Is Global Motion Really Based on Spatial Integration ... - Science Direct

Pergamon

004%6989(94)EOO17-F

k’isiun Rr.\. Vol. 34, No. IX, pp. 242.5 2430, 1994 Copyright #(‘I 1994 Elscwer Science Ltd Printed in Great Britain. All rights reserved 00426989194 $7.00 + 0.00

Is Global Motion Really Based on Spatial Integration of Local Motion Signals? ANDREW Receiwd

T:SMITH,*t I5 Nmlemhtv

1993;

ROBERT

J. SNOWDEN,*

in retGedform

I2 Januury

ALAN

B. MILNE*

1994

Previous studies have shown that a random-dot kinematogram (RDK) comprising dots, each of which takes a random walk in direction or speed over time, can appear to flow in a single direction. This has been interpreted as evidence for the existence of a co-operative network linking neurons sensitive to different directions/speeds and different spatial locations. We have investigated the possibility that global motion perception in such patterns might simply reflect motion energy detection at a coarse spatial scale (such that many dots fall in the receptive field of one energy detector) without the need to encode local dot motions on a fine spatial scale and then integrate their motions over space. We created random-walk RDKs and then spatially high-pass filtered them to remove low spatial frequencies. Perception of glohat motion was unimpaired for both direction and speed random walks, showing that the phenomenon is not reliant on low spatial frequencies and must, therefore, involve integration of local motion signals across space, as originally isolated. Motion

perception

Global

motion

Kinematograms

INTRODUCTION The process of motion detection in the mammalian visual system appears to show a high degree of spatial localization. Physiological studies show clearly that receptive fields in the early stages of visual processing are small and respond only to motion at restricted locations. Psychophysical studies similarly suggest localized analysis of motion: for example linear spatial summation of contrast sensitivity for moving patterns occurs over a range of less than I deg at all but the very lowest spatial frequencies (Anderson & Burr, 1987). In the realm of computation, models based on the principle of local estimation of motion (e.g. Adelson & Bergen, 1985; van Santen & Sperling, 1984) have been very successful. A large, rigidly-moving object is thus presumed to activate many independent motion sensors operating at different locations in space. However, several factors suggest that this initial local analysis of motion must be followed by some degree of integration of motion signals across space. Firstly, our perception is normally of coherent rigid motion rather than a multitude of point motions, as if a single direction and speed were assigned to an identified rigid object. Secondly, it is known that local motions can be grouped together to yield perception of form (e.g. Wallach & O’Connell, 19.53;

*Vision Research Unit, School of Psychology, University of Wales College of