matterwhich
ball you sit on or in which of the 4 directionsyou look. Now, you
could evenbuild the nenvork, or a piece of it, out of spheresand
wires: you would be able to hold in your hands a structurethat is
intrinsically 4-dimensional, though of course collapsed ("proiected")into the 3 dimensionsof spacein which your handsmove.
(An architect similarly collapseshis buildings into the z-dimensional spaceof his drawing board.) You could even sit on your
nerwork and squashit into a z-dimensionalfelt. It would not matter. A louse finding its way along the wires would still notice the

4 0 | V E HI CL E 8

Figure 15
A four-dimensional
cube.Eachedgeis markedby threeblackdotson a
line, connected by a wire.

4-dimensionalconnectivity,provided it had the necessarymathematical acumen.

The point I want to make is the specialvirtue of nenyorks as
opposed to solids. once you have decidedto representspaceby
discontinuous,discretepoints within the vehicle,you can represent
"neighborhood" by meansof linesconnecing the points.This gives
you the freedomto mimic all sortsof spaces,includingspacesthat a
human mind cannot imagine.can the vehicleimaginesuchspaces?
we must turn to the philosophersagain.Let us ask a philosopher
whethervehicle 8 is endowedwith the a priori conceptof space,for
this is a familiar questionto him. only, in this casethe philosopher
cannot iust closehis eyesand look insidehimselffor an answer.He
will have to invent experimentalsituations in which the vehicle
could demonstrateits proper use of an internal representationof
space.A simple test: move the vehicle from its presentposition a

4r I Space,Things,and Mouements
ceftain distancein a certain direction, and then again in another
direction.If the placewherethe vehiclewas beforehad somefavorable connotations,it might want to go back. tUfill it move back
exactlythe way it came,or will it choosethe diagonal,which is the
quickestway to get there?If it has an internal representationof.z'
dimensionaleuclideangeometry (that is, if it has z-dimensional
spacebuilt in a priori), it will head directly toward the goal.
Now this internal representationof spaceis somethingthat we
could very easily wire into the network within the vehicle.Just
imagine a z-dimensional sheet made of a material which has
everywherethe sameconductancevaluefor electriccurrents.This is
definedas the current (in amperes)divided by the voltageapplied
(in volts) for a wire of a certainthicknessand a certainlength.Now
if we apply a voltage differencebetweent'wo points on the sheet,
the current that flows through the materialis strongest(the current
area,is highest)alonga straight
density,current per cross-sectional
we let one of the nvo points
line connecting
representthe place where the vehicle is and the other point the
placewhereit wants to go, we caneasilyconstructa devicethat will
determinethe best coursefor the vehicleby way of a simplemeasurementof current densityin different directionson the sheet.
So we would conclude that Vehicle 8 does have the a priori
conceptof z-dimensionalspace.Could Vehicle 8 embodythat of 3and 4-dimensionalspaceas well? To wire an internal representation of 3 dimensionsinto the vehicle,we could use a block of the
samematerial out of which we madethe z-dimensionalsheet,with
many electrodesembeddedin it to producevoltagedifferencesand
measurecurrents.But for 4 dimensionswe alreadyknow that we
haveto resort to 4-dimensionalnetworks, sincewe are not able to
make (or even imagine) 4-dimensionalblocks. In principle, this
could still measureshortest
doesnot make much of a difference.\U7e

42 | VEHICLE 8
vehicleshow off its built-in a priori conceptof higherdimensional
space.If the vehiclecould talk, we would ask it to roratein its mind
a 4-dimensionalcube,let us say90 degrees,aroundone of the axes.
There are such exercisesin human IQ tests,using z-dimensional
picturesof 3-dimensionaldice with three sidesshowing.The three
sidesare decoratedin differentways.The questionsareof this sort:
is cube A just another view of cube B, C, D, or E? Somehumans
havetrouble with 3-dimensionaldice, all with 4-dimensionalones.
But a vehicle endowed with a nerwork like the one in figure 15
might very well pass the IQ resr for 4-dimensionalcubesif the
questionwas posed in a languageit could understand.
I can hear myself talking to the philosophers.again.
The point I
am making is that orderly representationof spacein a vehicle is
more than iust convenienceof construction.It provides for easy
tests of reality.
have seenhow easyit is to knit nerworks that
will react to imagesmoving at certain speeds.If thesecan be taken
as imagesof obiectsin the world outside,the velocity ofthe movement of the imageswill stay benueencertain reasonablebounds, dictated by the physical laws gove-ing the movemenrof the obiects.
In particular, there won't be any movement of infinite velocity;
therewon't be any suddendisplacement.Continuity of movement,
no matter at what velocity, is a primary criterion for the physical
reality of an obiect. Also, the continuity and certainregularitiesof
the changeof shapeof a shadowindicatethat the shadowis castby
a solid obiect.This, too, could be fairly easilydetectedby a nenvork
with z-dimensional connectivity.And of courseidentity of shape
irrespectiveof movement (a strong clue for obiectskeepinga certain geometrical relation with a given vehicle)can also be detected
by such networks. we will take up this point againfrom a different
point of view in the next chapter.Here it was sufficientto show that
in our vehicle,just as in the physicsof relativity, the recognition,or
even the existence,of obiects is related to the dimensionalityof
space,internal and external.



We will improve on our vehiclessome more, along the

linesoutlinedin the constructionof the precedingbrand 8, but with
a different intention this time. We will try to furnish our vehicles
with a convenientset of ideas referring to the shapesof things,
especiallyto shapesas we seethem with our eyes(and as a vehicle
seesthem if it is equippedwith a good camera-typeeye).
First of all, if we want to considershapeindependentlyof color
and other irrelevantdetails,we must producean outline drawing of
thingsin the visualfield of the vehicle,asa draftsmanwould with a
pencil.(Webster'sdictionary definesshapeas"the quality of a thing
that dependson the relative position of all points composingits
outline or external surface.") This is not very difficult if things
stand out clearly againsttheir backgrounds-for instanceif these
thingsare birds in the sky or vehicleson a white sheet.We can then
use the trick of lateral inhibition, which we have already learned
(figure r4). Only sharp boundarieswill be passedon to the next
level,therebyproducing a pure line drawing. If the interior of the
is quite homogeneous,
sayall black, therewill be
only the outline or shape.
Let us construct detectorsfor elementaryproperties of shape.

44 | V E HI CL E g

Figure 16
A detectorfor bilateralsymmetry.Thereis an array of elements
which an image is projected. Elements symmetrically spaced with respect to
the midline enhance each other. There will be a strong activation of the
array for bilaterally symmetrical images.

The first property that comes ro mind is bilateral symmetry. Irs

detector is easyto construct and enormouslyvaluable (figure r5).
Again we make an arrayof thresholddevicesonro which a picture
of the external world is proiected by meansof a suitablecamera
system(we can filter the picture first through a nerwork with "lateral inhibition" to enhancerelevantdetail).One half of it receivesa
picture of the right half of the visual environment,everythingto the
right of the vehicle;the other half receivesa picture of the left half
of the world. Now we connect by a wire each pair of threshold
devicesoccupyingsymmetricalpositionson the right and left sides.
Through the wire the threshold devicesinfluenceone another in
such a way that when they both receiveinput, they becomemuch

o, Jtr

45 |

t1 t}tn^ -'ll

5a u''arP

,rt pl{: 6,

moreactivethanwheno'ty or,fif themis
shape(witha verticaloi,
thatwhenthevehiclefacesa symmetrical


' -- v \ J

of symmetry,such as an upright human figure seenfrom the front

or from behind), there will be much more activity in this array of
thresholddevicesthan there will be in any other case.For every
elementexcitedon one sideof the vehicle,its symrhetricalelement
on the other side will also be excited,with the consequentreciprocal enhancement.
Let's not talk about an upright human figure; that introducesan
unintendedaestheticaspect.Think only of a world populated by
vehiclesof the variouskinds that we havebeenbuilding.Up to now
we have not talked much about the exterior appearanceof our
vehicles,although we haveimplicitly assumedthat the vehiclesare
madeof rwo halves,mirror imagesof one another:two motors,one
on eachside,rwo nostrils,a symmetricalcasinglike an automobile.
Of coursesuch vehicles,seenfrom the side, are not symmetrical:
their senseorgans are in front, their motors are in the back, and
their prevalentmovementis always in the same"forward" direction. Nor are the vehiclessymmetricalin the up-down direction if
they move around on surfaces,as our vehiclesmostly do; for reasons connectedwith gravitation, there will be wheels (or other
instrumentsof locomotion) on the side of the vehiclesfacing the
ground, the so-calledunderside.
But there are good reasonsfor the vehicleto be symmetricalin
the direction perpendicularto both the "front-back" and the "opdown" directions-along the axis definedby the pair of concepts
"right" and "left." We have seen this early on in the casesof
Vehicles2,3, and4, which showedsurprisinglylifelike behavioron
the basisof paired, very simple,symmetricalconnectionsbetween
two senseorgansand nvo motors.The kind of behaviorassociated
with two symmetricalreinsgoverningthe motors is one in which an
obiect is isolated from the environmentas a partner in behavior.
The vehicle'smovementsare directed by feedback,either turning

46 | VEHICLE g
the vehicle toward the object or turning the vehicle awayfrom the
Consider the first case: feedback that makes the vehicle turn
toward the obiect. An observermight say that our vehiclehas that
obiect on its mind or our vehiclepaysattentionto that object.Well,
what if the object is anothervehicle?What would the situationlook
like to that vehicle, and how should it react? Obviously the situation in which a vehicle seesanother headingdirectly toward it,
whether in an inquisitive, a friendly, or an aggressive
mood, is a
specialcaseand well worth specialattention.The detectorfor bilaterally symmetrical shapes,which we have iust described,proves
helpful here: we may connectit to the output in sucha way as to
trigger the mechanismsthat govern the appropriatereactionsto
"another vehicle facing me" or "another vehicle having me in
mind." (Perhapsone shouldreactivatethe beautifulterm "confrontation": fronts coming together, facing each other.) In fact, it is
clearthat bilaterally symmetricalconfigurationsin a natural world
containing only vehicles(and no other man-madeobiects,such as
churchesor monuments)would mostly signify just that: a parrner
in interaction with the observer.
is a relation bet'weenbilateral symmetryin sensory(especially vi
and the concept of "thour"
n of the
has been
e buildersof temples and churcheswho, by
evoke the presence of an a

always facing the

ln conversatlon

. The sameprinciple
rved in
suchas orchids,adopt bilater
rical sh
in order to be acceptedas "partners" by insects
rs keyed to this type of symmetry.
I want you to note that somethingnew and very important has
crept into our discussionof a detectorwith bilateralsymmetry.\U7e
decidedto give our type 9 vehiclesa systemof connectionsbetween
correspondingpoints on their right and left sides.In order to ex-

47 | Shapes
plain how usefulsucha systemwould be,we had to invokenot only
the externalappearanceof other vehicles(which our vehiclemight
meet)but their behavior as well. Things are getting complicated:
we are no longer working on individualstaken by themselvesbut
on the membersof a community in which there are complicated
interactionsbetweenvehiclesof the sameor of different kinds.
Everyimprovementthat we invent for the latestbreedof vehicles
put in circulation will either force othersout of businessby a processof Darwinian selection(seeVehicle 5) or make otherschange
their behavior through learning (seeVehicle z). Of course,this
makes it difficult to foreseewhat will actually work out as an
"improvement." Sometimesthe net effectwill be contrary to what
we expect, due to unforeseenreactions of the environment. But
certain great inventions will survive all vicissitudesand will be
I suspectthat the detectorof bilatimmuneto all shrewd defenses.
eral symmtry,which providesinformation about "being in someone'sfocusof attentionr" belongsto this category.Evenin biology
with all its complicated interactionsbenveenspecies,the symmetry
detectorhas remainedof primary importance.An insectin search
of a sexual mate does not really care if it gets occasionallysidetracked by an orchid as long as its symmetry detector servesthe
right purposein the mafority of cases.
Other insects fall for different kinds of flowers, for those with
radial symmetry, like daisies.We can also construct radial symmetry detectorsfor our type g vehicles:thesedetectorsmight indicate singularities in the world, sources from which something
emanatesin all directions.A radial symmetrydetectorcould also be
basedon the fact that no movementis perceivedon approachinga
patternlike that of figure 17. The picture remainsidenticalto itself.
A fundamental category of form is periodicity. A repetitive pattern may signify many important situations.It may signifya collection of identical individuals.Then again, a periodic pattern left on
the ground may be the track of a vehiclemoving by somesort of

48 | VEHICLE s

Figure 17
A patternthat is invariantto changes
of scale.A vehicleapproaching
centerof the figurehasa constantvisualinput (providedwe makethe
of perceived
may be usedasdiagnostic
for figureswith radialsymmetry.

periodic steppingmechanism.or the pattern may be generatedby

some oscillatory movement in the form of a standingwave-an
indication of stored energy.For all thesereasonsperiodicpatterns
are happeningsof great importance in this world; they are iust
as fundamental as bilaterally symmetricalor radially symmetrical figures. So we should equip our vehicleswith detectorsfor
This can be done in various simple ways. For instance,we can
give them periodic templateswith different spacingand let them
match the picture of the environment with the templatesby the
mathematicalprocessof cross-correlation.This is the principle of
Fourier analysis.Its technicalrealizationdoesnot requiretoo much
ingenuity.Another interestingdetectorof spatiallyperiodicinput is
implicitly containedin the nenvork describedin the previouschap-

49 | Shapes
ter as lateral inhibition. We have seen that such an arcay of
threshold devicesneglectscontinuous excitation and enhancescontrasts. It gives maximal output for patchesof excitation spaced
sufficiently far apart so that they won't disturb eachother by inhibition. For a periodic pattern, the spacingis determinedby the length
and strength of the inhibitory connections.If we test the lateral
inhibition devicewith striped patterns, we will notice that it gives
the sameoutput no mafferhow the stripesare orientedif the inhibition works in all directions.
Taken together,vehiclesof types 8 and 9 have provided much
new evidencefor our law of uphill analysisand downhill synthesis.
A problem that taxes the minds of psychologistswhen they deal
with real animalsor humans,that of inborn concepts,found many
solutions when we attacked it from the downhill, synthetic direction. We built very simple homogeneousnetworks and then discoveredthat they contain implicit definitions of such conceptsas
3-dimensionalspace, continuous movement, reality of objects,
multitude of obiects, and personal relation. More and more we
are losing our fear of philosophicalconcepts.
The exercisesin syntheticpsychologycontainedin this chapter
dealmostlywith visualinput. It is of courseeasyto imaginea priori
conceptsin other categoriesof input, suchasthe tactileor olfactory
inputs.It is quite elementaryto provide the vehiclewith detectorsof
aural periodicity. Th.y would detect various frequenciesin the
purely time-dependent(nonspatial)input derivedfrom one of the
vehicle'sears (microphones).The a prioris of frequency,the socalled resonators,have been basic to human auditory theory for a
long time.


Getting ldeas

The time has come to sit back and considerthe strange

variety of vehiclesthat populate our laboratory. Th.y all go about
their businessaccording to certain rules, someof which we understand, becausewe invented them ourselves,and some we don't,
becausethey emergedfrom a sort of Darwinian evolutionaryprocess.The objectsof their interestare definedby simpleproperties
such as smelland color, or by more abstractproperties,suchasthe
periodicity of their coloring or the symmetryof their outline. Formal propertiesmay stand for evenmore abstractdefinitions,as we
have seenin the caseof bilateral symmetrysignifyingthe situation
of "somebodyhaving me in mind."
Someof our vehiclesseemto move around smoothly, as if attracted and repelled by the sourcesof various fields of force
superimposedon one another.Othersappearto make suddendecisions, rousing themselvesfrom a rather phlegrnatic condition to
take off on isolatedventures,after which they resumetheir stateof
rest. The vehiclesseemto know their environmentrather well, so
much so that they are able to reachsomeobfectswith closedeyes,
so to speak,apparentlyon the basisof someinternalizedmap on
which the obiect'slocation is recorded.On the whole, thesevehicles


I GettingIdeas

are surprisinglysmart, especiallyconsideringthe limited amountof

intelligencethat we, their creators,haveinvestedin them.
But do they think? I must frankly admit that if anybody suggestedthat they think, I would object.My main argumentwould be
the following: No matter how long I watchedthem, I neversawone
of them produce a solution to a problem that struck me as new,
which I would gladly incorporatein my own mental instrumentation. And when they cameup with solutionsI alreadyknew, theirs
neverremindedme of thinking that I myselfhad done in the past.I
requiresomeoriginality in thinking. If it is lacking, I call the performanceat best reasonablebehavior.Even if I do observea vehicle
displayinga solution to a problemthat would not haveoccurredto
me, I do not concludethat the vehicleis thinking; I would rather
supposethat a smart co-creatorof vehicleshad built the trick into
the vehicle.I would haveto seethe vehicle'ssmartnessarisingout
of nothingr or rather, out of not'so-smartpremises,beforeI concludedthat the vehiclehad done somethinking.
But this doesnot meanthat we cannotcreatevehiclesthat would
satisfythis condition. We shall do this gradually, starting with the
problem of hauing ideas.Let us take one of the vehiclesof type 7,
the oneswith the Mnemotrix connectionsthat introducethe effects
of experienceinto the brain. This vehiclehas beenaround for some
time and has absorbeda great dealof knowledgeabout the world.
This knowledgetakes the form of statisticalcorrelationsbetween
elementaryeventsin the vehicle'ssensoryspacesor statisticalcorrelations benveen more complex events representedin some
thresholddevicesof its interior (or betweenelementaryeventsand
Supposethe vehiclehas learnedthat certain objects,A, B, C, D,
are situatednearthe rim of the tabletop on which it lives:a brokendown vehicle,a light, a battery, a hill, a supply of screws.It has
learnedto associatetheseobjectswith the concepts"margin of the
universe"and "dangerouscliff." On its occasionalexcursionsto-

S z I V E H I C L Er o
ward the margin of the universe,it will alsohavenoricedthe neighborhood relationsamongpairs of theseobjects:the screwsare next
to the hill, the light next to rhe battery, and so forth. One day, after
enough excursions,the vehicle will suddenlyrealizethar all these
paired associations(A next to B, B next to C
Z next to A)
make senseif the whole situation is seenas a closedchain. The
vehiclenow has the idea of a finite boundeduniverse,with objects
A to Z marking the marginal closed line. Once this "image" or
"idea" is generatedout of individual itemsof knowledge,it is there
to stay. It may, in fact, be immediately recordedon the maps,
whose use we have discussed.If so, we will observethat forever
after the vehiclemovesaround much more expertly.
We must be careful,however,not to let the processof acquiring
new ideasinterferewith the detailedknowledgethat our vehiclehas
assiduouslycollectedand carefully storedin many associativeconnections during its lifetime. We know that this may happen in
humans who are overly dedicatedto the developmentof ideas.
Th.y tend to connectmany individual casesinto generalcategories
and then usethe categoriesas if they were things,losingthe potential for categorizingin other ways by rememberingeachinstance.
In the exampleof the discoveryof the margin of the universe,I
can seethis danger.The idea of a closedchainof obiectsmay be so
strong that it keepsthe imagesof theseobiectspermanentlyactive
in the vehicle's brain. The consequenceis that associationswill
developbetweeneveryobject on the margin-and everyother object on the margin. The serialorder that led ro the original ideawill
therebybe lost or at leastsubmergedin a systemof much stronger,
massiveassociations.The way out in our casewould be to let the
excitation circulate in the closed chain associations.This would
strengthenthe associationsrepresentingthe serialorder of the objects and would not allow cross-association
to develop.
Here are somemore examplesof ideasthat may arisein vehicles.
There are coins lying around on the floor in the universeof the

St I GettingIdeas
vehicles.Someof the coinsare decoratedwith a pictureof a human
headand others are decoratedwith a number. One of our vehicles
hasalreadylearnedro recognizeand to distinguishthe two typesof
coins.That is, there are distinctpatternsof activity, saynno different thresholddevicesbecomingactivewhen one or the other kind
of coin is seen by the eye of the vehicle.Now one of the coins
showinga human head is flipped around by the vehicle-and suddenly it showsthe number.This happensagain and againuntil, by
the learningprocessthat we havealreadyincorporatedin our vehicles,an associationis formed: "head, flipping, number." Of course,
the associationalsoworks the other way around. once the association is acquired,the vehicleknows that, after adding the action of
flipping to the sightingof "number," the picture of the headwill be
seen.It may also be reinforcedby the contraryexperience,
when the
flipping of coins showing the number revealsthe head.
\U7emay call the whole complex of head-flipping-numberand
number-flipping-headthe idea of a coin with nrro faces.It arisesin
the vehiclealthough the two facesof a single coin are never seen
together.The idea of a coin with two facescan ariseevenif there
are some coins around with human headson both sides,as long
as thesecoins escapethe vehicleduring the phaseof "getting the
Here's another example.Moving through a garden,a vehicle
finds out that flower number one of a row is a sourceof food,
flowers z to 7 are not (they are poisonous),flower number 8 is
againa sourceof food and so areflowers r j, zz, andso forth. After
a while it may happenthat in the brain of the vehicleonly one of 7
thresholddevices(connectedin a circular fashion)alwaysbecomes
activein temporal coincidencewith the finding of a sourceof food
in a flower. This is again "getting the idea": that particular
thresholddevicewill be associatedwith the food finding sysremwith the consequentadvantageof being able to predict sourcesof
food without havingto investmuchenergyin the processof sniffing

s4 | VEHICLEro
must suppose,of course,that the time it takesfor one
thresholddeviceto becomeactive after anotherhas beenactivated
is exactly the sameas the time it takesto get from one flower to the
next or, better still, that the advancementof activity by one step in
the ring of threshold devicesis triggeredby eachflower.
All of this is not complicatedin principle but boring to carry out
in detail. \U7erely on the processof Darwinian selectionthat, starting with the vehiclesof Wpe 6, has introduceda greatvariety of
different patternsof connectionsinto the vehicleswithout our even
noticing it (although*. do recognizethe vehicles'greatlyincreased
complexity of behavior).We canwell imaginethat the vehiclecould
get the idea "edible flower" evenif the only flowersthat wereedible
were those whose ordinal numberswere squareor whose ordinal
numberswere prime. Thereis, however,a complicationin the cases
of squaresand of prime numbers.If thesenumbersget too large,the
vehicle has to perform a long and intricate dance between one
flower and the next in order to find out whetherthe flower's number is square or prime, leaving marks on the earth and retracing
them accordingto complicatedrules.We haveseenthis before,at
the end of the chapteron Vehicle 5, which also had its limitations.
No such difficulry arisesif the vehiclehas to find out whether a
number is evenor odd, or whether a numberis a multiple of six or
of eleven,as long as the vehiclecan count to eleven.
In this chapter we were only interestedin the generalidea of
"getting ideas." Readerswho want to know exactlywhat kind of
to calculatenumbersthat
nerwork of thresholddevicesis necessary
are square,or prime or whatever,must read the textbooksof automata theory.


Rulesand Regularities

Most of you will not yet be convincedthat the processof

getting ideasas it was describedin the previouschapter has anything to do with thinking. It is not surprising,you will say, that
occasionallysomethingclicks in the workings of a fairly complicatedbrain and from then on that brain is ableto perform a trick
(an algorithm, as somepeople say) that can be used to generate
of numbersor of other images.It is also not
surprisingthat thesemay occasionallymatchsequences
of eventsor
things in the world of the vehicle.
I will show you that this is just one step in the direction of
creating behavior akin to thinking. In the following chapterswe
will introduce more elementsof the thought process,making new
vehiclesto show new tricks, new typesof performance.In the end
our vehicleswill surpriseus by doing somereal thinking.
We want to equip Vehicle rr with a brain about which it can be
said-in a more radicalway than it could be said about previous
editions-that it is a model of the world; We already introduced
partial aspectsof this model idea,when we talked about the usefulnessof internal mapsrepresentingexternalspaces(Vehicle8), and
when we describeda learning process(Vehicle7) that discovers

things in the environment and establishestheir internal models
(called concepts).But this is not enough. These things move
around, bangagainsteachother, associateand dissociate,grow and
break.We havealtogethermissedthesedynamicaspectsup to now.
\Wewill introducethesedynamicsby improving on the systemof
Mnemotrix connectionsalready introducedin the type 7 vehicle.
You will rememberthat theseconnectionsbetweenelementsin the
vehicle'sbrain were of different strengthsand could be mademofe
effectivewhen the elementsthey connectedwere often activated
together.This turned out to be very convenient,becauseso manyof
the facts about the world that are interestingand important to us
(and to the vehicles)may be expressedasthings or eventsthat tend
to occur together.For this reasonit is unlikely that we will give up
the trick of associativelearningin any further developmentof more
But we soon discoverthat there are important piecesof knowled{e about the world expressedin a different form: eventsthat do
not presentthemselvesat the sametime but in succession-pairsof
events,of which one is always the first and the other the second,
like lightning and thunder, swinginga hammerand hitting the nail,
or, in the world of vehicles,meetinga sourceof food and tastingthe
food. When we discovera pair of suchevents,we tend to think that
one is the causeof the other, whateverthat means.But this may
lead to wrong interpretations,for instancewhen both eventsare
producedby a third hidden event,only with different delays.Most
of the time, however,when two eventsregularly occur in succession, it is no accident.And it certainly is useful for a vehicleto
know what to expectwhen eventsoccurthat haveimportant,possibly dangerous,consequences.
\Uflecould useour old supply of Mnemotrix wire togetherwith a
little electronicsto incorporate into the vehicles'brains all those
delayedcoincidencesof eventswe have beendescribing.I7hat we
want to achieveis a connectionbetweenthe rwo internal represen-

57 | Rulesand Regularities
tatives of an event A and an event B such that, when the representative A is activated by the input, the representative B is activated by
the connection, but not vice versa. The connection would then
represent the fact that "B often follows A" or, if you wish, the
causal tie benveen A and B. This would force us to do a rather
complicated wiring for every such connection. In order not to burden our constructive imagination too much, w prefer to buy a
different sort of wire, called Ergotrix, which conducts in one direction only and has an increased conductance when it is interposed
berween elements that are active in successionwithin a brief time.
We must be careful, of course, to install the wire in the right direction, conducting from the element that tends to be active first to the
one that tends to be active second.
Once again we will see to it that all of this happens auromarically. Plenty of Ergotrix wire will be installed between as many
elements as possible so that whatever sequencesoccur can be recorded in the system. Of course there will be no lack of opportunity
for learning. tU7ithall the movement in the world around the vehicle, with all the natural laws operating, and with all the other
vehicles displaying fairly regular behavior on the basis of all the
tricks that we (or the processesof evolution) have built into them,
many sequencesof events will repeat themselves and they will be
worth learning.
You may ask why we did not use Ergotrix wire in the first place
(Vehicle 7) when we first gave our vehicles the capacity to learn,
starting with those complexes of properties that frequently occur
together becausethey belong to one "thing." We used the Mnemotrix wire, which is ideal for associations, because it couples elements in a symmetrical fashion; once coupled, each of the properties
can recall the other in quite the same way. For each Mnemotrix
connection we could have used two Ergotrix wires (one for each
direction) to obtain almost the same result. But there are two reasons to leave things as they are.
First of all, we don't want to go back in evolution and change

things that have alreadyproved to be convenient,sincewe might
losesomeadvantagethat we havenot evenrealized.(Rememberthe
of uphill analysisand downhill synthesis:we run the risk of not
any longer what we previouslyput together.)Sectunderstanding
a good idea to keepthe nvo processes
conceptu4 t
ally separated-the associationsof elementaryproperties into
things or conceptson one hand and the sequencingof conceptson
the other hand, one controlled by the Mnemotrix, the other by the
Ergotrix system.The two kinds of learning produce two different
X kinds of knowledge,likegeographyand historyror systematiczool.-{ "W and animal behavior, referring to what kinds of things exist
to how they develop and interact.
If we let our imaginationsgo and try to work out in detail what
kinds of things the Mnemotrix systemwill discoverin a real world,
and what kinds of dynamic laws will be incorporatedin the Ergotrix system,we soon discoverthat the two kinds of knowledgeare
perhapsrelatedmore than we had assumedinitially for reasonsof
First of all, it would seemthat the process
of abstracting things from the environment-concept formation at
the most elementarylevel-must occur prior to the processof discovering the dynamic properties of these things. For the laws of
of eventsrefer to the developmentand to the combination of things rather than their elementaryproperties.This is familiar from our own human experience:listeningto a new language
we want to learn, we must first discoverindividual words, or roots
of words (somethinglike the morphemesin linguisticterminology),
before we can evenhope to discoverthe rulesthat governtheir use.
Also, in the developmentof a scienceit is often apparent how the
discovery and denomination of phenomenaprecedesthe definition
of the laws of their transformation.Chemistryhad to go through a
descriptive phasebefore the physics underlying the variety of subst'ancescould be understood. Zoology had to be taxonomic before
it was organizedby the theory of evolution.
V '

t .

59 | Rulesand Regularities
On the other hand, purely descriptiveclassificationis not only
boring, it is also potentially misleading.It may lead to the wrong
categorieswhen it is not guidedby at leastthe intuition of a theory
of the underlyingprocesses.
A centuryof microscopicanatomyhas
filled the librarieswith thousandsof beautifullyillustratedvolumes
that are now very rarely consultedbecausethe descriptivecategories of the old histology have beenlargely supersededby the new

acousticallywell-definedepisodes(the syllables,which the naive

listenercanrecognize),it is certainlytrue that a better,moregeneral
definitionof morphemesor words is derivedfrom grammar.\U7ords
(I usethis term loosely)are the segments
of speechthat we discover
as the ultimate particles of grammar. If we had no idea or no
experienceof grammar, we might neverdiscoverthat theseare the
rUflemight propiecesthat are shuffled around to form sentences.
posea different, incorrect segmentationof speech,for example,a
segmentationinto syllablesin a languagewith polysyllabicwords.
Words becomemeaningfulinsofar as they are usedin a grammatical system.
In other words, abstractingmeaningfulchunksfrom the environment (things,events)and discoveringthe rulesof their behaviorare
two processesthat condition eachother and are necessarilyinterlaced,like the learningof the vocabularyand the learningof grammar in a languagecourse.
Coming back to Vehicle rr, it seemslike a good idea to let the
discoveriesof the Ergotrix systeminfluencethe learningprocessin
the Mnemotrix system,on whoseinitial abstractionsit in turn depends.I don't want to work this out in detail, but somethinglike
the following schemewould clearly be possible.We have already

6 o I V E H I C L Er r
describedthe conditionsfor the strengtheningof an Ergotrix wire.
These conditions are fulfilled when an element,say a threshold
device,at one end of the wire becomesactiveshortlybeforeanother
elementbecomesactiveat the other end. We havealso seenthat it
is mostly groupsof suchelements,stronglyinterconnectedand representing "things," that become active in succession.Now let's
introduce the rule that whenever the Ergotrix wires become
strengthened,the Mnemotrix wires within eachof thesegroupswill
also becomestrengthened.
Thus conceptsare establishedin the vehicleespeciallywhen they
appearin regularsequences.
How would this look to us?We would
notice, observingthe apparentlyerratic behaviorof a vehiclein its
world, that the vehicledisplaysparticularly well definedreactions
to eventsthat are known to haveconsequences.
Take, for example,
a vehicle approachingan obstacleat high speed. would not be
surprisedto seethe vehiclepromptly reactto its perceptionof the
danger of a collision. Similarly, Vehicle r r will quickly remember
which of its own behavior pafterns regularly and quickly elicit a
reaction from other vehicles.We observethat afteran initial learning period Vehicle r r will either produce thesebehaviorpatterns
frequently or pointedly avoid them. It will use them as signals.It
will also learn thosesignalsthat regularlyprecedecertainbehavior
patterns of other vehicles.After a while Vehicle r r will react to
thesepremonitory signalsiust as it reacted,beforethe learniqg,to
the behavior that regularly followed the signals.
But it would take prolongedobservationto noticethis particular
aspectof learningin the vehicles.As a matterof fact, we might not
have suspectedit if we had not introduced a piece of our own
philosophy into the construction of thesevehicles.As our brain
children becomemore efficient,we notice that the "law of uphill
analysisand downhill synthesis"becomesmore and more compelling. For the time being, take the messagein this form: sinceyou
were not satisfiedwith the first meagershowingof intelligencein

6r I Rulegand Regularities
our vehicles,we started adding a few more tricks, hoping that they
would convinceyou a little more. The first uick we tried was the
coding of the environment in thoseterms that yield a maximum of
correlations and logical structure, in other words, in the most
meaningful terms.


Trains of Thought

At this stage,if you want to be critical, it is easyfor you

to maintain that up to now you havenot discoveredanythingin our
vehiclesthat goesbeyond ordinary learning.True, thesecreatures
seemedto becomemore and more able to dealwith the adversities
of their environment,not only by a processof Darwinian selection
but also by activeassimilationof information from the world. But
thinking is different.It is a processthat cango on for a long time, as
everyonewho has done someconsciousthinking knows. Thinking
can be observedin other people as well, when we get verbal or
nonverbalevidencefor a succession
of mentalstatesthat are guided
by somecriterion of plausibility or logic-mental statesthat reflect
the exploration of various blind alleys and eventual arrival at a
result. Sometimeswe seemto notice suchmentaloperationevenin
a monkey or in a dog. But not yet in a vehicle.
The possibility of sustaining long successions
of distinct brain
statesfor the purposeof exploring knowledgealreadyincorporated
in the brain is what we will introduce in a new brand of vehicle.
which we will call Vehicle rz.
First a remark on pathology. All the later vehicles,beginning
with Wpe 7, are in constant danger of running into a condition

I Trainsof Thought

quite analogousto epilepsy(which is alsoone of the most common

forms of derangementof animal brains).The strengtheningof the
connectionsbenveenthe elementsof the brain, which is at the basis
of associativelearning, embodiesthe danger of reciprocalactivation beyond control. In a population of elementsin which excitatory connectionsabound,if the numberof activeelementsreaches
a certain critical level, chancesare the remaining ones will also
becomeactivated.Theseelements,in turn, keepthe first setactive.
A maximal condition of activity is then establishedand maintained
until the supply of energyis exhausted.This maximal activation
makesno sensein terms of the information ordinarily handledby
the brain, which is keyed to patternsof partial activation of the
elements.Necessarilythe result is disorderly,ineffectivebehavior.
There are various ways of dealingwith this danger,and I propose
the following for our vehicles.
Let everythreshold devicein the vehicle'sbrain be touchedby a
specialwire through which we can control its threshold.If we set
the thresholdshigh, the thresholddeviceswill becomeactiveonly
whenthey are very stronglyactivatedby the input they receivefrom
other thresholddevicesor from the sensors.For a lower threshold,
lessinput will suffice.So if we watch the operation of the brainand in particular the total amount of activity in it-we can always
preventan attack of epilepsyby raisingall the thresholds.If thereis
not much activity, we can lower all the thresholdsand thereby
encouragethe circulation of activity through the brain. It is of
coursequite easy to let this happen automatically. All we need
(figure 18) is a box that receivesas its input the number of active
brain elementsat that moment and calculatesappropriatethresholds, which it then setsfor the whole brain. In real life, the input
for this thresholdcontrol devicemight be the rate of changeof the
number of active elements,in order to give it an opportunity to
foreseethe catastrophicexplosionof activity beforeit happens.But

64 | V E HI CL E tz

r __>

Figure 18
B is tlhebrain,whichreceives
input I andelaborates
an outputo. At the
sametimeit signalsthelevelof activityA in itsinteriorto a special
@for the elements
in B.
for purposesof illustration it will sufficeif the thresholdcontrol
deviceworks just on the amount of aciviry in the brain.
The effect of this global negarivefeedbackon rhe activity of a
vehicle'sbrain is illustratedin figure 19, which showsthe number
of active elementsas a function of the numberof activeelementsa
moment earlier.
the activiry is low, it will againbe low at the
next moment. (For very low excitation, there may even be a tendencyfor the activiry to die our, sincea minimum densiryof active
elementsin the brain is requiredto activatethe next setof elements,
but this is not shown in figure r9.) For very high levelsof excitation-that is, for a very large number of activeelements-we may
imagine that the thresholdsare immediatelyset so high that the
activity will drop to a very low level at the next moment. Intermediatelevelsof activiry will lead to maximum activity atthe next
moment (seethe middle part of the curvein Egurer9). Later on we
will come back to this curve, which has interesdngphilosophical
implications. First let us watch the operationof a brain that con-


nr,l^Aw f.zt! ai.V''


65 | Trains of Thought

Figure 19
The function describingthe next number of activeelementsA;..,,1
presentnumberof activeelementsA;. It can be seenby iteration (follow the
linesstartingfrom the arrow neara) that the statesof a brain controlledby
sucha rule are quite unpredictable.

tains many learned associative connections while it is being controlled by the feedback of a threshold control device.
We have already noticed that the vehicle's brain has a tendency
to explode into fits of activity because of the abundance of reciprocal activation between its elements, a situation reminiscent of the
chain reaction in a block of uranium. But most of these explosions,
if everything works out the way we have planned, should take place
within limited groups of elements that are tied together by particu-

+c, t,


R./ Lo,UJ.,


4 ftrTe- a,,h,,tl
,,1-rrrl r.'"r,,{


rections.Suchsetsof elementsariseas
"concepts" representingthings or eventsthat haveoften presented
themselvesin the environmenr.
Let one suchthing appearin the sensoryspaceof Vehiclerz. The
explosion of activity will happen in the correspondingset of
threshold devicesresponsiblefor that concept.This impliesan increaseof the number of active elementsin the brain, and the
thresholdcontrol devicewill immediatelyreact to it by raising all
the thresholds.A moment later many elementsthat werepreviously
activewill be silent. But the elementspertaining to the conceptin
questionare likely to stay active.This is becausethe strong reciprocal connectionswithin the set, once activated,guaranteea very
high level of excitation for eachelementof the set.This level is so
high that the activity of the elementsmay survivethe raisingof the
thresholds.Thus the first interestingeffectof our recentinnovation
is the focusingof individual concepts-of parternsthat havetheir
own internal consistency-at the expenseof backgroundactivity.
We greatlyappreciatethis effectin a well-functioninghumanbrain,
where it is often called the rocusrNc oF ATTENTToN.
But there is more. You rememberthat we haveinstallednot only
Mnemotrix wire for concept formation but also Ergotrix wire,
which representswithin the brain the relation of temporalsuccession, of consequence
or causality.Thus the elementsnow activein
the lone surviving concept after the automatic raising of the
thresholdsalso have someErgotrix wires attachedto them. These
Ergotrix wires lead to the elementsthat have often beenactivated
after the conceptin question,the consequences
of the activeconcePt'so to speak.Obviously, there will be more than one possible
next step for all but the most determinedsituations.
So we must ask ourselveshow the vehicle'sbrain finds the concePtthat follows the one it presentlyholds.The choice,it turns our,
is quite automatic.Among all the elementsactivatedby the presenr
concept through the Ergotrix wires, there will be some groups

67 | Trainsof Thought
strongly connectedby Mnemotrix wires becausethey again form
concepts.Thesegroupswill of courseignitewith particular alacrity
becausethe internal connectionswithin eachgroup will provide an
explosivekick to the activationfrom externalsources,that is, from
the active concept. Now you can see what will happen. The
threshold control, alarmed by all this growth of activity, will
quickly raise the thresholds,smotheringmost of the activity and
leavingonly the most resistantgroup of elementsactivated.As we
have already seen,this will be the group with the strongestreciprocal connections.In termsof conceptswe may put it this way: the
next concept, among all the concePtsthat are possible consequencesof the presentone, will be the most consistentor familiar
one-the one most strongly establishedby experience.
Note that with all these budding and growing explosionsthe
thresholdshave beenraisedabovethe level at which they were set
for the previousconcept.It is thereforevery likely that the previous
conceptwill be extinguished.So the systemwill not swing back into
its former condition but will end up with a different concept.This
embodiedin Ergotrix
new conceptwill haveits own consequences
wires. And thesewill againmaterializein a new concePtby way of
the sequenceof eventsthat we havejust described.The processwill
continue as long as you wish or as long as the chain of concepts
doesnot lead back to the conceptfrom which it started.
The upshot is something very much akin to thinking, to that
processso familiar to our introspection,where imagesappear in
successionaccording to rules reflecting the relations between the
things they stand for. This processgoeson in our minds when we
try to figure out the bestway to get from one point to another in
a familiar city by leaing our imagination produce successionsof
streetcorners(or other landmarks)whoserelationsof geographical
proximity we haveexperienced.It is alsoone of the tricks we useto
of possiblemovesin a game of chess,
determinethe consequences
of somestatementin a discussion.This chainor the consequences


68 | vnulcLEtz
ing of internal statesis exactly what we plannedto introduceinto
the brain of Vehicle rz to make its meditationslook more lifelike,
more like our own, not only in the time they take but also in the
unforeseenroutes they can follow.
There is an importanr properry that the brain of vehicle rz
shareswith the brains of our fellow men.consider againthe curve
of figure 19, which showsthe numberof activeelementsas a function of the number of active elementsa moment earlier.The exact
shapeof the curve is not very important, as long as it has a maxh,lh I imum and cuts the diagonal (A1: Ai*1). Startwith a certainvaluea
<4, > 1 on the abscissaand find the ordinate of the next value & on the
curve.Put that value b againon the abscissaand find c, and so on.
You will be surprisedto find that the succession
of valuesa, b, c
in generalquite
unpredictable.Now you will rememberthat figure 18 describesrhe
effectof thresholdcontrol on the activity of the brain of Vehiclerz.
r!(/emay takea, b, c .. . asthe numberof activeelements
in the brain
in successive
moments of time. If there are very few elements,the
will by necessitybecomerepetitiousafter a short while.
But for a fairly largebrain the succession
will be truly unpredictable
to an observer,for any practicablestretchof time.
I hopeyou realizewhat this means.If you could observethe inner
workings of the vehicle'sbrain, say, by watchinglight'bulbsconnectedto the threshold devices,and theselight bulbs lit up every
time the correspondingelementbecameactive,you could not even
predict how many lights would light up in the next moment, let
alone what kind of pattern they would form. (For any givennumber there are of course many constellationswith that number of
activeelements!)At this point we'should againinvite our philosophers to comment.
I would claim that this is proof of rnnr \nLL in Vehiclerz. For I
know of only one way of denyingthe power of decisionro a creature-and that is to predict at any moment what it will do in the

69 | Trainsof Thought
future.A fully determinedbrain should be predictablewhen we are
informed about its mechanism.In the caseof Vehiclerz, we know
the mechanism,but all we can prove is that we will not be able to
foreseeits behavior.Thus it is not determined,at leastto a human
I know what the philosopherswill reply. Th.y will say that althoughthis may look like freewill, in fact it is not. \U7hatthey have
in mind when they use that term is the real power of decision,a
forceoutsideany mechanicalexplanation,an agentthat is actually
destroyedby the very attempt to put it into a physicalframe.
To which I answer:whoevermade animals and men may have
beensatisfied,like myself,a creatorof vehicles,with somethingthat
for all intentsand purposeslooks like freewill to anyonewho deals
with his creatures.This at least rules out the possibiliryof petty
exploitationof individualsby meansof observationand prediction
of their behavior. Furthermore,the individuals will themselvesbe
unableto predict quite what will happenin their brainsin the next
moment.No doubt this will add to their pride, and they will derive
from this the feeling that their actions are without causaldetermination.



And indeed-following up the last sentenceof the previous chapter-it may be said that the internal rumblingsof Vehicle
rz are at leastaimless,if not random, constrainedas they are only
by the rules of plausibility stored in the vehicle's memories
(Mnemotrix and Ergotrix) but not determinedby them.
I am sure that most of you will not believethat "aimlesssuccession of images" is an accuratedescriptionof what goeson in your
minds most of the time. You will not be impressedby our vehicles
aslong as thereis no evidenceof somepurposeguidingtheir behavior and some direction in their thinking. Theseare virtueswe are
pleasedto seein our children.Why shouldn't we try to modify our
brain children,the vehicles,in this direction?It won't be difficult in
principle, and it meansa lot to thosephilosopherswho like to think
that goal-directedbehavior is the one property that gives living
beingstheir very specialstatuswithin the physicaluniverse.
There are two aspectsof goal-directedbehavior we must consider.First, the goal lies in the future. For instance,the eatingof the
mouseis the goal determiningthe movementsof the cat now. We
have the specialcaseof an event defined for a later time having
earlier effects, quite contrary to the effects that we are used to

7r I Foresight
consideringin physics.Second,the goal is desirableby its very
definition.We cannottalk about goalswithout first getting straight
the conceptsof good and bad.
Let us take the first problem first, that of actingtoward the future
or in accordancewith an event in future time. This is obviously
nonsenseif we take it to mean an action that is now a consequence
of somethingthat will happenonly in the future. However, it is an
entirely different matter-and it doesmake sense-if we take it to
mean an action that is a consequence
of somethingwe expect to
happenin the future, sincethat expectationmay well be available
before the action is planned. There is no violation of the law of
causality in this. All we need is a mechanismto predict future
eventsfast enough.o that they will be known beforethey actually
There are of coursesafepredictions-and othersthat are not so
safe.We have no problem predicting the future of a rolling stone
onceit is on its way down the slopeof a hill. But it is not so certain
whether a dog will leaveits comfortablepillow when it is shown a
pieceof cake.Other motionsare practicallyunpredictable,like that
of a child playing in the middle of the street.Yet the principle of the
prediction is very similar in thesecases.We haveseenenoughrolling stonesand hungry dogs that the perceptionof one situation
immediatelybrings to mind its consequences.
eventsare all we need for prediction, togetherwith a mechanism
forcing them to speedup in the reproductionwhen necessary,for
example,in dangeroussituations.Complicationsmay arise when
severaldifferent predictionsare approximatelyequally likely. In a
good prediction there must be the possibilityof predicting various
outcomes,given a certain situation, and of keeping the various
outcomesin mind in parallel. This is what we do when we drive
through a streetwhere children are at play.
Now we want to incorporatepredictioninto the vehiclesof type
r3. Clearly, the prerequisitesare all therein previoustypesof vehi-

cles.We were careful to reproduceinsidethe vehicles'brainsmany
rules and regularitiesthat govern the world. This way we could
speakof the vehicles'brains as modelsof the world, as miniature
editionsof external,public space.Their brainswerepopulatedwith
patterns of activity that mimicked the activities of real obiects in
their environment.We noticed that thesebrains (as modelsof the
environment)really cameto life only when the dynamicaspectsof
the world were also represented,so that a given functionalstateof
the elementsof the brain would evolveinto the next stateaccording
to the samerules that make the world evolvefrom one momentto
the next. lUtledid this by using Ergotrix wire, which activatesthe
elementsof the brain in the sameorder asthe sequence
of eventsto
which they correspond.And we implicitly assumedthat the Ergotrix wires would be trained to reproducesequences
of activationat
the samepaceas the original occurrenceof the sequences
of events.
But this is a somewhatgratuitous assumption:the Ergotrix wires
could work faster,or slower,than the sequences
that are impressed
upon them. Let them reproducethe sequences
at a more rapid pace
and you will have a brain that works as a predictor (figurezo).
We want to take a closer look at what goes on in a vehicle
equippedwith such a predictor. Rememberthat the thresholddevicesin the brain are ugder the influenceof two kinds of input: first,
directly or indirectly (via interposedfilters) from the sensesand,
second,from one another. Only the latter kind of influenceis
mediatedby Mnemotrix or Ergotrix wires.Considera certainstate:
the vehiclein quiet contemplationof the world, the thresholdcontrol at rest, and the thresholdsset high enoughso rhat only a few
ideas stand out over the background. (Theseideas are of course
represented by groups of active threshold devices with their
Mnemotrix cross-connections,
The evolution of the vehicle'smental stare may be affectedin
three ways. First, meditation. Even if the brain is at equilibrium,
with the thresholdsfixed, it cannot be entirely at equilibrium be-

;2 to-1

t;-1 t;2 t;3


t;3 $'2 tl

to td'l S2 q

Figure zo
A predictor with someauxiliary equipment. The flow of life is represented
by the film (or tape) being unwound from the reel marked future and
endingup on the reel markedpast. Only one momentof time, tro,is available
as input to the machine.The input is stored, however, for three units of
time on the endlesstape of a short-term memory. From there both the
presentinput and the contentof the short-termmemoryare relayedto the
predictor which computesthe future three units of time ahead.The predictor containsstatisticalinformation about the past embodiedin the Ergotrix
wires in its interior. The prediction for to * 3 is stored on another shortterm memoryuntil it is readyto be compared,by a specialcomparator,to
the real input t and to the input threeunits of time back.(Thisdepth in time
of the comparatoris desirablein order to assess
the dynamicpropertiesof
the predictor.) The comparator in turn emits signalsthat may modify the
predictor or switch it off (broken arrow).

T 4 | V E H I C L Er 3
causethe Mnemotrix connectionsbetweenthe activeelementswill
slowly grow in power, the longerthe ideais on. But this may not be
apparentfor a while, unlessnew elementsare recruitedto makethe
idea more ponderous,thereby upseaingthe equilibrium and making the brain go on a thinking tour, as we have alreadyseen(Vehicle rz).
Second,things may happen in the environment.The vehicle's
mental state will changeaccordingto new input from its sensors.
The transition from one stateto the next will be aided by the Ergotrix wires in the caseof a sequenceof eventsthat has occurred
before,but the Ergotrix connectionsare too weak by themselves
effect the transition without the help of the sensors'input.
Third, the sensorsmay signal a condition of the environmentthat
has alwaysevolvedin a certainway in the past. The Ergotrix connectionsin this casewill be very strong.And the next stateof the
vehicle'sbrain will be entirely determinedby them. As a resultthe
vehiclewill be blind to the real input that follows. Most of the time
this will not hurt the vehiclebecausethe sequenceof eventswill be
the sameas it has been in the past.
But occasionallythe rare event happensand the input clashes
with the internal prediction. This will result in a garbledcondition
that cannot developfurther in any coherentway. We want to avoid
this, especiallyin view of the fact that discrepancies
and expectationsare interesting and should be analyzedin detail.
Eventuallywe would like to provide the vehiclewith a devicethat is
turned on by iust these discrepanciesand amendsthe systemof
rulesusedfor the prediction,so that the vehiclewill know betterthe
next time it meetsthe samesituation.
First, we provide the vehicleswith two separaterepresenationsof
the environment,one in the predictor,the other in an equallylarge
ensembleof elementsthat receiveonly the fresh input from the
sensorsand do not elaborateon it. Thesenvo half brains are connected point to point to each other, so that the discrepanciesbe-

7S I Foresight
tween their states of activity can be detectedas easily as the
two drawingsif you hold them one on top of
the other againstthe light. The technicalrealizationis easy.Saythe
two half brains are connectedby inhibitory connectionsbetween
correspondingpoints. Therewon'r be much activity if the two parternsof activationare exactlyequal,becauseof the reciprocalinhibition. But if one of the nvo representations
not presentin the other, this will stand out strongly.
\U7ewant our vehiclesto be imaginative,but mainly realistic.
That's why in the caseof conflictinginformation we want to take
the information from the realistic half brain more seriouslythan
that from the predictor.We may incorporatea rule: when in doubt,
believethe sensors.And we do this by introducing a mechanism
that simply turns off the predictor in casesof conflict.But we want
to go one stepfarther; we want to educatethe predictorto make it
more realistic.This is not as easyas it may sound.Rememberthat
the eventin the environmentthat causedthe predictor to make the
wrong prediction belongsto the past by the time the clashbetween
the two half brains revealsthe mistake.
Thus we want something like shoft-term memory (figures zo,
zt), a third representationof the environmentlaggingbehind the
other two, so that, if necessary,
the past is availableat any time a
few stepsback. such a mentalechois not difficult to incorporatein
the vehicle'sbrain. Just connectevery elementof the sensoryhalf
brain with another elementthat becomesactive one unit of time
after the first, and with yet another set of elementsthat becomes
activetwo units of time later, and you havean efficientshort-term
Now with a few additional piecesof equipment,we can greatly
improve the predictor by making it more flexibleand open to new
We do not worry about occasionalwrong predictions,
especiallyif the mistakesare not fatal ones.Knowledgeis incorporated all the time in the Mnemotrix and Ergotrix connections,and

76 | veHtCLE t7

Figure zr
Learning by internal repetition of one-time evenrs.I is the input. D is the
Darwinian brain which emits judgmentson the desirability of the input and
sets two switches accordingly (through the broken arrows). On the left:
normal operation, with the Darwinian brain quiescent.The realistic brain
R feedsthe predictor P and also an open chain of nvo delayelementsd. On
the right: the Darwinian brain D signalsemotionalinput. The two switches
are thrown to the right, the predictor is no longer fed by the realisticbrain
but receivesthe contents of the nvo delay elementsd to which in turn it
givesouput. The information precedingthe emotion reverberatesthrough
the predictor and the delay elementsuntil the Darwinian brain calmsdown
again and setsthe switches to the normal position.

the statistical knowledge about the world they represent is never

complete (by its very statistical nature!). But this piecemeallearning
may not be sufficient when the one occasional deviation from the
statistics is a very important one (in the good or in the bad sense).
Say, for example, that most of the time green vehicles display peaceful behavior, but there is an occasional green vehicle whose aggressiveness is particularly vicious. It would be wrong to associate the

Z7 | Foresight
property"ggo/opeaceful"with the color greenand to reactindifferently to the sight of green,sincesooneror later an encounterwith
the greenmaverickis bound to take placeand the victim mustbe on
the defensive.It is betterthen to give specialweight to the rare but
decisiveexperienceand to consider green vehiclesas generally
How is this done?We are talking about "good" and "bad" as if
theseconceptswereeasyto define.Of coursethey arenot, but there
is a way out of this difficulty. Rememberthe vehiclesof our earlier
models.Th.y were fairly simplemindedcomparedto the ones we
are now developing,but they were efficient. The type 6 vehicles,
which underwenta processof Darwinian selection,know one thing
for certain: the avoidanceof dangerand the searchfor advantage.
And they know this eventhoughno one (neitherthe vehicles'buildhasany idea of a definitionof good
ersnor the vehiclesthemselves)
or bad. The type 6 vehiclessimply move forward toward good
things and back away from dangerousthings. But this is all we
Catch one of those Darwinian vehiclesof type 6, take away its
motors, and you have a detectorfor good and bad. The wire that
went to the forward motor signals"good" and the wire that went
to the backward motor signals"bad." So we can incorporatethe
brain of Vehicle 6 into the brain of Vehicle r 3 and therebyprovide
it with important, ancient,intuitive knowledge.
We can now put the piecestogether. Short-termmemory, two
stepsback in time for everythingthat happened,is alreadythere.
The predictor is there. A switch that momentarily turns off the
*I hope you rememberthat we are only talking about liale machines.It would be
wrong to cite the usefulnessof one-instancelearning in vehiclesas a iustification for
prejudice and superstition in human behavior. We do have vastly more complex
brains that enable us to make the diagnosis of good and bad independent of
superficialmarkers such as the shapeor the color of the casing.

78 | VEHICLEt3
predictor in the caseof a conflict berweenpredictionand reality is
also there.The Darwinian evaluatoris ready to signalparticularly
sinisteror ioyous events.The new trick: figure zr.
whenever the Darwinian evaluatorD signalsan unpleasantturn
in the real courseof events,or a very pleasantone, the predicting
half brain P is disconnectedfrom the input it normally receives
from the realistic(sensory)half brain, R. Insteadthe predictinghalf
brain receivesits input from the short-termmemorytwo stepsback.
Soit will go againthrough the rwo instanrsprecedingthe important
happening.At the sametime its output is connectedto the input of
the short-term memory. So it will receiveover and over again via
the short-term memory the successionof the two events,a and b,
until the Darwinian evaluatorD hascalmeddown and everythingis
switchedback to normal.
The net effect is that successionsof events leading to strongly
are incorporatedfirmly in the MnemotrixErgotrix systemeven if they occur only rarely. The internal reverberationsetup by the Darwinian evaluatorarrificiallymakesup for
their low frequency and turns them into high-frequencyeventsin
the inner workings of the brain.
we may relax now and observevehicle 13 in action. Its power
of prediction is quite apparenr when it follows a moving object
around, say another vehicle carrying a sourceof attraction. When
the obiect temporarily disappearsbehind an obstacle,vehicle 13
will head toward the place where it is likely to show up again. we
also notice more peculiarproperties.For no obviousreason,Vehicle 13 seemsto avoid certainplacesand vehiclesin its environment,
and it seemsto have an irrational affection for someother places
and vehicles.If we watch it long enough,we may find out that there
are indeed reasonsfor these idiosyncrasies.The vehicle may associatea one-timeeventwith this or that placeor vehicle,and act
accordingly.Vehicle r3 remembersfacts much aswe do, individual
facts and eventsof its past experience.This rememberingis differ-

79 | Foresight
ent from the memorywe haveconsideredbefore,which consistedin
the molding of behavior accordingto the unchangingrules and
regularitiesof the environment, perceivedthrough the statisticsof
many individual events.The vehiclesof type 13 derivetheir experiencesfrom rare but important happenings.They will be quite different, one from the other, becauseeachvehicle builds up its own
characterbasedon the particular experiencesof its early life.


Egotism and Optimism

As time goes on, we grow affectionatetoward the diversifiedcrowd of our vehicles,from the very simpleones to the
more complex modelsdisplayinginterestingsocialinteractionsand
sometimesquite inscrutablebehavior.we can play with them, we
may get to know them personally(and they may ger ro know us),
we can teasethem, testthem,teachthem tricks,and let themlove or
fight each other. We do not feel, however, that they show any
personaliry,not eventhe most complex onesof rype r3. It is difficult to say what we mean by that.
Perhapswe would acceptthem more readily as partnersif they
gave more convincing evidenceof their own desiresand projects.
we notice that our fellow men usually seemto be after something,
when they go about their businessor when we conversewith them.
Dealing with people is interestingbecauseof the challengetheir
continuousinternal schemingseemsto provide.The systemof desireswe suspectbehindtheir schemingmay be part of what we call
the personality. It may be the lack of just such projects that we
notice in our vehicles.we cannot help feelingthat they are driven
by necessityrather than drawn by goals-in spiteof all the efforts

8r I Egortstn
and Optimism
we put into them, in spite of specialmechanismsthat are apt to
abolishlowly forms of causaliry,and in spite of the predictor that
seemsto draw motives from a future stateof the world.
Once we have noticed this, we can of course,in a last creative
effort, endow a new kind of vehicle,our last, Vehicle 14, with a
certainamount of systematicegotism,with a touch of the pleasure
principle, in order to make it look more like our fellow humans.
We proceedas follows.
Rememberthat our more sophisticatedvehiclesalready have
built into them many componentsthat comein handy for this new
project. With the introduction of the Ergotrix wires (Vehiclero)
predictionbecameone of the vehicles'mentalhabits.In Vehicler3
the updatingof the predictor was greatlyimprovedby a mechanism
giving greatweight to rarc but imporrantevenrs.This was achieved
by incorporatinginto the brain of Vehicle13 another,more primitive, Darwinian brain that contributedall the ancientinformation
about good and bad things its ancestorshad accumulatedthrough
the generations.
Still earlier,we had noticed (Vehiclerz) that the succession
mental statesdictated by the Ergotrix connectionswas essentially
random and quite unpredictable(perhapseven unpredictableas a
matter of principle becauseof the peculiarmathematicalproperty
we associatewith the function of figure r9). The randomnessof the
decisionsmadeby Vehicle rz in part reflectedthe statisticalnature
of the knowledgeincorporatedin the Ergotrix connectionsand the
continuous updating of this knowledgeby an ongoing learning
process.It also dependedon the very nature of the processthat
makesthe brain swing from one stateof hctivity to rhe next during
alternateepisodesof raising and loweringthe thresholds,auromatically imposedby the mechanismof thresholdcontrol. IUe will now
givethis processan optimistic slantso that the pump of thoughtsin
the brain of the vehiclewill producea succession
of more and more
pleasurablemental images.We will convinceourselvesin the end

that such optimism not only leads to nice dreams but also has
obiectively favorable consequences.
Ve will assumethat most of the time the uncertaintyas to the
next state, given a certain state of activity, is not only an uncertainty for the observerbut an inherentuncertaintyin the sensethat
the predictor points toward (at least) two statesthat are equally
likely as a continuationof the presentstateof the brain (and therefore of the world). Sucha dilemma in previousvehiclesmight have
been decided by random elbment built into the brain (for ex"
ample, by Geiger counter making its decisionson the basis of
whether or not it was hit by a cosmicray within the last tenth of a
second).But from now on we will imposethe following rule for
Vehicle 14: when choosingamong severalequallylikely next brain
states,choosethe most pleasingone.
You have alreadyguessedhow we want to achievethis. We hold
the presentstatefor a short time (no problem,short-termmemory
is already there) while the predictor is allowed to go quickly
through its various predictions.At the sametime the built-in Darwinian evaluator is asked to evaluatethesepredictionsfor their
favorable or unfavorableaspects.It will in generalcome up with
this is done,the
different valuesfor the different predictions.'U7hen
predictor quickly goesonceagain through its predictionsand stops
This is
at the prediction with the highest scorefor pleasurableness.
then the next stateof the brain.
We don't needmore than that. We may put the vehiclesback on
the table and meditateabout their behavior.A superficialobserver,
or an impatient one, will not notice anything special. We, as
creatorsof vehiclesand experiencedobserversof their behavior,do
notice subtle changesin our latest perfectedbrain children. We
know their tastes:we have ample opportunity to seewhich sources
of stimuli, which situationsand which other vehiclesthey are attracted by and which they avoid. Their reactionsto thesethings in
the past were quite direct and easilyobservablewhen the obiectwas

81 | Egotism and Optimism

in the vicinity of the vehicle.Distant sourcesand situationsdid not

seemto affect them much.
Now it is different with type 14 vehicles.Th.y move through
their world with consistentdetermination, always clearly after
somethingthat very often we cannot guessat the outset-something that may not evenbe there when the vehiclereachesthe place
it wants to get to. But it seemsto be a good strategy,this running
after a dream.Most of the time the chain of optimistic predictions
that seemsto guidethe vehicle'sbehaviorprovesto be correct,and
Vehicle 14 achievesgoals that Vehicle rj and its predecessors
"couldn't evendreamof." The point is that while the vehiclegoes
through its optimistic predictions,the succession
of internal states
implies movementsand actions of the vehicleitself. While dreaming and sleepwalking,the vehicletransformsthe world (and its own
position in the world) in sucha way that ultimatelythe stateof the
world is a more favorableone.
We observeat somestagehow one of the vehiclesof type 14 is
waiting for anothervehicleto appear.This orher vehiclecarriesa
very appealingsourcewhich Vehicle 14 intendsto tap. It seemsto
be waiting impatiently, sinceevery now and then it performs the
motions that are associatedwith the tapping, as if by anticipating
its own behavior in the presenceof the desiredevent, it could
acceleratethe event'soccurrence."This is very humanr" we say.
"Haven't we all felt an urge to run ro the door long before the
doorbell rings, when waiting impatiently for a beloved friend?"
Indeed,it is aberrantbehaviordictated by a very subiectivelaw of
causality,but it doesseemto reflecta basicattitude of humankind,
this irrational belief in the effectiveness
of one'sown actions.

This portfolio of vehicles,someplacidly at rest, most madly careening

over the landscapeof the aftist Maciek Albrecht's imagination,
illustrates only a few of the many marvelous"creatures" inspired by
Valentino Braitenberg'stext.




t '





r=r ///
) l5,iir







aa :r.\




| .:'







Biologicd Notes
on the Vehicles

The precedingfantasy has roots in science.I will now

sketcha few facts about animal brains that have inspiredsome of
the propertiesof our vehicles,and their behaviorwill then seemless
gratuitousthan it may have seemedup to this point. I have been
directly or indirectly involved in most of the researchI shall mention. Thesenotesshould not be taken as a treatiseon brain science
and quite personalessays.
but as a seriesof disconnected

The uirtues of crossedconnections

(Vehicleszt j, and 4)
Vehicles r to 4, the early ancestorsof the whole breed,
springfrom an attempt to understandthat very curiousbasicfact of
brain science,the crossedrepresentationof the world in the (vertebrate)brain. The generalprinciple is apparentin the projection of
visual spaceonto the brain. A million or so fibersof the two optic
nervescarrying signalsfrom both eyestoward the brain crosseach
other in such a way as to representin the left brain an image of
everythingto the right of the animal and vice versa in the right
brain. Just how many fibersof the right eye actually seepoints of


the right half of the visualfieldis a questionthat obviouslydepends
on the position of the eyesin the head.In a frog or a mousethe right
eye looks to the right and the left eyeto the left, but in a cat or a
monkey-animals with forward looking eyeslike ourselves-each
eye seesalmost equal portions of the right and left halvesof the
world. The fibers in the optic nervesin eachof thesecasesexactly
follow the rule that everythingfrom the right world goesinto the
left brain and vice versa, which makes for a rather more complicated schemein the caseof eyespointing forward. Incidentally,
the samerule is valid for the senseof touch, whereagaininformation from the left half of the skin is relayedto the right half of the
brain and vice versa.The motor systemis also crossed:the nerve
cells whose activity is most evidently associatedwith a certain
motor act are on the side of the brain oppositeto that of the limb
being moved. Thus there is some fustificationfor sayingthat the
two halvesof the world are representgd
in the oppositehalvesof the
brain. But why should this be so?
Sinceit was first discovered,the fact of crossedprojection has
presenteda puzzle,and variousexplanationshavebeenattempted.
These have ranged from simple mechanical interpretations to
elaborateconstructionsinvolving argumentsabout imageprocessing within the central nervoussystem.At the simplestlevel it has
beenarguedthat the abundanceof crossedfiber bundlesmakesthe
brain mechanicallymore stable,by a lacing or weavingeffect.Another very generalargumentplacesthe origin of fiber crossingsin
the transition from a primitive (hypothetical)brain with spherical
symmetry to the bilaterally symmetrical brain of most animals
(figure zz).lt is arguedthat as a medianplane becomesdefinedin
this transition, we may renamethe connections,initially supposed
to be random, as crossedand uncrossed:the longestand therefore
most important fiber bundleswill be the onesthat crossthe median
plane.Apart from weak points in this argument,it shouldbe valid
for invertebrateas well asvertebratebrains.But while crossedpro-


| Crossed Connections

Figure zz
A simple explanation of crossed connections in the brain.'When a median
plane becomes defined in an animal with spherical symmetry, if each of the
elements is connected to each of the others, there are more fibers crossing
the median plane than fibers staying on one side. This little difference has
been invoked as the ancestor of the much more imposing crossed connections in vertebrates.

jection may occur in invertebrate

brains, it does not seem to be the

general rule.

Possiblythe bestknown explanationof the crossedrepresentation of the world in the brain is that providedby Ramony Cajal to
accountspecificallyfor the crossingof the optic neryes,which he
interpretedas a correctionof the image inversionthat occurs for
reasonsof geometrical
opticsin cameraeyes(r). His argumentis as
Supposethe right and left halvesof the visual field are
projected,with optical inversion,onto the right and left retinae.If
thesetwo half imageswere projected by uncrossedfiber bundles
onto the right and left halvesof a commonreceivingsurface,there
would be a midline discontinuityin the mappingof the visual field
on this surface(figure4A). Ramon y Caial sawthe chiasmalcrossing as a simple meansof avoiding this discontinuity(figure 4B).
The other sensoryand motor systems,accordingto this theory,
adaptedsecondarilyto the crossedrepresentationof the world in


Figure z3
of thecrossed
Ramony Caial'sexplanation
the continuityof the
inversionin the lenseye:the crossingreestablishes
on the brainin the awkarrowB, whichwould otherwisebe represented
ward fashionof A.
the brain. If visual imagesof objectsto the right are beingprocessed
in the left brain, it is economicalto let the motor commandsfor
actions dealingwith theseobfects(and presumablyexecutedwith
the right extremities)also arisein the left hemisphere.
Severalobjectionsto Ramon y Cajal's argumentmay be raised,
both on the basis of the reasoninginvolved and in the light of
experimentalresultssincehis time.
for correctionof optir. Crossingis sufficient,but not necessary,
cal inversion.For example,a rSoonadstof uncrossedfiber bundles,
or the equivalent internal crossingof fibers within each bundle,
would permit correction without inversion of the image (figure
z+:A,B). Similarly, recurvedand uncrossedbundlesproiectingonto
the posteriorpolesof the optic lobeswould also correctfor discontinuity (C).
an advantagein
z. The cogencyof Cajal's argumentpresupposes
a continuous unbroken representationof the visual field in the

99 | CrossedConnections

Figure z4
(A),or rwisting(B)of individual
A weakness
of Caial'sargument:
(C)woulddo thesametrick.
or a

brain. If the receivingsurfaceis identifiedwith the visual part of the

midbrain, the so-calledoptic tectum-as Ramony Caial apParently
did-then experimentalstudiesof the topograPhyof the proiection
on the tectum are pertinent.R. M. Gaze,for example,found in the
frog an orderly proiection of the left visual field onto the right
tectum and vice versa (z). And the orientation of the two projections is indeedsuchthat a continuouspattern in the visual field is
representedagain as a continuous Pattern on the tectal surface,
evenif part of the pattern is seenby oneeyeand anotherpart by the
other eye.But it is difficult to seewhat usethe optic tectum makes
of this continuity, sincethere is no continuityof the gray substance
acrossthe midline. The fwo halvesof the tectal nerve net are quite
separate.It doesnot matter much how the nvo piecesare oriented
to one other when the connectionis madevia fiber bundlesof the
white substance.Hencethe optic tectum of the frog doesnot Provide a good basisfor an explanationof fiber crossingin terms of


The compound eye of the fly:

reconstructionof continuity in the
I have myself given evidencefor the correctnessof the
Cajal principlein anothersystemof multiplefiber crossingsfound
in the visual systemof the fly. There the complicatedweaveof the
fibers leadingfrom the compound eyeto the brain exactlycompensatesfor the disruption of the imageproducedby eachof the lenses
proiecting small inverted portions of the visual field onto the array
of the light sensitiveelements(3).
The compoundeyeof the fly is composedof about 3,ooonearly
identical subunits,called ommatidia, eachequippedwith its own
separateoptics and containing 8 separatephotosensitiveelements,
the rhabdomeres.Each rhabdomereis a specializedportion of one
cell, the so-calledretinulacell.The upperendsof 7 of theserhabdomeresin eachommatidium are arrangedin a very regular pattern,
localized in the focal plane of the inverting optical system.This
pattern is called retinula (small retina) for a very good reason:to
each rhabdomerecorrespondsa line of sight, and to the whole
retinula 7 lines of sight, which intersecta distal plane in a pattern
which is that of the retinula rotatedby r8oo.
The optical information discretelygatheredby the elementsof
the retinula, and transformedby the visualpigmentsinto the kinds
of signalsthat are conveyedby nervefibers,is carrieddown into the
first visual ganglion-the lamina ganglionarisor simply laminathrough a bundle of 8 fibers emanatingfrom the base of each
ommatidium. It will come as no surprisethat this nervebundleis in
fact twisted by r8oo. The portion of the visualenvironmentseenby
each ommatidium has been inverted by the lens optics and could
not fit continuouslyinto the global pictureprovided by the noninverting arrayof ommatidia (an ommatidiumpointing forward sees
a portion of the environment situatedin front of the animal, one

ror I TbeEyeof the Fly

pointing backward seesa posteriorportion of the visual field, and
so on) unlessit were first re-rotatedby tSoo in the fiber bundle
proiectingto the ganglion (figures25, z6).
There is even more precision to be discoveredin this system:
retinulaeof neighboringommatidia have their lines of sight so
orientedthat eachis parallel,with greatprecision,to anotherline of
sight in eachof 5 neighboringommatidia (+). This meansthat 7
retinula cells of 7 different ommatidia receivepreciselythe same
visual information. (Here I simplify slightly, leavingout retinula
cell number 8, which would complicatethe issuebut would not
changethe argument.)The law of the retina-to-lamina-proiection
this: all the elementsthat look at the samepoint of the visualfield
sendtheir axon into the samecompartmentof the ganglion(figure
z5) (s).

Figure z5
Diagram of the eye and visual ganglia of the fly. Co is the cornea, a set of
lenses (in reality about 3,ooo). Re is the retina, with three light-sensitive
elements ("rhabdomeres," in reality 7 f.or each lens) arranged with their
tips in the focal planes of the lenses.Each lens projects an inverted image:
the complicated weave of the axons below the retina, Rax, compensatesfor
this and reconstructs the global picture on the first visual ganglion, La.
There are further inversions of the picture in the outer chiasm (Che) and
then again in the inner chiasm (Chi) benveen the second (Me) and the third
(Lba, Lbs) visual ganglia, but these are not readily explained on functional
grounds. Che is an example of an inversion that is not benveen the two
halves of the brain but happens separately within each half.


The rigor with which this principle is carriedthroughis especially
astonishingin exceptionalregionsof the eye,suchas near the margin (where an ommatidium has fewer neighborsthan elsewhere)or
near the "equatorr" wherethe arrangementof the retinula changes
abruptly. Horridge and Meinertzhagendedicateda very diligent
study to the precisionof this wiring and found absolutelyno exceptions (6). It is easyto convinceoneselfthat learningplaysno parr in
the establishmentof this type of connectionbecauseone finds the
whole arrangementreadymadein the late stagesof pupation, long
beforethe compoundeyehasever receivedvisualinput (exceptfor
subduedand diffuselight, which may filter throughthe involucreof
the pupa).

Olfactory orientation: control of behauior

by symmetricalreins (Vehiclesr to 4)
It is nice to seein the precedingexamplehow a bit of
physics,the geometricalopticsof a lens,is incorporatedpreciselyin
a nervenet. But I havealsoarguedthat the lensin the veftebrateeye
provides no convincingexplanation for the crossedrepresentation
of the world in the brain (Z). I proposeda different explanation,
which takes as a starting point the one senseorgan that has an
uncrossedrelation with the cerebralhemispheres,
the senseof smell
(figure z7).Each of the two olfactory tracts (the bundlesof fibers
carrying signals from the nose to the brain) goesstraight to the
cerebral hemisphereon the same side. The connectionsfrom the
hemisphereto the motor systemare crossed,however,which means
that a certain smell has a stronger effecton the motor systemon the
sideoppositethe nostril it hits first or more strongly.This brings to
mind schemeslike thosein Vehicleszb and 3b, with all the properties we discussedthere.
The most important nervouspathway in our primordial verte-

rol I Olfactory Orientation




I'r \


Figure z6
Explanationof the fiber pattern Rax berweenretina and brain. rlt) zt4,
projec,4j arepointsin visualsPaceseenby threeadjoininglenses.Their
the originalorder
tion is in the order tzr,4tL,543.The fibersreestablish
rzt45brate ancestors may well have been that between the nose and a set
of muscles used for locomotion, since in the water the business of
following chemical gradients is certainly important. The details,
however, are not clear. First of all, we don't know what kinds of
motors these primitive vertebrates used. If they were propelled
pair of fins, the case is analogous to that of our
primarily by
vehicles of type z, in the sense that the thrust produced by the
motor on one side of the animal (or vehicle) makes it turn toward

ro 4 | B I O LO G IC AL N O T E S O N T H E VE H IC LE S


Figure z7
Crossing of visual (V) and tactile (T) input. Also the motor output
crossed (M). Only the olfactory input (O) is uncrossed.

the opposite side. The contrary is true for a fish, which relies mainly
on a bending of its body for locomotion. In this caserhe contraction
of the muscleson one side results in the animal turning to the same
side. Also, it is not clear whether it is advantageous for a fish to
turn toward the sources that activate its locomotion, a supposition I
used to explain the crossing between the olfactory input and the
motor output. Be this as it fr?y, this sort of explanation of the
crossedproiection of the world in the brain may have its merits. It
draws on a large body of observations on animal orientation and
locomotion under the influence of various chemical and physical
stimuli, with the older work well summarized in Fraenkel and
Gunn (8). one of the originators of this tradirion, JacquesLoeb, in
a succession of books propagated a mechanistic approach quite
similar to that of our vehicles (q). The outburst of zoological work
in this field was largely prompted by negative reactions to his ideas.
Besides Vehicles z and 3, the very simple, so to speak, one-


I ObiectFixationin Flies

dimensionalbehavior of Vehicle r has biological counterpartsin

both the older literature and somerecentwork on bacteria (ro).

Orientation and obiect fixation

in flies (Vehicle4)
Vehicle 4, with its nonlinear relation between sensory
input and motor output, also brings to mind someneurophysiology. Such input-output characteristicsare quite common at all
levels.Movement of an object is not perceivedvisuallyif it is too
slow or too fast and receivesan optimal responseat a certain angular velocitythat is well known for fliesand for men (rr).
When the nonlinear relation of input-output is further complicated by varying characteristicsof a set of detectorsdepending
on their position in a sensoryfield, the ensuingbehavior may becomequite complicated.Or, the other way around,it is sometimes
possibleto explain astonishinglycomplexbehavior,such as that of
a fly navigatingthrough a room and landingon a hanginglamp, by
invoking nothing but a setof almost identical,rather simple movement detectorswhoseoutput, weightedfor position, convergeson
a few motoneurons.This idea appearsin some recent work by
Reichardt(r z) and hasits precursorin earlyexplanationsof phototropism(rl).
Consideragain the compoundeyeof an insect.We have already
seenthat it is composedof a greatnumberof almostidenticalunits,
eachwith its lensand associatedsensoryand'neuralapparatus.It is
a fact that in many insectsmost of the information that entersthe
brain from the eyeis not about where light and dark spotsare in the
visual field but about where something moves and in what direction, independentof what it is that moves.This is of courseimportant information when an insectwants to control its own position
in its visual environmentduring flight. Rotation of all points of the


visualfield around a certainaxismost likely signifiessimplythat the
animal has itself been turning (in the opposite direction) around
that axis.
Forward movementof a flying fly producesa more complicated
visualflow field: the panoramastreamsthrough the visualfield in a
backwarddirection, with the flowing motion seeminglyemanating
from the point of forward profectionof the direction of flight. But
the velocity of the flow dependson severalfactors, including the
angle the line of sight forms with the direction of motion of the
animal and the distancefrom the eyeof the variousobieas forming
the visual environment.The measurement
of the velocityvectorsin
of the visual field of a flying insectobviouslyprovidesa
great deal of information, but we are hard put when we try to
invent schemesthat would extract the relevantinformation for the
life of the insect,and make useof it.
Once again,it may be simplerthan it looks: anotherinstanceof
the law of uphill analysisand downhill synthesis.One of the observations about flying flies was their tendencyto navigaretoward
isolatedobjects on a homogeneousbackground.They do this in
very"coinplicatedexperimentsby Reichardt.And they also do it in
real life when they settle on a branch or on somebody'snose.A
simpleexplanation is this (figurez8). Generally,perceivedmotion
in the visual field makesthe fly turn in the direction of the motion.
The effect of the perceivedmotion may be different for different
directions, however. Say, motion of an obfect in the backward
direction in the right half field makesthe fly turn toward the right
more vigorouslythan the forward motion of the sameobjectin the
sameposition makes the fly turn toward the left. Try to imagine
what happensif that obiect wiggles (or if the fly's head vibrares,
which has the sameeffect).With everywiggle toward the back the
fly turns toward the obiect a little more than it turns awayfrom it
with the wiggle in the opposite direction. In the end the fly will be

ro7 | Obiect Fixation in Flies

Figure z8
The turning tendencyinducedby movementin two oppositedirections
may not be of the samemagnitude(representedby the different lengthsand
directions of the arrows). This will produce a tendency to turn toward
wiggling objects on a stationary or void background. The differencebetween the forward and backward reactionmay vary accordingto position
in the eye (arrows).

In reality things are slightly more complicated but still of the

same nature. It seemsthat the difference benrveenthe effects of
forward and backward motion variesin a systematicway over the
visual field. Thus for everypatternin the visual field therewill be a
net turning tendency,compoundedout of the many contributions
of turning tendencyfrom eachpoint of the visual field. The complicatedtrajectoryof a fly in your room may be, in a way, a peculiar
sort of imageof that room, the velocityand the maneuversof the fly

being completely determined by the initial velocity of the fly and by
the distribution of visual detail in the environmenr.
Another well-known instance of nonlinear input-output relations is apparent in the reactions of many animals to the sight of
other animals or moving objects. This dependsin a curious manner
on the size of the other animal (or object): small specimenselicit
prey-catching behavior, very large ones elicit flight, and intermediate-sizeobjects are examined in more detail. Somethingof this
sort has been shown even at the level of electrophysiological studies
of single neurons in the visual sysrem of the toad (r4).

McCulloch-Pitts neuronsdnd
redl neurons(Vehicle5)
Vehicle 5 is, of course,an embodimentof the old McCulloch and Pitts "Logical calculusof the ideasimmanentin nervous
activity" (rf ). This was one of the great boostersof modernbrain
science.Its experimentalbasisis in electrophysiological
the spinal cord.n The influenceof one input nerve ("posterior
root") of the spinalcord on one output nerve("anteriorroot") is
under certain conditions "monosynaptic": the fibersof the posterior root directly contactthe motoneuronsfrom which the fibersof
the anterior root originate.
*A glossary may be helpful for readers who are not trained in
the biological sciences. Neuron: a special kind of cell devoted to signal transmission in the nervous
system. Dendrites: usually ramified appendages of the neuron, which carry signals
toward the central part of the neuron. Axon: a single usually ramified appendage of
the neuron which carries signals away from the center of the neuron. Nerue: a
bundle of axons. Synapse: the place where the axon of one neuron transmits signals
to a dendrite (or cell body) of another. Motoneuron.' a neuron of the central nervous
system connected to a muscle. Sensory flet4ron: a neuron directly connected to, and
influenced by, a sense organ.

rog I McCulloch-Pins Neurons


the cooperation of various input nerves in the activation of

spinal motoneurons was analysed, three facts emerged. They turned
out to be fundamental discoveries about the computational properties of synapses,even before the techniques of electrical recording
of single neurons were developed (r6). For some of the motoneurons, the conjoined activity of severalinputs is necessaryin order to
activate them. In other cases the fibers of one input nerve are by
themselves sufficient to reach the thresholds of the neurons. And
finally, a third kind of situation seemedto imply that some fibers
inhibit the motoneurons, in the sense that their activation from
other sources is made ineffective. These inferences from macroscopical input-output experiments were later confirmed with microelectrode studies; they were explained as consequencesof the
electrical properties of the neural cell membrane and of the influenceof chemical transmitter substanceson these properties (rz).
In their famous paper McCulloch and Pitts srylized the functional
relations of neurons connected by synapses as the fundamental
operations of the calculus of propositions: conjunction, disjunction, and negation (and, or, andnot). These are fundamental in the
sense that they were the first logical relations to be used for that
purpose in antiquity by the Greek philosophers. But they are not
unique; many other sets of such fundamental relations would do, or
even one single relation (there are two relations with this property:
not botb, and botb not), which could of course be called fundamental with much more right.
Is it an accident, then, that conjunction, disjunction, and negation were first defined by the philosophers and then rediscovered as
fundamental properties of neurons and synapsesin the spinal cord?
Or is the nervous system really constructed out of these operations,
with the consequence that the philosophers can only discover in
their own thinking the laws that make their brains tick? Or did
Sherrington describe the phenomena of facilitation, occlusion, and
inhibition in terms that were subconsciously suggested to him by


the philosophicalteachingsto which he wassubjectedin his schools
and perhaps implicitly through ideas incorporatedin the English
language?I haveno answer.
The McCulloch-Pitts theory of nervenetsis one of the roots of
the theory of automata(r8), so much sothat in the earlyyearssome
people who really had computing machinery in mind used the
words "neuron" and "synapse"and drew diagramsthat were originally intendedto depict real nervenetsin animalbrains (rg).
It was indeed practical to speak of neurons and of threshold
devicessynonymously,but there are good reasonswhy I preferred
the latter terms in the description of the vehicles' brains. Real
neurons have propertiesthat go far beyond the simple threshold
deviceswe used as building blocks for our vehicles'brains. True,
the most important signal by which patternsof activity are representedwithin animal brains is the "action potentialr" an explosive
event that happensin its entirety or doesnot happen at all and,
when it happens,is propagatedwith undiminishedintensity along
the fibers leadingto other neurons.This obviouslyimplies the concept of thresholdbecausea certainminimal intensityof excitationis
reqpired to set off the explosiveevent.But it is debatablewhether
thdsethresholdsplay the role that we assignthemwhen we think of
logical computationby meansof thresholddevices.First df all, it is
difficult to imagine such computation without a clock that keeps
ptrict order in time. In the McCulloch-Pitts theory, as in digital
domputers,the temporal coordinateis represented
by a sequence
discreteinstants,with all the changesin the activity of the nenvork
happeningbetweenone instant and the next.
In real brains this is hardly so. The exactpoint in time at which
an action potential arisesin a neuron dependsnot only on the time
at which the excitationreachesthe neuronbut alsoon the intensity
of excitation (figure z9). Just as the potential acrossa condensor
reachesa certainvalue faster the strongerthe current that charges
the condensor,the critical levelof the potentialacrossthe nervecell


I McCulloch-Pins Neurons

Figure z9
Dependenceof the action potential on the intensity of stimulation. The
shapeof the rapid excursionof the potential E (abovethe dashedline) does
not changewith different intensitiesof stimulation, while the chargingtime
required to meet the threshold is shorter for higher intensity stimulation

membrane(which triggersthe spike)is also reachedmore quickly

when the excitationis strong.Thus the amount of excitationabove
the thresholdis lost in the ordinary thresholdelementof a comPuting'devicebut not in the brain where it translatesinto the time of
occurrenceof the action potential. A consequenceof this is the
desynchronizationof action potentialstriggered by synchronous
excitationin a block of nervetissue.Whether two spikeswill meet
or not at a certainsynapticiunction, and hencewhetherthe logical
operationperformed by that junction will occur, may dependon
just theseunwanteddelays(figure3o). The simpleinterpretationof
a nervenet as an automaton with a fixed structure, operating synchronouslyon a discretetime scale,thereforebecomeslesslikely.
This is not to say that neuronsmay not occasionallytrigger allor-nothingreactions.Very quick actions,suchasoccur in situations

of danger or in reaction-time experiments in a psychological laboratory, or in sports, must be governed by sequencesof very few
action potentials in the neurons of the motor system.tUfith neurons
producing action potentials at a frequency of the order of ro or roo
per second, the reaction to a stimulus that occurs in less than o.r
seconds must be triggered by the first, or by the first few action
However, in many other situations, well studied by neurophysiologists, the signal within the brain corresponding to a sensory
stimulus is a burst of action potentials rather than a single action
potential. In such bursts, very commonly the frequency varies with

Figure 3o
The effect described in figure z9 applied to a small nerye net. In this figure
either A, B, or C alone, or two or three of them in combination, activate
the interneuron 11 which has a threshold equal to one unit of excitation.
But the more input elements that are active, the sooner interneuron 11will
produce its potential. Thus when A, B, C, and D are active at the same
time, coincidence of the output of 11 and 12 at E may not occur any more.
This is to show that threshold elements on a discrete time scalemay not be
an accurate description of neurons.


I Euolution

the intensityof the stimulus. Borrowing terms from computerengineering:there is an analogueprinciple involved in this which is
quite foreignto the digital operationof the McCulloch-Pitts nerve
net, or to the automation of automata theory. I(e are far from
understandingthe code or, what is more likely, the differentkinds
of codesthat are usedbet'weennervecellsin the brain. One point of
information theory, however, remainsvalid: all messages
can be
representedin theory by discretesignalson a limited number of
elements.This is the reasonwhy the vehicles'brains, made out of
thresholddevicesvery different from live neurons,may still display
somevery lifelike propertiesof information handling.
Most readerswill have recognizedthe vehicleleavingmarks on
the beachas a very elementaryversionof a Turing machine.For
thosewho are not familiar with this concept,I recommendeither
Turing's original articles or Minsky's book of. t967, or the very
friendly introduction in J. Sampson(zo).
The very last sentenceon Vehicle 5 must by necessityremain
cryptic,becausethe idea is not yet fully worked out.'What I refer to
is the increasein computing power of a brain that is endowedwith
the power of learning.No doubt this makesit possiblefor the brain
to write down its own record in the pattern of its interneuronal
connections,and then read it off again.But the way this is done is
very differentfrom the way a Turing machineworks with its tape,
printing, and readingheads.

Euolutioz (Vehicle6)
The gamewe are playing to generatethe vehiclesof type
6 hushesup most of the complexity of Darwinian evolution. My
aim was not to make propagandafor this theory. It is all too obviously correct for the people who are enchanted by its power
of demystification,while others will foreverinvent difficultiesand


The SelfishGeneby Dawkins(zr) is a book on
evolution that should appeal to psychologists.It is not weighted
down by an obscuredesirethat biology shouldnot arisefrom physics after all. Of coursethere are the classicsof evolutionarytheory
(zz). Preparedby Dawkins, even the reader with no interest in
biology will enfoythe information-generating
capacityof the evolutionary processin an adventurein psychology.

Memory (Vehicles7, ro, rr, r 2, rj, 14)

Beginningwith Vehicle 7, we considereda property of
nerve tissueunmatchedby anything in present-daytechnology,the
distributed memory acting on the logical structureof the network
itself. In nearly all technicalrealizations,includingelectronicsimulations of nerve nets, information that goes into the memory is
depositedseparately,outsideof the computingmachinery,often in
equipmentthat is entirely different from that doing the computing.
This is becauseneither Mnemotrix nor Ergotrix wire are commercially available.Indeed,if an engineerreadsabout our vehicles,I
am sure he will be irritated by the glib way in which I haveassumed
the feasibilityof somethingwhich to him would appearas the main
technicalproblemto be solved.oHowever,I am not alone.Uttley's
"conditional probability machine" assumeselementswith properties similar to a pieceof Ergotrix wire (z), andSteinbuch's"Lernmatrix" doesnot work without Mnemotrix junctions(24).
Thesemodelsand many others,notably the very influential (verbally formulated)modelby D. O. Hebb (zj), wereall createdunder
the impressionthat "association" is the most important principle
by which information about the environmentis incorporatedinto
* ProfessorStefanoCrespi Reghizziin Milan did read the manuscriptand was irritated. I thank him cordially for his comments.


I Memory

the brain. When thingsoccur together,the neuronsthat signaltheir

occurrencewill also be somehowconnectedin the brain. Is this
assumptioncorrect?RichardSutton and Andrew Barto (26) argue
that it is perhaps too simple an assumption,in view of emerging
information about the complexity of the operation of individual
neurons.And they arguethat associationmay not be sufficientto
producethe effectsthat must be explainedin cognitivepsychology.
\U7ehave already discussed(in Vehicles r 3 and 14) the aspectof
prediction,which theseauthorsstress.
At this point we ask insteadwhether there is any direct physiologicalevidence,basedon microelectrodestudieson singleneurons,
which makes the phenomenonof associationmore concrete,evidence beyond the almost inescapablebut indirect assumption
derived from psychology.The answer, since Hubel and Wiesel
(27) is yes, there is suchevidence,at leastof this form: artificially
inducedsquint in kittens, which disrupts the normal cooperation
betweenthe rwo eyes,has the effect that someof the normal connectionsbetweenthe eyesand the cortical nerve cellswill not be
formed.Apparently, the pattern of theseconnectionsis molded by
The principle that can bestexplain theseobservationsis the following (figure 3r). A cortical nervecell that is at first diffusely but
weaklyconnectedto a largenumberof input fibersfrom both eyes,
with time and experiencepicks those fibersfrom the right and the
left eyethat mostly carry the samesignals.The cortical nerve cell
then makesstrong connectionwith them at the expenseof the other
input fibers.This way it is assuredthat individual nervecellsof the
visualcortex receivesignalsfrom correspondingportions of the nvo
retinas and hence from the same point in the visual field. The
principleof associationis apparentin this: relatedactivity leadsto
the making of.a connection.At a more macroscopiclevelthe physiology of associationwas establishedbeforethe introduction of the
microelectrode(zS). The pairing of electricalstimuli to different




from the leIt eye

input flom the right eyc


Figure 3r
Refinementof the projection of visual input onto the visual coftex by a
learning process.Fibers from both eyesreach the cortex in a rough topographicalorder, such that bundlesof fibersfrom correspondingplacesof
the two eyes are intermingled in the same compaftmentof the cortex.
Subsequentlyindividual cortical neuronspick fibersfrom both eyeswhich
are mostly active at the sametime (X) and make srrongconnectionswith
them (dots). Thus cortical neuronsbecomeconnectedto retinal elements
having exactly the samecoordinatesin the right and left eyes.
parts of the brain had the consequence that one of the two loci
yielded behavioral effects of stimulation previously associatedonly
with the other locus.

In searchof an engram: the anatomy

of memory (Yehicles7, rr)
In a way these results are quite obvious and could be
expeited. Granted that signals are carried by fibers and synapses
from the senseorgans through the brain to the muscles,how would

rr7 | TheAnatomyof Memory

we explain that a few neuronsin the acousticcentersof a rat's
brain, respondingto the soundof abuzzer,at first haveno influence
on the motoneuronsof the rat's forepaw but after sometraining
regularly lead to movementthere, if not by supposingsome anatomical rerouting, that is, changesin the synapsesof the nensork?
It is an entirely different matter if we ask what exactly has happenedin the tissue,and at what level. There are still peoplewho
think that the growth of new dendritesis involved,or the degeneration of alreadyexisting ones,while othersprefer to think in terms
of axonalgrowth or degeneration.It is fashionablenow to saythat
the changesprobably take place in already existing individual
synapsesthat are ready to learn. But this hardly explains the
changesin the sizeof the brain, which someclaim are associated
the acquisitionof information: the more information is acquiredby
the brain, they say, the biggerthe brain becomes(29). In fact the
synapsesoccupy only a tiny fraction of the volume of the brain.
It is no longer fashionable,luckily, to imaginethat the information of complex experiencesresidesin individual moleculesof the
brain. Thereis of courseplenryof ribonucleicacid in the cell bodies
of nervecells,and it is not quite clearwhat it doesthere.But this is
not a good reasonfor supposingthat its large information storage
capacity,normally devoted to geneticinformation, in the brain
codeshappeningsof the individual's life. This idea is actually irritating sincethe additional mechanismsit implicitly requiresare
more complicatedthan the facts it intends to explain. How is the
information about the face,the name,and the utterancesof somebody whom I iust met distilled down to the minusculecodeword
that fits into one moleculeof one cell (which one?)in my brain?
And worsestill, how is the macroscopicpattern of actionpotentials
in nervecellsthat signalsmy meetingthat person againcompared
to the minusculetrace left earlierso that I may be ableto recognize
him? tUfillit end up in the samecell in a parallel strandof ribonu-


cleic acid? And how do I get the information out of the molecule
when I want to describethat person?
The experimentalapproachesto the questionof the anatomical
nature of the engram are all marred by a fundamentaldifficulty.
(Engramsare the memory traces postulated many years ago by
psychologists,long beforetherewas any hopeof everfinding one in
the brain.) Supposewe have some idea about the anatomical
changesresponsiblefor memory and want to prove it. We present
someinput to one animal, but not to another animal, in order to
useit as a control. It is not that the control animalhasexperienced
nothing while the experimentalanimal receivedits input: it hashad
its own experiencesand its own thoughts.In order to comparethe
tracesleft in the brains of the two animals by the two different
inputs, we would have to know exactly where the information
endedup in the rwo animals.In truth we don't. Most likely, two
inputs that may haveentirely different meaningsare representedin
the brain in quite the sameway, as diffusepatternsof activity in an
enormousnetwork of neurons.
Supposethe engram were embodied in changesthat are very
easily visible in the electron microscope,or possibly in the light
microscope:a changein the thicknessof axonalterminals,a change
in the number of synapticvesicles,or a changein the amount of
pre- or postsynapticthickening.No matter what kind of information we presentedto the animal, in the end we would expectto see
somesynapsesof one kind and someof the other, for information
must be representedin a pattern of elementsin different states,or
else it wouldn't be information at all. But different patterns can
only be distinguishedif they are understoodin everydetail.In other
words, they cannot be distinguishedat the presentstageof our
knowledgeof the brain.
There is of coursethe possibility of imposinginput of a brutally
abnormal kind, by keeping an animal entirely in the dark or by
keeping one of its eyesblinded. In thesecases,which are called

rr9 | TheAnatomyof Metnory

deprivationexperiments,anatomicalchangescan indeedbe found
in the brain, but it remains questionablewhether they are of the
samekind as the changesunderlying memory in the normal upbringingof an animal. It would not be surprisingif the development
of the brain iust doesnot happenin the normal way in a mousethat
neverseesthe light of day, perhapsfor reasonsconnectedwith an
abnormal condition of the hormonal systemrather than lack of
There are some animals,however,in which the deprivationexperiment,so to speak, is a part of normal development,and the
controls are also furnished by nature.t07hatI have in mind is the
comparisonof brains in relatedspecies,
suchas rabbit and hare,rat
and guineapig-of which one is born very immature and the other
is born at a much later stagein its embryonicdevelopment.Ratsare
born tiny, naked,blind, helplesscreatures,while guineapigsresemble the adult animal in their appearanceand their behavior immediatelyafter birth. If the developmentof the brain is studiedin
the nvo speciesfrom early stageson (3o), it appearsthat there are
no greatdifferences,exceptthat the exit from the uterushappensat
a later date on the developmentalcalendarin one casethan it does
in the other.
There are some stagesin the developmentof the rat brain that
happenafterbinh and can be influencedby environmentalstimuli:
this finding was sometimeshailed as a paradigmof the anatomyof
learning.The same episodein the developmentof the guineapig
brain takes place when the fetus is entirely (or almost entirely)
shieldedfrom environmental influencesin the maternal womb.
Thus the structuresthat develop at that time (for example,the
dendritesand axons of cortical neurons, dendritic spines,most
synapses)at least in the guinea pig and presumably also in the
normal rat do not encodemessages
from the environment.
We haveto take a closerlook (lr)r the anatomicalchangesthat
subservememory must be finer than that. Schiiz presentssome


good candidates,subtledifferencesbeffMeen
the biologicallymature
but psychologicallyinexperiencedbrains of newborn guinea pigs
and the brains of adult experiencedanimals: differencesin the
shapeof synapseson electron micrographsand differencesin the
number of synaptic vesicles,as well as quite macroscopicdifferencesin the shapeof the "dendritic spines,"which carcymost of the
synapsesin the cerebralcortex. What makesthesechangesgood
candidatesfor memory traces is that the varianceincreaseswith
age.We would be disappointedif somechangein the structureof
synapsesaffected all synapsesof the cerebralcortex in the same
way. \U7ewould then call it an effect of agingrather than of learning, since memory tracesought to differentiatebetweenneuronsto

Maps and their use (Yehicles8, g)

It is all too obvious that Vehicles 8 and 9 have not
from creativefancy. They incorporatethe one aspect
of animal brains that has beenthe main themeof brain researchfor
the past hundredyears:the representationof externalspacesin the
spatial coordinates of the nervous system.We used to think in
the past that suchmapsof the world (or of the body surface)in the
brain were a prerogativeof the primary sensoryand motor fields,
for example,in the cerebralcortex. But recentlymore refinedtechniques have revealeda successionof visual, tactile, auditory, and
motor maps, coveringmost of the availablespacein the brain (lz).
One wonderswherethe sort of computationthat is not relatedin
any obvious way to geometricalspacetakesplace.Intuitivell w
have no use for z- or 3-dimensionalCartesiancoordinatesin the
context of languageor in the abstractworld of conceptsrelatedby
a multitude of associativeconnections,nowadaysoften described

rzr I Mapsand Their Use

underthe heading"semanticnets" (ll).Intuition may of coursebe
misleadingin this field,.as we are told by Lieblich and Arbib (f +)
and also by someof the discussants
of their paper,who point out
that simplecartographycannotbe the whole story evenin portions
of the cerebralcortex that are clearly related, point by point, to
somesensoryspace.Maps aremeaningless,
they warn us, unlesswe
have a processfor using them. The concept of the world graph,
which they propose,makesthe distinction lessdrasticbetweeninformation handling in geometricallydefined spacesand information handling in the abstractspacesof languageand the like.
Thereis no doubt in my mind about the functionalimportanceof
theseorderly representations,quite in the spirit of the tricks described in Vehicle 8, although in theory other explanationsare
possibleon embryologicalgrounds.If the problem is to connecta
million sensecellsto a million cellsin the brain, one of the simplest
solutionsis of courseto let a whole bundle of fibersfind its wxy,
insteadof specifyingthe addressfor eachindividual fiber.The preservedorder in the projection may just be due to the preserved
neighborhoodrelationsof the fibersin the bundle, and it would be
idle then to speculateabout the functional meaningof the resulting

This cannot be the whole story,however.The tricks requiring an

internal representationof neighborhood,which we introduced in
Vehicles8 and 9,have clearlybeeninspiredby functionalprinciples
known to operatein animalbrains.A greatdeal is known about the
of movementdetectorsin the visual systemsof various animals, including flies, as we have already seen.Cells that
respondto moving stimuli havebeenidentified in the retina of the
rabbit (lS). In a beetle(Chlorophanus)the propertiesand the arrangementof a set of visual movementdetectorswere definedin a
quantitative way by Hassenstein,Reichardt, and Varju $6), although the correspondinghistology could not be identified with
certainty. In the fly (lZ), much more is known now about the


various levelsof integrationin the visual ganglia,includingsome
neuroanatomyrevealingfiberpatternsof suchstupendous
that they seemto be taken out of somemechanicalvehicle'sbrain
Lateralinhibition alsohassolid experimentalfoundations.Since
it was discoveredin human visual bil, auditory,and tactile (+o)
perception,it was also describedas a principleof neuronalinteraction in the eye of the horseshoecrab Limulus Polyphemus(4r)
and after that in all too many other situations.Its simplicity and
powerful information-handling properties invited mathematical
formalization (42) and various speculationson its role as a basic
computationaldevicein central nerve nets,such as the cerebellum
(43) and the cerebralcortex (++).
A word about the idea that led to the consrrucion of figure r 5:
networks may be symmetricin any number of dimensionsand still
be housedcomfortably in the 3-dimensionalspacesurroundingus,
or even in the z-dimensionalspaceof a drawing. This was just
intended as a warning to neuroanatomistswho cannot abstract
from what they see.It is conceivablethat the exact analysisof a
pieceof nervetissuemay reveala connectivitynot at all apparentin
the externalshape,for instancea truly 4-dimensionalnerworkcompressedinto an ordinary 3-dimensionalbody. But I know of no
What happensis that occasionallya sensorymanifoldof more than
two dimensionsis proiectedonto the usualkind of cortex-likenerve
net which, for all we know, is essentiallyz-dimensional.Thereis a
well-known example in vision. Although eacheye receivesa twodimensionalpicture of the visual environment,the combinationof
the two picturesprovidesinformation about three-dimensional
visual space.And indeed,the,two picturesare brought togetherin
one and the same piece of cortex, the visual cortex, where this
information is presumably extracted. But before 3-dimensional
spaceis reconstructed,the t'wo picturesare projectedonto the corti-


I Mapsand Their Use

cal surfacein a peculiar wxy, with narrow stripes of the picture

from one eyealternatingwith stripesfrom the other eye,all on the
sameplane,sharingthe samez-dimensionalcoordinatesof the cortex (45). We still don't know how the third dimensionof space,
which is lost in the projection, is later regained through stereoscopicvision and where it is represented
in the brain, but it seems
almost certain that its representationis not orthogonal to that of
the other two dimensions,that is, to the plane of the visual cortex.
One savingthought: if the detectionof continuoustrajectoriesis
one of the points of the orderly representationof sensoryspaces,
the lossof one dimensionin the projectiondoes not matter much,
sincea continuoustraiectory in the original spacealwayshas as its
imagea continuousline in the proiection,and a discontinuousline
has a discontinuousproiection most of the time.

Shapes.The morphemesof uisual

Vehicle9 is especiallydedicatedto the memory of Gestalt
psychology.Under this denomination,which means nothing but
the studyof the conceptof shape,a group of brilliant psychologists
during the first third of this centurysetout to discoverthe laws that
make similar shapeslook similar to humans (+6). How right they
were in making an issue out of this problem becameclear to
everybody,including computer engineers,when, much later, they
tried to constructefficientmachinesfor the discriminationof forms
(+z) (enemyairplanes,handwrittenaddresses,
turbulentor nonturbulent cloud patterns).
Gestalt psychologistswere not so successful,however, in their
attempts at relating their discoveriesto functional principles of
brain physiology.Not enough was known at the time about the
neuronsand their connectionsin the brain, and what was known


was often presentedin a form that tendedto obscurethe computerlike aspects.Today much more is known about the brain, but
progress is slow in this field and we still have to rely mostly
are iust beginningto grasp some of the
on speculation(+8).
codewords the brain usesin categorizingshapes,the elementsof
meaning that we project into our visual environmentor, to use a
term frorn linguistics,the morphemesof visualperception.Here are
Clusteringis undoubtedlythe most fundamentalelementof form
perception,the most obvious morphemein the brain. The Pleiades
are perceivedas a unitary object in the sky becauseof the rather
uniform brightnessof a number of starsclusteringin that region.
And indeed,the morpheme"local densiry"in this casecorresponds
to a physicalreality,the gravitational couplingand commonorigin
of these stars. Another example: a number of sounds,all rich in
high frequency components,indicate the presenceof an animal
moving nearby in the underbrush.The neuronalactivity clustering
in the region of the acoustic system where high frequenciesare
representedis immediatelydiscoveredby other neurons,which relay signalsto an alerting systemin the brain.
Clusteringof neuronalactivity may be a factor evenafter several
stagesof abstractionfrom the sensedata, as when we immediately
perceivemovementof disparateobjectsin widely separatedparts of
the visual field when their movementis in the samedirectionand at
the samevelociry.This is the "common fate" phenomenonof Gestalt psychology,recentlyreproposedas a puzzlein neurophysiology by H. B. Barlow (+g). Here the clusteringis not in a region of
the brain where visual space is mapped but perhapsin another
region where we may supposean orderly representationof velocities occurs.
The detectionof clustershas a clear counterpartin neuroanatomy. The neurons in the brain are highly branched,star-shaped
objectswhose sizein many cases,notably in the cerebralcortex' is

rz1 | Shapes
larger by at leasta factor of ro than the separationof their centers.
Their dendritic trees are fairly uniformly covered with synapses,
severalthousandsfor eachneuron,through which they receivetheir
input. (Theyare also connectedto eachother.) Thus eachresponds
to the activationof a cloud of synapses
centeredaround it, and the
belongingto neighboringneuronsoverlapgenercloudsof synapses
ously. Somedendritic treesof smallerneuronsare evenfully containedwithin the dendritic spreadof largerneurons(figure3z). We


Figure 3z
Fromcajal, r9rr. Golgipictureof the upperlayersof the humanvisual
coftex. only a small percentage of the total neuronal population is shown,
all of them of the pyramidal kind. The size of their dendritic ramification
varies a great deal. only part of the apical dendrites are shown for some
enormous neurons of the lower layers, the spread of which greatly exceeds
that of the other neurons in the picture (ascending dendrites marked c).


realizehow we are able to seedensitiesof dots usingneuronswith
large dendritic spread at the same time as we resolveindividual
dots, and even the contours of individual dots, using the smaller
neuronsof the sameregion in the visual cortex.
There is evensomethinglike a neuronal"zoom" embodiedin this
structure. Figure 33 givesrise to the following observation.In the
etching above, the
reproduction of part of an eighteenth-century
smoothnessand curvature of the skin is admirably renderedby

Figure 33
The neuronal zoom effect. Ve integrate visually over the hatching, which
admirably rendersthe smooth skin in the engravingabove.But shifting our
glanceto the animalsbelow, we are ready immediatelyto count the legsof
the centipedeor to describethe shapeof the book scorpion'sclaws, details
far finer than the spacingof the lines of the hatching.

rz7 | Shapes
variations in the density of quasi-parallelblack and white lines.
Individual lines are seenonly if artention is drawn especiallyto
them. on the contrary, if we glanceat the legs of the centipede
below, the segmentsof the bodiesof the three arthropods, or the
shapeof the clawsof the book scorpion,all their structural details
are immediatelyrecognizedas suchwith the full spatial resolution
our eyesafford. Note that the periodiciry of the centipedeis narrower than that of the hatching aboveand the small numbers next
to the zoologicalillustrationsare no largerthan the spacingof the
lines of the shading.
This observationimplies that we are able to switch rapidly from
one set of filters to another, making available to the formperceivingmechanismdifferent bandsof the space-frequency
spectrum. In termsof the neuronsin the cortex (figure 7z),itseemsthat
setsof smallerand largerneurons(by a factor of at least 5) can take
over in the coding of the visual input, dependingon which set of
neuronsprovidesthe picture that makesthe most senseto the brain.
Another categoryof visual perceptionis the continuity of lines
and of traiectories.Like clustering,it is implicitly embodiedin the
structureof nervenets as we seethem under a microscope;it also
providesgood evidencefor the usefulness
of internal maps. It certainly is not difficult to invent a network of "neurons" with connections benveenneighbors(figure 34) providingfacilitation suchthat
the input becomeseffectiveonly if one of the neighboringelements
has receivedinput a moment earlier.Sucha network would give a
much strongerresponsefor a patch of excitationmoving smoothly
over its surfacethan for disjoint patchesor discontihuousmovement. This is a common type of connectivity(for example,in the
system of axon collaterals in the cerebralcortex), although for
somereasonthe facilitatingconnectionsbetweenneighborsseemto
be lesseasilydetectedin the electrophysiological
the inhibitory ones.

, vttupf

tl.w< :1 1n ,

r 2 8 | B T O L O G T C A LN O T E S O N T H E V E H I C L E S

Figure 34
A nerwork that responds to continuous trajectories. Neighboring elements
excite each other subliminally. They can be fully excited by the input (black
dots) only if a neighbor was excited a moment before. Thus only continuous trajectories (for example the one indicated by the stippling) are

It is clear that a network such as the one in figure j4 can provide

the most convincing clues for the distinction between real objects
and random noise or hallucinations, for the most common thing
that can be said about physical objects is that they move at reasonable speedswithout breaks in their trajectories. The disturbing
thing is that, at least in visual perception, the continuity of a line is
not necessarily detected at this trivial level. Kanizsa (So) has given
examples in which lines can be seenthat are not at all contained in
the pattern presented (figure 3 j). Th.y are apparently constructed,

rzg I Shapes




Figure 35
Contoursthat arenot presentin thepicturearereconstructed
of interpretation.
a: fromBrunswik,rg1'j; b: fromKennedy,
r974;c, d, ande: from Kanisza,
r974,all quotedin Metzger,r975.
in much the sameway for everyobserver,by someactiveprocess
that may have its roots partly in experienceand partly in inborn
mechanisms.We learn from this that it is somewhatartificial and
unnecessaryto draw a sharp line berweenperceptualand cognitive
It would be surprisingif it turned out that the visualcategoryof
bilateral symmetry is not relatedto the symmetricalbuild of the
two halvesof the brain and to their point-to-point connectionsin
the simpleway I suggested
in Vehicle9. This very strongelementof
form, mirror symmetrywith respectto a vertical line situated in
front of the animal (jz), has an obvious counterpartin neuroanatomy: the commissuresconnectingsymmetricalpoints of the


right and left brain. The most imposing commissure,the corpus
callosum, contains about zoo million fibers in man, about one
hundred times more than the fibers in the rwo optic nerves.By an
argumentof sheerinformation capacity,this systemof fibersmust
do more than simply comparethe visual input in the two halvesof
the visual field, which are proiected onto the rwo halvesof the
brain. No doubt bilateral symmetryis also an important properry
in other sensoryor motor contexts.Actually, the primary visual
area is one of the exceptionsin the general schemeof callosal
connections:it contributesvery few fibers to the corpuscallosum,
but the secondaryand tertiary visual areas,wherethe higher-order
figure analysisis sometimessupposedto take place,are abundantly
connectedby fibersberweensymmetricalpoints in the right and left
In somecaseseven the wildest speculationsdo not lead to satisfactory explanations.One good thing about computertechnology
is the possibility of immediatelytranslating speculationsinto machines.Their worth is therebyquickly revealedand the turnover of
ideas is increased.We can no longer fondle our ideasabout the
brain with the securefeelingthat their falsificationis beyondtechnical feasibility. Most ideas can be translated into computer programsand are thus easilyput to the experimentaltest.
And yet many aspectsof perceptionare still a mystery.Nobody
knows by what principle we are able to recognizewithout fail
individual human facesout of millions. Evenif we reducethe problem to that of the recognitionof profiles,we noticethat the perception and distinction of contours has by no means been fully
understood.Contours are extractedout of the original visualinput
most likely by the processof lateral inhibition, which is familiar to
us from the discussionof Vehicle8 (figure r4). No doubt they carry
most of the information we needin dealingwith the objectsof our
environment,as we all know from the useof drawingsas the most


I Shapes

Figure 36
Variouskindsof arrowheads.Their essence
is describedby the abstract
figureon the right.

widespreadmeansof nonverbalcommunication(at leastbeforethe

introduction of photographs).
But how contours are further analyzedin the brain is not at all
clear, exceptfor one thing: it probably is not done the way computer engineersdo it, judging from the meager successof their
attemptsto replacehumanobserverswith machinesin crucialsituations. I think what we humans do in the perceptionof contours
must embodyat leastthe two principlesillustratedin figures36 and
37.For one thing, we categofizeshapesroughly by the presenceor
absenceof appendages,which may interfere with our motor acts
when we deal with an obiect of that shape, and by the relative
positionof thoseappendages.
For instance,the dangerousfunction
of a barbed arrowhead is fully describedby the abstractscheme
that we call an arrowheadin technicalgraphicsand is quite independentof the detailsof the contour (figure36). On the otherhand,
we take in a surprisingamount of information about minute variations of curvature.We are able to detectimmediately the discontinuity in the secondderivativeof curvesthat are composedof four
arcs of circles, while something in our perception (the "inner
eye"?-the gazedoes no such thing) glides pleasantlyalong the
smoothlychangingcurvatureof the one true ellipse (figure37).

r 32 | B I O LO G IC AL N O T E S O N T H E V EH IC LE S

Which is the ellipse?

Figure 37
Detection of discontinuitiesin the secondderivative.Five of thesecurves
are composedof arcsof circlesof different radii. Only one correspondsto
a single algebraic expressionof second degree.It is easily identified by
everyone(from Scheffers,
rgrr) (53).

It is tempting to think of Hubel and Wiesel's line segment detectors in the visual cortex as the elements of a differential analysis of
curves in visual space (l+). I7e are told that in any small region of
the cortex about half a millimeter across there are representatives
of all orientations in the set of line segment detectors, and there are
some that see white lines and others that see black lines or even
dividing lines bet'ween white and black. Since half a millimeter
cortex in the central part of the visual field corresponds to little
more than the unit of resolution of the visual system, we get the
impression that besides location and color, orientation is another
dimension in which the visual input is coded at the elementary level.


I Structureof the CerebralCortex

But we know very little about the possiblemechanismsof interaction betweenthe neighboringline segmentdetectorswe must Postulate in order to explain our proficiencyin the detectionnot only of
curvaturebut also of changesof curvature(figure,7), The feature
detectorsof Hubel and'Wieselup to now explain only first, not
secondand third derivatives.

An inborn categoryof acoustic

form perception(Yehicles8, 9)
In acousticssomeof the inborn categoriesof perception
are well documentedin their relation to physiologicalfacts. To
most of us a melody played in different keys remainspractically
identicalto itself. This very astonishingfactiswell explainedby the
finding that in the cortex of the brain, as on a piano keyboard,
frequenciesare representedon a logarithmic scale(figure f 8) (SS).
The resulting translational symmetry for tone Patternscharacterized by constantfrequencyratios is one of the basicfactsof music.
It may reflect our ability to roughly recognizetheshapeof a solid
body by the acousticfrequenciesit emits when it is mechanically
solicited.This pattern is independentof the sizeof the obiect when
it is definedin terms of frequencyratios.

Structure of the cerebralcortezc(Vehicle rr)

We are getting to our more cognitivevehicles,numbers
ro to 14. From here on it becomesincreasinglydifficult to provide
direct justification for the vehiclesby pointing out exPerimental
facts about animal brains. Rather, the connectionis with both
kinds of psychology: seriousacademicpsychologyabout animal





Figure 38
An a priori of music: the logarithmic plot of acoustic frequenciesin the
brain. Above (from Tunturi, ry62): projection of the frequency scale
(numbersstand for kilohertz) on the middle ectosylviangfrus (MES) and
again on the anterior ectosylvian gyrus (AES) of the dog cortex. Both
proiectionsare linear on a logarithmicscalebetweenabout zjo and 8,ooo
(16,ooo) herrz. Doubling of the frequencycorrespondsto equal distances
on the cortex. Only for very low frequenciesthis relation breaks down.
Below (from Evans, ry68)z a similar plot in the ventral cochlearnucleusof
the cat. Logarithm of frequencyversuspositionin the nucleusis linear (55).


| Structureof the CerebralCortex

learningand behavioron the one hand, and the introspectivepsychology of thinking (of which we are all specialists)on the other.
However, the distinction between two kinds of association,
which is introducedin Vehicle r r, is not only appealingon philosophical grounds but may find an interpretation in terms of
neuroanatomy.To make this plausible,I will provide an introduction to the cerebral cortex by quoting from some of my recent
the numberof nerve
Accordingto somewhatdivergentestimates,
cells in both hemispheresof the cerebralcortex of man amountsto
about ro billion. The mafority of thesebelongto a type calledthe
pyramidal cell. It is characterized,amongother things, by an axon

exceedthe order of magnitude ro5, of which the largest number

belongsto visual input. It is difficult to estimatethe magnitudeof
fiber bundlesreachingthe cortex from other parts of the brain,
althoughthe numberof cellsin the thalamus,from which the greatest part of this so-callednonspecificinput to the cortex originates,
may serve as an upper limit. It does not exceed the order of
ro8. From this we may infer that the internal,cortico-corticalconnectionsof the cortex are at leastro times,perhapsroo timesmore
powerful than the connectionsof the cortex with the external
world. It follows that the cortex is a machinethat mainly works on
its own output or, to put it differently,works in a reflexivemode.
This great internal complexity, comparedto the complexity of
the input and the output, is characteristicfor the cerebralcortex.
The fact that the cortex of man (and of other mammals) is the
largestpiece of gray matter of the whole brain is related to this
complexity.Only the cerebellumcomescloseto the cerebralcortex




-s- e ' ",



with its surfacearea,but not with its volume.The optic rectum,the

most impressive"cortex" of lower vertebrates,
is far lesscomplex:
the numberof neuronsin the (frog)tectumis about the sameasthe
number of fibersenteringthe tectum.
There are gbod reasonsto considerthe most numerouscell type,
the pyramidalcells,as the basicneuronalequipmentof the cortex.
The great majority of the synapsesin the cerebralcorrex have pyramidal neuronson both the presynapricand postsynapricsides.It
is not entirely certain, but it is a fairly safe assumption,that the
connectionsbetweenthe pyramidal cells are excitatory. The reasons for this assumptionare the following:
r. The cerebralcortex (and especiallythe hippocampalregion)is
the piece of nervous tissue most susceptibleto epileptic activity
$Z).If enoughneuronsare activated,the most diversestimuli can
produce self-sustained
seizurelikeactiviryin the cortex. One way of
doing this is to make an electric current passthrough the tissue.
This presumablyexcitesindiscriminatelyexcitatory as well as inhibitory neurons. The fact that a seizureensuesshows that the
excitatory connectionsprevail over the inhibitory ones. It is reasonable,then, to make the pyramidal cells,the most numerouscell
type, responsiblefor the excitatory connections.
z. The fibersof the corpuscallosum,which are axonsof pyramidal cells, certainly make excitatory connectionssince they convey
epileptic activity from one side of the brain to the other ("mirror
focus" (S8)). Their excitatory nature has also been directly observedby electrophysiologicalmeans(Sg).
3. The axonsof cortical pyramidal cellsthat reachdistantplaces,
such as the spinal cord, make excitatoryconnections.
A pyramidal cell of averagesize(in the mouse)has about j,ooo
synapsesover which it receivesexcitation.This is shown by measurementsof the length of dendrites,by counting the number of
so-calleddendritic spinesper length of dendrite,and from the electron-microscopicobservationthat most "spines" receiveonly one

r17 | Structureof the Cerebral Cortex

synapse.The number of synapsesthat the axon of a pyramidal cell
with all its branches makes is about equal to this number. The
question arises as to divergence and convergence in this system of
synaptic connections bet'weenpyramidal cells. The question can be
formulated thus: from how many different neurons do the 5,ooo
afferent synapses of a pyramidal cell derive, and to how many
different cells does one cell distribute its 5,ooo efferent connections? The answer: from about j,ooo and to about irooo results from
geometrical considerations, particularly from the straightness and
the sparsebranching of the axon collaterals, which only allow multiple connections with the dendritic tree of another pyramidal cell
in the rare case that a collateral happens to run parallel to a dendrite (5o).
The overall picture is one of alarge cortical mixing machine that
transmits signals from every cell to as many as possible other cells
and inversely allows signals from many other cells to converge on
each cell.
The connections between pyramidal cells are collected in two
distinct systemsof fibers (figure lg). The fibers of the A-system are
the axons of pyramidal cells, which traverse the white substance
and enter the cortex again in different places in order to terminate
(mainly) in the upper layers of the cortex. There they make synaptic
connections with the so-called apical dendrites of other pyramidal
cells. The B-systetn consistsof branches of the pyramidal cell axon,
which stay within the cortex and make synaptic contact with the
so-called basal dendrites of neighboring pyramidal cells.
The assumption that both the A- and the B-terminals of the
excitatory pyramidal cell axon terminate mainly on other pyramidal cells has not been proved directly by electron-microscopical
observation, but it is inescapableon quantitative grounds. The bulk
of the postsynaptic sites are furnished by dendritic spines of pyramidal cells. The greater part of the axonal presynaptic specializations again belong to pyramidal cells. The majority of the afferents

r 3 8 | B I O LO G IC AL N O T E S O N T H E V EH IC LE S
special case I
ol fact.input

general case! $r Sr-conneaions

long range'A: short range'B

special casc 2
othcr sansory inpl

Figure 39
The skeleton cortex: pyramidal cells with long-range (A) and short-range
(B) connections. The olfactory input to the upper layers and other sensory
input to the middle layers are also shown (from BraitenberE, r97g).

of a pyramidalcell must come from pyramidalcells,and viceversa.

The main hypothesisabout the role of the pyramidal cells is
supported by indirect evidence:if among the afferent fibers of a
neuron therearesomethat often becomeactivesimultaneously,the
synapsesof thesefibersare strengthened.
I havealreadymentioned,
in the discussionof memory, the observationsupporting this assumption. The projection of correspondingpoints of the right and
left visual fieldson the visual cortex dependson a learningprocess
in which a fiberfrom eacheyeis evidentlyconnectedto one and the
samecortical neuron-presumably a pyramidal cell-in virtue of
rheir similar activity panerns (figure 3r) (6). Rauscheckerand
singer (52) showed that this happensaccordingto a rule quite
similar to the one postulated by Hebb.
I assumethat eachpyramidal cell is capableof discoveringcor-


| Structureof the CerebralCortex

relatedactivity among its afferentfibers all over its dendritic tree.

The constellationof afferentswhose synapsesare strengthenedby
the learning processconsistsin generalof afferentson the apical
dendritesas well as afferentson basaldendrites.Due to the connection of the apical dendritesto distant neurons(A-system)and that
of the basal dendritesto neighboringneurons (B-System),in each
elementarylearning processin one pyramidal cell the information
concerningthe condition of the whole cortex is brought into relation with information within the context of the area.
This can be further interpreted:the things of our experience,the
"terms" of the cortical representationare composedfrom different
detectedby the apicaldendritic
sensequalitiesand are consequently
treesof pyramidal cells as constellationsof activity in their longrange cortico-corticalafferents.On the contrary, the rules of the
evolutionand modificationof theseterms are more likely specified
in terms of individual sensorymodalities and are therefore contained within the confinesof cortical areas.It would then be the
businessof basal dendritesto detect theserules in the activity of
afferentsthey receivefrom neighboringcellsof the samearea.The
distinctionof nvo parts of the dendritic tree, typical of the cortical
pyramidal cells, accordingto this view reflectsthe logical distinction betweenterrns andrelatiozs benrreenterms. The unitary learning process,which we assumeto involve the whole of the dendritic
tree, implies that the learning of terms and the learning of their
relationscondition each other. This is the origin of the idea that
madeVehicle II smarterthan its predecessors.
If we assignto the pyramidal cells the task of learning sets of
afferentswith correlated activity, we may even derive from this a
role for the inhibitory "stellatecellsr" which seemto be preferentially located in the placeswhere the external input to the cortex
meetsthe pyramidal cells-in the fourth layer of the cortex and
particularly in the primary sensoryregions.Let us assumethat a
pyramidal cell can only learn to recognize (and therefore to


strengthenits synapseswith) setsof afferentsthat tend to become
activetogether.It cannotlearnto recognizea constellationof activity characterizedby somefibersbecomingactive,and somefibersat
the sametime remaining inactive.In fact, in order to learn a coniunction of suchnegatedand non-negatedrerms,the learningmechanism inside the neuron would have to be considerablymore
complicatedthan if it only had to recognizeconjunctionsof positive
terms. Still, most of the conceptswe learn consistof negatedand
non-negatedqualities: man is a featherlessbiped,a ring is a disc of
a particular material with absenceof that material from a central
region,and so on.
This difficulty is best dealt with by imagining that each input
fiber, besides reaching some cortical pyramidal cells directly
through excitatory synapses,inhibits others via inrerposedinhibitory interneurons.This makes available to the cortical learning
mechanisma set of pyramidal cellsstandingfor the corresponding
input being active, and another set signaling the corresponding
input being inactive. The learning processmay limit itself ro rhe
detectionof simultaneousactivity of membersof thesetwo setsand
thus conjoin negatedand non-negatedterms with the sameease
with which it conjoins non-negatedterms.
I recently proposed a model for orientation and directionsensitiveline detectorsin the visual cortex, which assignsto a setof
inhibitory neuronsin area r7 this role of switchingthe sign of the
input (61). This may be just aspecial caseof a generalprincipleof
the cortex.

Cell assemblies;embodiments
of ideas (Vehicles7, ro)
Vehiclesro ro 14 operarewith ideasthat standfor things
or situationsin their environment,and the ideasare represented

r4r I CellAssemblies
the brains of the vehiclesby groups of active elementsthat are
somehow tied together by reciprocal activating connections.\U(e
may ask: why are things not representedby single elementsor,
quite abstractly,by patterns of activity which are not constrained
by the condition that the active elemenrsexcite each other? The
answer is of course: becausethe vehiclesare caricaturesof real
brains and at presentit is again fashionableto think of groupsof
connectedneurons,so-calledcell assemblies
(64), as the carriersof
individual itemsof meaningor, if we wish, asthe morphemesin the
languageof the brain.
It is impoftant to realizethat there is no logical reasonfor this
interpretation. Supposethe point is internal representationof
thingsin sucha way that any largeenoughsubsetof the detailsthat
characterizea thing will be sufficientto evokethe thing in its entirety. This is undoubtedly a good principle, since it makes for
economicaluseof channelcapacityunder the assumptionthat the
existenceof things is the main redundancyin the world, with
"things" standing for bundles of details each of which, when it
presentsitself, raisesthe probabiliry for the rest of them to present
themselvesas well.* Under thesecircumstances
a Hebbian assembly of all the neuronswhich individually representthe detailsof a
thing is indeeda good codewordfor that thing, sinceit representsin
its internalexcitatoryconnectionsthe conditionalprobabilitiesthat
characterizethe thing itself. But a singleneuronthat receivesexcitatory synapsesfrom a set of afferenrsrepresentingthe detailsof a
thing is also a good codeword, vry much for the samereasons,if
its thresholdis set appropriatelysomewhereberweenthe value for
"all afferentsactive" and that for "one afferentactive." It, too, will
*If you want to grasp the meaning of terms such as
channel capacity, redundancy,
and conditional probability, you must read more information theory than I can pack
into a footnote. There are many introductory
Shannon and l7eaver's Mathematical

texts, one of the best still being

Theory of Communication



respondto subsetsof the detailsthat characterizethe thing, the size

of the subsetbeing determinedby the threshold.
The reasonswhy we think again in terms of cell assemblies
more empirical.Cell assemblieshaverecenrlygainedsupporrfrom
neurophysiologyin nrvo ways. First, many years of recording responsesof single neurons to sensorystimuli have shown that no
very complicated or very unique input is needed to activate a
neuron. The most efficient stimuli for cortical neuronsare rather
elementary configurationsof the sensoryinput, such as moving
lines in narrow regionsof the visual field (d5) or changingfrequencies in certain delimited regions of the acoustic spectrum (66).
These simple "features" cannot independentlycarry meaningbut
must be related to meaningful eventsin the same way as the
phonemes of linguistics are related to words or sentences.The
whole meaningfulevent must be signaledin the brain by a set of
neurons, each contributing a particular aspectwhich that event
may have in common with many other events.B.}lh"tt aknfi n(ttr*
The secondline of evidenceis derivedfrom rhe neurophysiology
of learning. It was one of Hebb's points that cell assemblies
representing things in the brain are held togerherby excitaroryconnections betweenthe neurons of which they are composedand that
these connectionsare establishedthrough a learning process.The
most natural way in which such learningcould take place is the
transformation of a statistical correlation,say, a frequent coincidence of a certain set of elementaryfeaturesin the input, into
synapticconnectionsbetweenthe correspondingneurons.We have
already seenhow somerecentobservationson the plasticity of the
connectionsof singleneuronscan indeedbe explainedby invoking
such a mechanism(62).
The anatomyof the cortex, as I havejust sketchedit, alsomakes
good sensein termsof the theory of cell assemblies
(68). If we wanr
to be ready to build up plenty of cell assemblies,
we needplenty of
neurons. The cerebral cortex in fact contains about as many

r43 | ThresholdControl
neuronsas the rest of the brain. Theseought to be richly connected,
with a high divergenceof signalsfrom eachneuron to as many as
possibleother neuronsin the cortex.We havealreadyseenthat the
pyramidal cells do their best in this respect.This divergence(and
correspondingconvergence)is necessaryin order to provide as
much freedomaspossiblefor the choiceof partnersin the development of cell assemblies.
Finally, most synapsesought to be excitatory, sincecell assemblies
areheld togetherby excitatorysynapses.
About three out of four synapsesin the cortex are of type I, presumablyexcitatory (6g1.

Thresholdcontrol and the pump

of thoughts (Vehicles r 2, 11 , 14)
Threshold control is what makes Vehicle rz special.
There is only indirect evidencefor this in animal brains, but it is
difficult to seehow they could work otherwise, especiallymammalian brainswith their enormouscollectionof corticalneuronsreciprocally connected by positive feedback, an explosive situation
indeed.Threshold control may be more than a necessaryevil; in
fact, it doesintroduceinterestingdynamicsin an otherwiseall-toorigid mechanismof "cell assemblies"and associations.I have
shownthat somethingakin to thinking may resultfrom the dynamics of thresholdcontrol (Zo),and Palm (7r) has gonea long way
toward showingthe potentialitiesof this principle.That this may be
chaotic (zz), and therefore unpredictable,you may or may not )95t f
acceptas sufficientexplanationof the freedomof thought.
on the input sideof the thresholdcontrol we needa mechanism
that can quickly discoverthe explosiveignition of cell assemblies.
A cell assemblymay be composedof neuronsdistributedover wide
regionsof the cerebralcortex; therefore, this mechanismshould
receiveinput from the entire coftex. The piece of gray substance



accompanyingthe cortex of the hemispherethroughoutits extentis
the caudatenucleus,and it has been shown to receiveinput from
everywherein an orderly topographicalproiection (73).
A recent study by Wilson, Hull, and Buchwald (74) provides
evidencefor very effectivepropagationof input from differentparts
of the cerebral cortex throughout the caudatenucleus,quite in
accordancewith the idea of the caudatenucleusas the detectorof
overall cortical activity. I suggestthat the striatum-caudatecomplex is part of the mechanismfor cortical thresholdcontrol probaloop.
bly via the paleostriatum-thalamus
The last two vehicles,introducing the idea of prediction,repeat
I am unableto say
by psychologists.
what hasoften beendiscussed
whetherthe idea of optimistic prediction is original with me (ZS);I
presumeit has occurredto others.It seemsto me sufficientto take
away any aura of mysteryfrom goal-directedbehavior.



Caial, S. R. y. 1898. Structureof the optic chiasmand generaltheory of the

crossingvisual pathways. (Estructuradel kiasmaoptico y teoria generalde los
entrecruzamientosde las vias nervosas.)Reu.trimest. micrograf.Lll, quoted
in Caial,r9rr.

L. Gaze,R. M. 1958. The representationof the retina on the optic lobe of the
ftog. Quart. ]. Exp. Physiol. 4rt zo9-Lr4.
t. Braitenberg,Y.1967. Patternsof proiectionin the visual systemof the fly. I.
Retina-Laminaproiections.E*p. Brain Res.3: z7r-298.
4. Autrum, H. J., and I. Wiedmann.1962.Experimentson the opticsof the insect
eye (appositionaleye). (Versucheriber den Strahlengangim Insektenauge(Appositonsauge).)
Z. Naturforsch. ry: 48o-482.
Kirschfeld, K. 1967. The proiection of the visual environment on the array of
rhabdomeresin the compound eye of Musca. (Die Projektion der optischen
Umwelt auf das Raster der Rhabdomereim Komplexaugevon Musca.) Erp.
Bruin Res.3: 248-27o.

Vigier, P. r9o7, The mechanismof the synthesisof the visual perception in

compound eye of dipterans. (Mecanismede la Synthese des impressions
lulumineusesrecueilliespar les yeux composesdes dipteres.) C.R. Acad. Sci.

Paris tzz-r24.
6. Horridge,G. A., and I. A. Meinertzhagen.ry7o.The accuracyof the patterns
of connexionsof the first- and second-orderneurons of the visual systemof
Calliphora.Proc. Roy. Soc.B. ry5: 69-82.

1 46 | RE F E RE N C ES
Progressin Brain Res.
7 . BraitenbergrY. ry65. Taxis, kinesisand decussation.
r7i 2ro-222,.
Braitenberg, V. 1968. On Chiasms. lnz Neural networks, edited by E. t(.
Caianiello. Berlin, Heidelberg, New York: Springer-Verl"g,pp. 34-42.
Fraenkel, G. S., and D. L. Gunn. ry6r. The orientation of anitnals. Kineses,
taxes and cornpdssreactions. New York: Dover PublicationsInc.
[.ocb, J. rt9o. Hcliouopism in animals and its analogy with plant heliotropism. (Der Heliotropismus der Tiere und seine Ubereinstimmungmit dem
Heliotropismus der Planzen.) Wiirzburg. p. rr8. Quoted in Fraenkel and

Gunn, 196r.
Roessler,O. E. r98r. An artificial cognitive-plus-motivationalsystem.Progr.
Theor. Biol. 6z t47-t6o.
Koshland, D. E. r98o. Bacterial chemotaxisas model behauioralsystez. New
York: Raven Press.


Fermi, G., and V. Reichardt, ry67. Optomotor reactionsin the fly, Musca
domestica.Dependenceof the reaction on the spacefrequency,th velocity and
the luminance of moving periodic patterns. (Optomotorische Reaktion der
Fliege Musca domestica.Abhengigkeit von der Reaktion der Wellenlinge, der
Geschwindigkeit,dem Kontrast und der mideren l-euchtdichtebewegterpe-



riodischerMuster.) Kybernetik z: t5-28.

Reichardt, W. r97o. The insecteye as a modelfor analysisof uptake,transduction and processingof optical data in the nervoussystem.34. Physikertagung
1969, Salzburg,Plenarvortriige.Stuttgart: B. G. Teubner.
Fraenkel, G. S., and D. L. Gunn. 196r. The oilenution of animals. Kineses,
taxes and contpdssreactions. New York: Dover Publications Inc.
Mast, S. O. 1923. Photic orientation in insec$, with spccialreferenceto the
dronefly, Eristalis tenax, and the robber-fly, Erax rufibarbis. /. Exp. Zol. 382

Ewet, J. P. r98o. Neuroetbology. An innoduction to the nanrophysiological
fundamenuls of behauiozr. Berlin, Heidelberg,New York: Springer-Verlag.
r 5 . McCulloch, W.S., and V. H. Pins. r94j. Alogical calculusof ideasimmanent
in nerryousactivity. Bull. Math. Biophys.Szrr1-r1r.
1 6 . Creed, R. S., D. Denny-Brgwn, J. Eccles,E.G. T. Liddell, and C. S. Sherrington. tgtz. Rfiex actiuity of the spinal cord. Oxford: Clarendon Press.
1 7 . Eccles,l. C. 1964. The physiology of syndpses.Berlin, G<ittingen,Heidelberg,

New York: Springer-Verlag.

r 8 . Kleene,S. C. 1956. Representationof eventsin nervenetsand finite automata.


| References

ln: Automata studies,edited by c. E. Shannonand J. Mccarthy. princeron,

New Jersey:PrincetonUniversity Press.
Arbib, M. A. ry64. Brains,machinesand mathetnatics.New york: McGraw
rg. Von Neumann,J. 1956.Probabilisticlogicsand the synthesisof reliableorganisms from unreliablecomponents.lnzAutomata studies,edited by C. E. Shannon and J. Mccarthy. Princeton, New Jersey:Princetonuniversity press.
zo. Turing, A. M. on computable numbers, with an application to the Entscheidungsproblem.Proc.London Math. Society.Ser.z, 4z: z3o-z6S Ggl6)
and 4z: 514-546 (tglZ).
Minsky, M. L. t967. Computdtiorr: finite and infinite qnachines.Englewood
Cliffs, New Jersey:Prentice-Hall.
Sampson,l. R. ry76. Adaptiue inforrution processing.
An introductory surzey. New York, Heidelberg,Berlin: Springer-Verlag.
2r- Dawkins, R. ry76. The selfishgene. oxford: oxford university press.
zz. Darwin, ch. 1859. The origin of species.New American Library. Mentor
Fisher, R. A. ry58. The genetical theory of naural selection. New york:
Mayr, E. ry7o. Population, species,and euolutioz. Cambridge, Mass.: Harvard University Press.
Li. uttley, A. M. 1956. conditional probability machines and conditioned
reflexes.ln: Automata studies,edited by c. E. shannon and J. Mccarthy.
Princeton,New Jersey:Princeton University Press.
24. steinbuch, K. ry69. The learning matrix. (Die krnmatrix.) Kybenetik. n
LS. Hebb, D. O. ry49. Organization of behauior.New York: Wiley and Son.
26. Sutton, R. s., and A. G. Barto. r98r. Toward a modern th*ry of adaptive
nctworls: expectationand prediction. Psychol. Ran 88: r15.
L7. Hubel, D. H., and T. N. Viesel. 1965.Binocularinteractionin striatecortexof
kinens rearedwith artificial squint. J. Neurophysiol.zlz ro4r-rojg.
T. N., and D. H. Hubel. 1965. comparison of the effectsof unilateral
and bilateral eye closureon cortical unit responsesin kitten. J. Neuropbysiol.
z9z rozg-ro4o.
28. Baer, A. r9o5. Contemporaneouselectrical stimulation of nvo regions of the
cerebralcortex of an unrestraineddog. (Uber gleichzeitigeelektrischeReizung
zweier Gro8hirnstellenam ungehemmtenHunde.)Pfliiger'sArch. Ges.physiol. to6: 54-567.



Loucks, R. B. 1933. Preliminaryreport of a techniquefor stimulationor de'

struction of tissuesbeneaththe integumentand the establishingof conditioned
reactionswith faradization of the cerebralcortex.J. Comp- 439444.
Bennett,E. L., M. C. Diamond, D. Krech, and M. R. Rosenzweig.t964.
Chemicaland anatomicalplasticityof brain. Sciencet46z 6to-6t9.
Diamond, M. C., D. Krech, and M. R. Rosenzweig.1964.The effectsof an
enriched environment on the histology of the rat cerebral cortex. J. Comp.
Neur. rz3: rrr-r2o.
Valsh, R.N., O. E. Budtz-Olsen,L. E. Penny,and R. A. Cummins.1969'The
effectsof environmentalcomplexity on the histologyof the rat hippocampus./.
Comp.Neur.r37z 36r-766.
of the threemain typesof glial
Szeligo,F., and C. P. Leblond. ry77. Response
cellsof cortex and corpus callosumin rats handledduring sucklingexposedto
enriched, control and impoverished environments following weaning. /.
Comp. Neur. rTzz 247-264.
3 o . Schtz, A. 1978. Some facts and hypothesesconcerning dendritic spinesand
learning. ln: Architectonics of the cerebralcortex, edited by M.A. B. Brazier
and H. Petsche.New York: RavenPress,pp. rzg-ri1.
Schtiz,A. r98r. Prenatal maturation and postnatal changesin the guineapig
cortex: a histological study of a natural deprivation experiment.I. Prenatal




devclopment.(PrdnataleReifung und postnataleVerinderungen im Cortex des

Meerschweinchens:Mikroskopische Auswernrng eines nanirlichen Deprivationsexperimentes.I. Prinatale Ennyicklunil ]. Hinforsch. zzz 91-ur.
Schffz,A. r98r. Prenatal maturation and postnatal changesin the guineapig
coftex: a histological study of a natural deprivation experiment.II. Postnatal
changes.(Priinatale Reifung und postnatale Verinderungen im Cortex des
Meerschweinchens:Auswerfirng einesnat0rlichen Deprivationsexperimentes.
II. PostnataleVerinderungen.) J. Hirnforsch. zLz rrr-rz7Seefor instancethe impressivecollectionof visual areasin the papersby Tusa,
Palmer and Rosenquist, by Van Essen,Maunsell and Bixby, by Allmann,
Baker, Newsome and Petersenand by Gros, Brucc, Desimone,Fleming, and
Ganas in the volume Cortical sensoryorganization, Vol. z: multiple uisual
areas, edited by Clinton N. Voolsey. Clifton, New Jersey: Humana Press,
Minsky, M. ry25. sunantic infornation processing.cambridge, Mass.: The
MIT Press.


| References

of spaceunderlyLieblich,I., and M. A. Arbib. 1982.Multiple representations

ing behavior.The Behauioraland Brain Sciences.
5: 627-659.
15. Barlow, H. 8., R. M. Hill, and W. R. Levick. 1964. Retinal ganglion cells
respondingselectivelyto directionsand speedof imagemotion in the rabbit. /.


Physiol.ryT 377-407.
B., and W. Reichardt.r955. Systemtheoreticanalysisof temporal
16, Hassenstein,
factors, of sequenceand sign in the perceptionof movement by the beetle,
Analyse der Zeit, Reihenfolgenund VorChlorophanus. (Systemtheoretische
des Riisselkiifers Chlorozeichenauswertungbei der Bewegungsperzeption
phanus.)Z. Naturforcch. rtb: Srt-1z'4.
Hassenstein,B. 1958. Perceptionof movementof figural and irregular patterns. (Uber die lgahrnehmung der Bewegungvon Figuren und unregelmi8igen Helligkeitsmustern.)Zeitschrift f. uergl. Physiol. 4o: 556-592.
Reichardt,W., and D. Variu. 1959.Transferfunctionsin the visualPerception
im Auswertesystemftir das Beweof movement.(Ubertragungseigenschaften
Z.f. Naturforschung.4b $o): 674-689.
37. Fermi, G., and W. Reichardt, ry63. Optomotor reactionsin the fly Musca
domestica.Dependenceon wave length, velocity, contrast and averageluminance of moving periodic patterns. (Optomotorische Reaktion der Fliege
Musca domestica.Abhengigkeit der Reaktion von der Wellenliinge, der Geschwindigkeit, dem Kontrast und der mittleren leuchtdichte bewegterperiodischerMuster.) Kybemetik zz 15-28.
Reichardt,V., and T. Poggio.1975.Visualcontrol of orientationbehaviourin
the fly. Part I. A quantitativeanalysis.Quart. Reu.Biophys. 3; 1rr-375.
Poggio,T., and W. Reichardt.ry76. Visualcontrolof orientationbehaviourin
the fly. Part II. Towards the underlying neural interaction. Quart. Reu. Biophysics9, jt 177-4j8.
38. BraitcnbcrgrY. r9Tj. On the texture of btains.New York, Heidelberg,Berlin:
j9. Mach, quotedin Radiff, F. 1965.Mach bands.San Francisco:Holden Day.
40. Bekesy,G. v. r95o. Experiments in hearing.New York: McGraw Hill.
4r. Hartline, H. K., and F. Ratliff. r957. Inhibitoryinteractionof receptorunits in
the eyeof Limulus.J. Gen. Physiol.4o: 757-367.
42. Reichardt,V. 196r. On the optical resolutionin Limulus. (Uber das optische
von "Limulus.") Kybemetik r 59-69.
Variu, D. 1965. On the theory of lateralinhibition. lnz Cyberneticsof neural
editedby E. R. Caianiello.Rome:C.N.R.

Szentagothai,J. ry67. ln: Eccles, Ito and Szentagothai:The cerebellumas a
neuronal machine.Berlin, Heidelberg, New York: Springer.
44. Beurle,R. L. r956. Propertiesof a massof cellscapableof regenerating
Proc. R. Soc.Lond. Ser.B. 24o: jj.




Wilson, H. R., and J. D. Cowan. r97j. A mathematicaltheory of the functional dynamicsof cortical and thalamic nervoustissue.Kybernetik 13 3j,
Hubel, D.H., and T. N. lViesel.rgTz.Laminarand columnardistributionof
geniculo-corticalfibers in the macaquemonkey.J. Comp. Neurol. l.46: 4zr-

Probleme.Berlin: Springer.
46. Krihler, W. rg33. Psychologische
ITertheime4M. t923. Experimental contributionsto the theory of form. (Untersuchungenzur Lehre von der Gestalt.) Psychol.Forsch.4.
Petermann,B. 1929. Die Vertheimer-Koffka-K<ihlerscheGestalttheorieund
das Gestalproblem. l*ipzig: Barth.
For recent developmentssee:
Gibson,J.J. rgSo. Thefirceptionof the visualworld.Boston:Houghton Mifflin

Julesz, B. ry7r. Foundation of cyclopeanpelceptioa. Chicago and London:

The University of Chicago Press.
Metzger, W. xgZS.Gesetzedes Sebens.Frankfurt: Verlag Waldemar Kramer;
and a seriesof contributions of the ltalian school: Kanisza,G., F. Metelli, G.



Vicario, and P. Bozzi, in the Riuista di Psicologiaand the Giomale ltaliano di

47. Minsky, M., and S. Papert. 1969.Perceptrons.Cambridge,Mass.: The MIT
Braddick, O. J., and A. C. Sleigh. 1983. Physicaland biological processingof
images.Berlin, Heidelberg, New York: Springer.
48. Braitenberg,Y. 1983.ln searchof morphemesin the brain. Giornaleltaliano
di jzr-j4o.
49. At the EuropeanNeuroscienceSocietymeering,Brighton, r98o.

Kanisza,G. ryz+. Contours without gradientsor cognitivecontours. Giornale

Iuliano di Psicologia t.
Brunswik, E. rg3 j. Quoted in Merzger 1975.Seenote 46.
Kennedy,J. M. rgZ+. Quoted in Metzger 1975.Seenote 46.


Lieblich, I., and M. Arbib, 1982. Seenote 34.

Sz. Barlow, H. 8., and B. C. Reeves.ry78.Ttte versatilityand absoluteefficiency
of detectingmirror symmetry in random dot displays.Vision Res.Vol. 19:


I References

53. Scheffers,G. r9rr. Lehrbuch der Mathenatik, z,ndedition. Leipzig: Veit and
54. Hubel,D. H., and T. N. Wiesel.1959.Receptivefieldsof singleneuronesin the
cat's striatecortex.J. Physiol.(Lond.) r48: 574-59r.
Hubel, D. H., and T. N. Viesel. ry77. Fundonal architecnrreof macaque
monkeyvisual cortex. FerrierLecture.Proc. R. Soc.Lond. B. r98: r-59.

Tunturi, A. R. 1962. Frequencyarrangementin anterior ectosylvianauditory

cortex of dog. A*. J. Physiol.zo1: r85.
Evans,E. F. 1968. Upper and lower levelsof the auditory system:A contrastof
strucnrreand function.ln: Neural nehaorks,edited by E. R. Caianiello.Berlin,

Heidelberg,New York: Springer.

56. Braitenberg,Y. t978. Cortical architectonics:general and areal. ln: Architectonicsof the cetebtal cortex, edited by M. A. B. Brazier and H. Petsche.New
York: RavenPress,pp. 44t-465.
Braitenberg,Y. 1978. Cell assembliesin the cerebral cortex. lil Lecture notes
in biomathematics,Yol. zr, edited by R. Heim and G. Palm. Berlin, Heidelberg,New York: Springer,pp. r7r-r88.
57. Jasper,H. H. 1969. Mechanismsof propagation: extracellular studies.In:
Brain mechanismsof the epilepsies,edited by H. H. Jasper,A. A. Vard, and
A. Pope.Boston: Little, Brown and Company.
58. Morell, F. 196r. Lasting changesin synapticorganization produced by continuous neuronal bombardment.In: CIOMS sqposiurn on brain mechanisms
and learning,edited by A. Blackwell, pp. 375-392.
59. Renaud, quoted by Gloor, P. 1972.ln: Synchronization of EEG actiuity in
epilepsies,edited by H. Petscheand M. A. B. Brazier. Wien, New York:
Seenote 56.
6o. Braitenberg,Y. 1978. Cell assemblies.
6t. Hubel, D.H., and T. N. Wiesel.1955. Binocularinteraction.Seenote 27.
T. N., and D. H. Hubel. 1965. Comparisonof the effects.Seenote 27.
62. Rauschecker,
J. P., and V. Singer.r98r. The effectsof early visual experience
on the cat's visual cortex and their possibleexplanation by Hebb synapses.
Physiol.jro: zrS-2r9.
$. Braitenberg,V. 1983. Explanationof orientationcolumnsin termsof a homogeneousnetwork of neurons in the visual cortex. NeuroscienceAbstracts.9:
64. Hebb, D. O. 1949. The organization of behauior.New York: John Wiley.
65. Hubel,D. H., and T. N. Wiesel.1959.Receptivefieldsof singleneuronesin the
cat's striatecortex.J. Physiol.(Lond.) r48: 574-59r.

Evans,E. F. 1958.Uppe.rand lower levelsof the auditorysystem:A contrastof
structure and function.ln: Neural netutorhs,editedby E. R. Caianiello.Berlin,
Heidelberg, New York: Springer.
Aeftsen, A. M. H.J., and P. I. M. Johannesma.r98r. The spectrotemporal
receptive field. A functional characteristicof auditory neurons.Biol. Cybem.
42| r33-r4t.
62. Hubel, D. H., and T. N. Wiesel. ry65. Binocularinteraction.Seenote 27.
Wiesel,T. N., and D. H. Hubel. 1965.Comparisonof the effects.Seenote 27.
Blakemore,C., and G. F. Cooper. r97r. Modificationof the visual cortex by
experience.Brain Res. 3t: 766.
Hirsch, H. V. 8., and D. N. Spinelli. r97r. Modificationof the distributionof
receptive 6eld orientation in cats by selectivevisual exposureduring development. Exp. Brain Res. 13: r-43.
68. Braitenberg,Y. 1978.Cell assemblies.
Seenote 56.
69. Wolff, J. R. ry76. Quantitative analysisof topographyand developmentof
synapsesin the visual coftex. Exp. Brain Res.Suppl.r 259-263.
Uchizono, K. ry66. Characteristicsof excitatory and inhibitory synapsesin the
central nenroussystemof the cat. Nature zo7: 642.
Braitenberg,Y. 1978. Cell assemblies.Seenote 55.
Braitenberg,Y. 1977,On the texture of brains.Neuroandtomyfor the cybernetically tninded. Berlin, Heidelberg, New York: Springer.
7 r . Palm, G. 1982. Neural assemblies.An alternatiueapproachto artificial intelligence. Berlin, Heidelberg,New York: Springer-Verlag.

\1q "'
hvclR r

Myrberg, P.J. 1958.Iteration of real polynomialsof the znd degree.(Iteration

der reellenPolynomezten Grades.)l. Am. Acad. Sc.Fenn.z5rA: r-ro.
Webster, K. E. 1965. The cortico-striatal proiection in the cat. J. Anat. 99:
Dray, A. r98o. The physiologyand pharmacologyof mammalianbasalganglia. Progr. Neurobiol. r4t z2r-jjS.

7 4 . Wilson, J. S., C. D. Hull, and N. A. Buchwald. ry83.Intracellularstudiesof

the convergenceof sensoryinput on caudateneuronsof cat. Brain Res.z7o:
V. Gehirngespinste.Neuroanatomie fiir kybernetisch Interes75.
siefte. Berlin, Heidelberg,New York: Springer,1973. RevisedEnglishtranslation: Oz the texture of bains. Neuroanatomy for the cyberneticalb minded.
Berlin, Heidelberg,New York: Springer, ry77.