Brain Research, 386 (1986) 93-104 93 Elsevier BRE 12101 Relative Motion Sensitivity in Cat Striate Cortex as a Function of Stimulus Direction P. HAMMOND, B. AHMED* and A.T. SMITH** Department of Communication and Neuroscience, University of Keele, Keele ( U. K.) (Accepted 25 March 1986) Key words: Vision - - Cat - - Striate cortex - - Relative motion sensitivity This paper presents the second stage of an investigation into the relative motion sensitivity of complex cells in the striate cortex of lightly anaesthetized adult cats. Relative motion between an oriented bar and a background of random visual texture was generated by moving both stimuli, in-phase or in antiphase, with dissimilar speed. Three configurations were compared: motion of both foreground bar and background texture in preferred directions for either (a) the bar or (b) the background, and also (c) where background motion was other than orthogonal to bar orientation. Providing foreground stimuli elicited substantial responses, sensitivity to relative motion was qualitatively but not quantitatively predictable from discrete responses to foreground or background alone, whatever the angular inclination between their respective directions, suggestive of strongly non-linear interactions between the two. Where the foreground evoked no response, or depressed firing, the pattern of sensitivity to relative motion could not be predicted. INTRODUCTION With some 30 years of hindsight, it is generally ap- preciated that the visual system does not operate in the simplistic manner suggested by early single neu- rone studies of sensitivity to isolated oriented con- tours 2°-22. Rather, it routinely processes complex and dynamic visual scenes, amongst which interac- tions between foreground objects and their back- grounds are commonplace. We have therefore at- tempted to assess the contribution of striate cortical neurones to this processing, by studying their sensi- tivity to foreground motion against textured, rather than uniform, backgrounds which may move simulta- neously with, or relative to, the foreground. Studies of neural sensitivity to relative motion are limited in number. In the cat's superior colliculus, Mandl studied relative motion sensitivity only be- tween a moving disc and a stationary patterned back- ground 23. Rizzolatti et al. investigated extrarecep- tive field influences on responses to stimuli within the receptive fiek., in the superior colliculus 27, lateral su- prasylvian visual area 25,26 and striate cortex 26. The closest parallel with our own studies is the work of Frost's group who reported sensitivity to relative mo- tion between a foreground spot paired with back- ground texture in the pigeon tectum 4"5 and in the cat lateral suprasylvian cortex 6. Their main finding was that neural responses were suppressed by in-phase relative motion and enhanced by relative motion in antiphase. However, except for our own previous re- ports 17A9, no comparable measurements are avail- able for the striate cortex. Because complex (but not simple) cells in the cat's striate cortex are responsive to motion of random vis- ual texture 12, we chose to investigate the sensitivity of such cells to monocularly viewed relative motion between a foreground bar and background texture. We used a paradigm in which the bar moved in opti- mal directions at a fixed velocity whilst the back- ground texture moved relative to it, in the same (in- phase) or opposite (antiphase) directions ~7'19. The responses of virtually all complex cells in the striate * Present address: Department of Physiology, Faculty of Medicine, University of Kuwait, P.O. Box 24923, Kuwait. ** Present address: Department of Psychology, Univ~ersityCollege Cardiff, P.O. Box 78, Cardiff CF1 1XL, U,K. Correspondence: P. Hammond, Department of Communication and Neuroscience, University of Keele, Keele, Staffordshire ST5 5BG, U.K. 0006-8993/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)