Welcome to Scribd. Sign in or start your free trial to enjoy unlimited e-books, audiobooks & documents.Find out more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Intermodal Regulation of Heading

Intermodal Regulation of Heading

Ratings: (0)|Views: 0|Likes:
Warren, W.H. (1998). Perception of heading is a brain in the neck. Nature: Neuroscience, 1(8), 647-649.

From abstract in article: How can you see where you are heading, given that your eyes and head can move relative to your body? Extra-retinal information about neck movements is part of the answer.
Warren, W.H. (1998). Perception of heading is a brain in the neck. Nature: Neuroscience, 1(8), 647-649.

From abstract in article: How can you see where you are heading, given that your eyes and head can move relative to your body? Extra-retinal information about neck movements is part of the answer.

More info:

Categories:Types, Research, Science
Published by: Laboratory in the Wild on Jul 09, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

08/04/2013

pdf

text

original

 
 news and views
nature
 neuroscience
volume 1 no 8 december 1998647
How do you see where you’re going? Asyou move about the world, a pattern of motion called optic flow is generated atyour moving eye. Somehow, from thisflux of light, you are able to perceive yourcurrent path of travel, avoid obstacles, andreach your destination. The question of how this is done is of interest to psychol-ogists studying perceptual-motor func-tion, neuroscientists pursuing theunderlying neural mechanisms, comput-er scientists building visually guidedrobots, and indeed anyone who can walk or drive.One answer was proposed 50 years agoby James Gibson
1
. He noted that whenyou travel on a straight path, the opticflow necessarily forms a radial patternwith a focus of expansion lying in yourcurrent direction of self-motion or head-be detected by a mobile eye in a mobilehead. To use optic flow as a source of directional information, the system mustsomehow compensate for these move-ments. Previous work has shown thatextra-retinal information about move-ments of the eye within the head con-tributes to the perception of heading. Inthis issue(pages 732–737),Crowell,Banks, Shenoy and Andersen report thestriking finding that the visual system alsomakes use of extra-retinal informationabout head turns.To visualize the problem, imagine try-ing to build a robot that can guide itself through the world based only on visualinformation from a video camera. Whenthe robot travels on a straight path, theing (
Fig.1a
). Consequently, you could usethis focus as a ‘point of aim’ to steertoward goals and around obstacles. Inpractice, however, things are more com-plicated, because this flow pattern must
Perception of heading is abrain in the neck
William H. Warren
How can you see where you are heading, given that your eyesand head can move relative to your body? Extra-retinalinformation about neck movements is part of the answer.
William H. Warren is at the Department of Cognitive and Linguistic Sciences, Box 1978, Brown University, Providence, Rhode Island 02912, USA.e-mail: Bill Warren@brown.edu
Fig. 1.
Examples of retinalflow fields for travel parallelto a ground plane. Each vec-tor represents the instanta-neous retinal velocity of apoint in the environment.
(a)
Radial velocity field pro-duced by translation onstraight path (s). The focusof expansion lies at the ‘X’.
(b)
Velocity field producedby simultaneous translationon straight path (s) androtation about a verticalaxis to fixate the ‘O’. Thesame instantaneous flowfield is produced by travelon circular path (c), whilelooking along the tangent tothe path.
inhibit neurogenesis in the adult humanbrain, we might eventually find a way toinduce it in specific neurodegenerative dis-orders or following stroke and trauma, byblocking or bypassing the inhibition of mitotic activity. An additional benefit fromdiscovering such inhibitory factors may bein the treatment of malignancy, where thegoal is just the opposite, to prevent celldivision. Finally, a promising approachmay be the transplantation of stem cells tothe affected brain areas. Recent grafting of human fetal neural stem cells into new-born mice demonstrates that such cells canboth survive and, remarkably, migrate anddifferentiate into locally appropriate phe-notypes—oligodendrocytes, astrocytes andneurons
13,14
. In these experiments, how-ever, the host animals were newborns, inwhich the essential developmental cues arestill preserved. For example, radial glial cellscaffolding, an essential substrate for long-distance migration of the cortical neu-rons
15
, is absent in the adult cerebrum. It istherefore unclear whether transplantedneurons would migrate or differentiateappropriately in the adult brain. Further-more, even if they do, this may not be suf-ficient to restore their function. Neuronsin affected structures operate mainlythrough their precise, long-distance con-nections, which would also need to be re-established. To generate such connectionsin the large adult human brain may not bea trivial task. Nevertheless, the wordimpossible is not in the vocabulary of con-temporary neuroscience.
1.Rakic, P.
Science
227
, 154–156 (1985).2.Ericksson, S. E.
et al. Nat. Med.
4
, 1313–1317(1998).3.Altman, J. & Bayer, S. A.
 Restorative Neurol.
6
,203–225 (1993).4.Lois C, & Alvarez-Buylla, A.
Science
264
,1145–1148 (1994).5.Kempermann, G., Kuhn, H. G. & Gage, F. H.
Proc. Natl. Acad. Sci. USA
94
, 10409–10414(1997).6.Kuhn, H. G., Dickinson-Anson, H. & Gage, F. H.
 J.Neurosci.
16
, 2027–2033 (1996).7.Kempermann, G., Kuhn, H. G. & Gage, F. H.
 Nature
386
, 493–495 (1997).8.Eckenhoff, M. E. & Rakic, P.
 J. Neurosci
.
8
,2729–2747 (1988).9.Kirschenbaum, B.
etal.Cereb.Cortex
4
, 576–589(1994).10.Gould, E.
et al. Proc. Natl. Acad. Sci. USA
95
,3168–3171 (1998).11.Nottebohm, F. & Alvarez-Buylla, A.
 Restorative Neurol.
6
, 227–236 (1993).12.Rakic, P.
Science
l83
, 425–427 (l974).13.Flax, J. D.
 Nat.Biotechnol.
16
, 1033–1039 (1998).14.Brustle, O.
 Nat. Biotechnol.
16
, 1040–1044(1998).15.Rakic, P.
Science
241
, 170–176 (1988).
scs
ab
 © 
 
1998 Nature America Inc. • http://neurosci.nature.com
   ©   1   9   9   8   N  a   t  u  r  e   A  m  e  r   i  c  a   I  n  c .  •   h   t   t  p  :   /   /  n  e  u  r  o  s  c   i .  n  a   t  u  r  e .  c  o  m
 
648nature
 neuroscience
• volume 1 no 8 • december 1998
 news and views
motion on the video screen forms a radi-al pattern like that of Fig. 1a—the ‘trans-lational component’ of flow. This patternwill be familiar to anyone who has seen
Star Wars
and watched the MillenniumFalcon catapult into hyperspace. In thelast ten years, research in visual percep-tion has shown that humans judge theirdirection of heading from such radialflow patterns with an accuracy better than1 degree of visual angle
2,3
. At the sametime, neuroscientists located an area inthe visual cortex of the macaque monkey,known as MSTd, that contains neuronsselective for specific kinds of flow pat-terns, including expansion, rotation andspiral motion
4,5
. Ken Britten and col-leagues recently reported that micros-timulation of MSTd biases the macaque’sheading judgments
6
, suggesting that thisarea may indeed mediate the perceptionof heading.Now imagine that the robot is simul-taneously rotating its camera to track anobject in the scene, for example fixatingon a stop sign while moving down thestreet. This dramatically alters the flowpattern on the screen (
Fig.1b
) because therotational component’ destroys any spe-cial feature of the flow field in the direc-tion of translation. How can the robot seewhich way it is heading in this case? Inprinciple, such flow patterns could be ana-lytically decomposed into their transla-tional and rotational components.Numerous computational models havebeen derived for solutions based on reti-nal flow, many of them predicated on thefact that motion parallax is unaffected byrotation
7,8
. If, however, the robot hadaccess to non-visual information abouthow fast its motors were turning the cam-dom-dot display simulating translationand rotation. This time, however, theycompare this condition with an actualhead turn to track a point in a radial flowdisplay. As with eye movements, they findthat an active head rotation yields veryaccurate judgments of the future path.In this case, there are three possiblesources of extra-retinal information—proprioceptive information from theneck muscles, efferent informationabout the motor commands to the neck muscles, and vestibular canal informa-tion about head rotation. By cleverlymanipulating active and passive headand body rotations, the authors are ableto titrate out most of these sources. Forexample, vestibular information was iso-lated by having subjects view a radialflow display while the full body was pas-sively turned in a motorized dentist’schair (with no neck movement; see
Fig.2
). Similarly, neck proprioceptionwas isolated by fixing the head in aheadrest and rotating the chair under-neath it (together with the radial flowdisplay), while keeping the neck musclesrelaxed. To add efferent information tothis situation, the head was freed and thesubject was required to make active headturns to keep a head-mounted laserpointer on a stationary target, while themotorized chair (and display) counter-rotated underneath.Contrary to what might have beenexpected, efferent information was notrequired to compensate for the rotationin half the subjects. Neither neck pro-prioception nor vestibular informationalone was sufficient, and the combina-tion of two sources yielded mixed results,typical of such ‘cue conflict’ situations;yet all three together produced accurate judgments in nearly all observers. Onewould like to understand the nature of these interactions and the origin of indi-vidual differences in more detail. Theauthors provide an important clue byshowing that the combined effect of thisinformation was not simply additive, rul-ing out a traditional linear model of ‘cuecombination’.Taken together, such extra-retinalresearch points toward a consistent pic-ture in which movements of the eye inthe head and the head on the body aresimply compensated by extra-retinalinformation—as if we only gave ourrobot non-visual information about thecamera turns. However, I think such apicture would underestimate the roleof visual information in solving theproblem. This claim hinges on theera, this could be used to estimate the rota-tional component independently. In suchan extra-retinal solution, the rotationcould then be subtracted from the flowpattern to recover the translation.There is evidence, much of it fromBanks’ lab
9,10
, showing that the visual sys-tem does indeed use extra-retinal infor-mation about eye movements to recoverheading. When people are shown com-puter displays that simulate a combinedforward translation and rotation over arandom-dot ground plane and asked to judge their future path of travel, they tendto make large heading errors that increasewith the simulated rotation rate (at >1degree/s). As one might suspect fromlooking at Fig. 1b, observers typicallyreport traveling on a curved rather than astraight path
11
, suggesting that they can-not recover the direction of translationfrom the flow pattern alone. However,when they make an actual pursuit eyemovement to track a point in a radial flowdisplay, heading errors are small, indicat-ing that the visual system makes use of extra-retinal information about eye move-ments. It is known that MSTd neuronsreceive eye movement signals, and recentwork in Andersen’s lab
12
has shown thatin a subset of MSTd expansion-sensitiveneurons, the preferred focus of expansionactually shifts within the receptive fieldduring a pursuit eye movement, in a man-ner that would partially compensate forthe rotation.What is startling about the report of Crowell and colleagues is that extra-reti-nal information about something asremote as the neck can have the same role.The authors use a similar perceptual pro-tocol, in which their subjects view a ran-
Fig. 2.
The experimentalset-up used in the ex-periments by Crowelland colleagues. Subjectsviewed an optic flowdisplay while in a motor-ized dentist’s chair. Thisallowed the authors tomanipulate active andpassive head and bodyrotations, and thus iso-late the contributions ofdifferent sources ofextra-retinal information.Photograph provided byJames Crowell.
 © 
 
1998 Nature America Inc. • http://neurosci.nature.com
   ©   1   9   9   8   N  a   t  u  r  e   A  m  e  r   i  c  a   I  n  c .  •   h   t   t  p  :   /   /  n  e  u  r  o  s  c   i .  n  a   t  u  r  e .  c  o  m

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->