Brain Mechanisms and Spatial Vision

468 Pages·1985·29.779 MB·English

by  David N. Lee, David S. Young (auth.), David J. Ingle, Marc Jeannerod, David N. Lee (eds.)

Brain Mechanisms and Spatial Vision
NATO ASI Series
Advanced Science Institutes Series
A Series presenting the results of activities sponsored by the NATO Science
Committee, which aims at the dissemination of advanced scientific and technological
knowledge, with a view to strengthening links between scientific communities.
The Series is published by an international board of publishers in conjunction with the
NATO Scientific Affairs Division
A Life Sciences Plenum Publishing Corporation
B Physics London and New York
C Mathematical and D. Reidel Publishing Company
Physical Sciences Dordrecht and Boston
D Behavioural and Martinus Nijhoff Publishers
Social Sciences DordrechtlBoston/Lancaster
E Applied Sciences
F Computer and Springer-Verlag
Systems Sciences BerlinlHeidelberglNew York
G Ecological Sciences
Series D: Behavioural and Social Sciences - No. 21
Brain Mechanisms and
Spatial Vision
edited by
David J. Ingle
The Rowland Institute for Science
Cambridge, MA 02142
USA
Marc Jeannerod
Laboratoire de Neuropsychologie Experimentale
INSERM - Unite 94
Bron, France
David N. Lee
Department of Psychology
University of Edinburgh
Edinburgh, UK
1985 Martinus Nijhoff Publishers
Dordrecht / Boston / Lancaster
Published in cooperation with NATO Scientific Affairs Division
Proceedings of the NATO Advanced Study Institute on Brain Mechanisms and Spatial
Vision, Lyon, France, June 16-25, 1983
Library of Congress Catalog Card Number: 84-27297
ISBN-13: 978-94-010-8743-8 e-I SBN-13: 978-94-009-5071-9
001: 10.1007/978-94-009-5071-9
Distributors for the United States and Canada: Kluwer Boston, Inc., 190 Old Derby
Street, Hingham, MA 02043, USA
Distributors for the UK and Ireland: Kluwer Academic Publishers, MTP Press Ltd,
Falcon House, Queen Square, Lancaster LA1 1RN, UK
Distributors for all other countries: Kluwer Academic Publishers Group, Distribution
Center, P.O. Box 322, 3300 AH Dordrecht, The Netherlands
All rights reserved. No part of this publication may be reproduced, stored in a
retrieval system, or transmitted, in any form or by any means, mechanical,
photocopying, recording, or otherwise, without the prior written permission of the
publishers,
Martinus Nijhoff Publishers, P.O. Box 163, 3300 AD Dordrecht, The Netherlands
Copyright © 1985 by Martinus Nijhoff Publishers, Dordrecht
Softcover reprint of the hardcover 1s t edition 1985
v
Preface
This volume contains chapters derived from a N.A.T.O. Advanced
Study Institute held in June 1983. As the director of this A.S.I.
it was my hope that some of the e1ectrophysiologists could express
the potentialities of their work for perceptual theory, and that
some perceptionists could speculate on the underlying "units" of
perception in a way that would engage the imagination of physio
logists. The reader will have to be the judge of whether this was
achieved, or whether such a psychophysiological inter1ingua is
still overly idealistic. It is clear that after the revolution
prec~pitated by Hube1 and Weisel in understanding of visual cortical
neurons we still have only a foggy idea of the behavioral output
of any particular species of cortical detector. It was therefore
particularly unfortunate that two persons who have made great
strides in correlating interesting facets of cat cortical physio
logy with human psychophysics (Max Cynader and Martin Regan of
Dalhousie University) were unable to attend this meeting. Never
theless, a number of new and challenging ideas regarding both
spatial perception and cortical mechanisms are represented in this
volume, and it is hoped that the reader will remember not only
the individual demonstrations but the critical questions posed by
the apposition of the two different collections of experimental
facts.
David Ingle
April 1984
VII
TABLE OF CONTENTS
PREFACE V
D.N. Lee and D.S. Young
Visual Timing of Interceptive Action 1
J.J. Koenderink
Space, Form and Optical Deformations 31
K. Nakayama
Extraction of Higher Order Derivatives of the Optical Velocity
Vector Field: Limitations Imposed by Biological Hardware 59
J. Todd
The Analysis of Three-Dimensional Structure from Moving
Images 73
B. Rogers and M. Graham
Motion Parallax and the Perception of Three-Dimensional
Surfaces 95
A. Yonas and C.E. Granrud
The Development of Sensitivity to Kinetic, Binocular and
Pictoral Depth Information in Human Infants 113
B. Timney
Visual Experience and the Development of Depth Perception 147
J.R. Lackner
Human Sensory-Motor Adaption to the Terrestrial Force
Environment 175
R. Schmid, A. Buizza and D. Zambarbieri
Visual Stabilization During Head Movement 211
D.J: Ingle and B.L. Shook
Action-Oriented Approaches to Visuo-Spatial Brain Functions 229
VIII
A. Cowey
Disturbances of Stereopsis by Brain Damage 259
M. Jeannerod
The Posterior Parietal Area as a Spatial Generator 279
J. Paillard and B. Amblard
Static Versus Kinetic Visual Cues for the Processing of
Spatial Relationships 299
H. Kennedy and G. Orban
Behavioural and Neurophysiological Correlates of Visual
Movement Deprivation in the Cat 331
G.A. Orban
Velocity Tuned Cortical Cells and Human Velocity
Discrimination 371
P. Hammond
Visual Cortical Processing: Texture Sensitivity and Relative
Motion 389
B.J. Frost
Neural Mechanisms for Detecting Object Motion and Figure
Ground Boundaries~ Contrasted with Self-Motion Detecting
Systems 415
G. Jansson
Perceptual Theory and Sensory Substitution 451
INDEX 467
VISUAL TIMING OF INTERCEPTIVE ACTIGN
David N. Lee and David S. Y0ung
Department. of Psycho10gy
University of Edinburgh
Edinburgh EH8 9JZ
Scotland
1 . INTRODUCTION
One of the most remarkable aspects of motor skill is the
precision with which actions can be timed. In this chapter we
will be concerned with the control of actions whose timing is
dictated by how the organism is moving relative to the
environment, as in locomotion, or by how an object is moving
relative to the organism, as in catching or hitting something. We
will refer to this as extrinsic timing. It requires predictive
information about the relative motion of the organism and objects
and surfaces in the environment, information that is often only
available through VlSlon. Time to contact (meaning, in general,
time to nearest approach) is a particularly important predictor.
Consider, for example, catching a moving object with one hand
an ability which starts developing as early as 18 weeks of age
in human infants (1). In catching a ball not only has the hand to
be positioned, properly oriented, on the trajectory of the ball
but also the fingers have to start closing just before the ball
arrives. If hand closure starts too late, the ball will bounce
off the palm too early and it will rap the knuckles. With a
moderate speed ball, timing has to be accurate to about 25ms (2),
whereas in fast ball games like cricket the required precision is
probably around 10ms. The timing of the catch is clearly under
visual control, but this is not to say that the ball needs to be
watched all the way to the hand: providing it is visible about
200-300ms before contact, reasonably precise timing can be
achieved (3, 4). Nonetheless, sight of the ball over the last
2
lOOns to contact can significantly improve performance (5). It
would appear, therefore, that the timing of a catch is normally
under continuous visual control, with the visual information
~etting progressively more precise as contact is approached.
A particularly dramatic example of extrinsic timing in
locomotion is ski jumping. After accelerating to around 60mph
down the steep in-run in crouched posture skiers have to
forcefully straighten up just before they reach the lip of the
in-run in order to launch themselves. Very precise timing is
required. In a film analysis of 14 Olympic ski-jumpers it was
found that on average they started straightening when their time
to contact with the lip was 194ms and the standard deviation of
this time interval across the 14 jumpers was a mere 10ms (6).
The ability to time interceptive acts rapidly and precisely,
as shown in these examples and in many other sports as well as
animal activities (e.g. Ingle, this volume) suggests that
visuo-motor systems have evolved which are particularly efficient
at detecting time to contact and gearing actions to the
infor~ation. In seeking to understand how the systems might work,
it is logical to start with an examination of the optical input.
This we do in the following section, showing how time to contact
under constant approach velocity is specified optically. The
perception of time to contact does not, in principle, require
perceiving distance and speed. In Section 3, we extend the theory
to how actions are timed when approach velocity is not constant,
as when something is falling. Section 4 describes an experiment
testing the theory. Applications of the ideas to the problem of
motor control in hemiparetics and to the training of young
children in crossing the road are presented in Sections 5 and 6.
In Section 7 the theory is extended to visual control of locomotor
acts and Sections 8 and 9 report related experiments.
2. TIME TO CONTACT AND THE TAU-MARGIN
As we have said, organisms need predictive information in
order to time their actions. Consider a person about to catch a
ball. His actions might be driven by a complex predictive system
which takes into account any available information about the
ball's current position, velocity, acceleration, rate of change of
acceleration, and so on. However, an over-complex system could
entail unnecessary delays in making use of the information as well
as over-sensitivity to nOise, and as such would be unlikely to
have evolved. A better system is robust rather than over-refined
and takes advantage of the most reliable and rapidly available
information.
3
Ball
Point of
observation t
D
-
!
-r-
Figure 1. How the tau-margin is optically specified for a ball,
diaDeter D, on a collision course. At time t it is distance r
from the point of observation. By definition, the tau-Dargin =
-r/}. (The dot denotes rate of change with tiDe.) The solid
angle A = rr D2/4r2; differentiating, A = -1rD2r/2rl. Dividing
these equations sives 2A/A = -r/0 = the tau-Dargin.
The physical quantity that we propose as an estimate of the
time to contact of an object heading for an organism is its
cHst30ce away divided by its speed of approach. This would be an
exact estimate of the time to contact if the relative acceleration
of the object and organism were zero right up to contact. We call
it the tau-margin, after the optic variable T which specifies it.
We emphasize that the tau-margin does not have to be obtained from
a distance and a velocity through a division; that is merely a
convenient way to define it. Indeed, the physical relationship of
the object and organism at a moment in time can be described with
the tau-margin as a primitive quantity and distance and velocity
as derived quantities. We suggest that an organism trying to
intercept a moving object exploits such a description.
Since an object to be intercepted is not always on a
collision course with an organism's point of observation (e.g.,
when playing tennis) we define the tau-margin in general to be the
time to the nearest approach of the object to the pOint of
observation if the relative velocity of the two were to remain
constant. To specify the relative velocity requires a frame of
reference, a set of directions that are fixed relative to each
other. Furthermore, in order that the tau-margin is defined the
frame of reference must not rotate relative to the environment.
If, for example, it were locked to the organism's eye then the
relative velocity and hence the tau-margin - would change in
value whenever the eye rotated. The reference frame may however
move without rotation with the organism through the environment.
In what way is the tau-margin, which partially describes the
spatio-temporal relationship of the ball to the catcher at any
moment in its flight, useful for controlling action? If the