The role of human ventral visual cortex in motion perception

Sharon Gilaie-Dotan,Ayse P Saygin,Lauren J Lorenzi,Geraint Rees,Ryan Egan,Marlene Behrmann

doi:10.1093/brain/awt214

Abstract

Visual motion perception is fundamental to many aspects of visual perception. Visual motion perception has long been associated with the dorsal (parietal) pathway and the involvement of the ventral ‘form’ (temporal) visual pathway has not been considered critical for normal motion perception. Here, we evaluated this view by examining whether circumscribed damage to ventral visual cortex impaired motion perception. The perception of motion in basic, non-form tasks (motion coherence and motion detection) and complex structure-from-motion, for a wide range of motion speeds, all centrally displayed, was assessed in five patients with a circumscribed lesion to either the right or left ventral visual pathway. Patients with a right, but not with a left, ventral visual lesion displayed widespread impairments in central motion perception even for non-form motion, for both slow and for fast speeds, and this held true independent of the integrity of areas MT/V5, V3A or parietal regions. In contrast with the traditional view in which only the dorsal visual stream is critical for motion perception, these novel findings implicate a more distributed circuit in which the integrity of the right ventral visual pathway is also necessary even for the perception of non-form motion.

Highlights

IntroductionThe visual system is traditionally segregated into two main processing streams, the ventral ‘form’ visual pathway, and the dorsal ‘visuospatial’ pathway, sometimes referred to as the ‘motion’ pathway (Ungerleider and Mishkin, 1982; Mishkin et al, 1983; Van Essen and Maunsell, 1983; Van Essen, 1985; Maunsell and Newsome, 1987; Maunsell et al, 1990; Goodale and Milner, 1992)
To assess the contribution of the ventral stream to motion perception, we compared the performance of patients with that of controls on a series of tasks that have proven effective in uncovering impairments in motion perception (Vaina et al, 1990; McLeod et al, 1996; Spencer et al, 2000; Milne et al, 2002; Saygin, 2007; Gilaie-Dotan et al, 2011)
To characterize sensitivity to motion coherence, we calculated the proportion of coherently moving dots required to detect the fast coherent motion (27.27/s) embedded in dots moving in random directions (Green, 1961; Levinson and Sekuler, 1976; Meteyard et al, 2008; Gilaie-Dotan et al, 2013)

Summary

Introduction

The visual system is traditionally segregated into two main processing streams, the ventral ‘form’ visual pathway, and the dorsal ‘visuospatial’ pathway, sometimes referred to as the ‘motion’ pathway (Ungerleider and Mishkin, 1982; Mishkin et al, 1983; Van Essen and Maunsell, 1983; Van Essen, 1985; Maunsell and Newsome, 1987; Maunsell et al, 1990; Goodale and Milner, 1992).

Objectives

Methods

Results

Discussion

Conclusion