Visual segmentation by contextual influences via intra-cortical interactions in the primary visual cortex

Stimuli outside classical receptive fields have been shown to exert a signi ficant influence over the activities of neurons in the primary visual corte x. We propose that contextual influences are used for pre-attentive visual segmentation. The difference between contextual influences near and far fro m region boundaries makes neural activities near region boundaries higher t han elsewhere, making boundaries more salient for perceptual pop-out. The c ortex thus computes global region boundaries by detecting the breakdown of homogeneity or translation invariance in the input, using local intra-corti cal interactions mediated by the horizontal connections. This proposal is i mplemented in a biologically based model of V1, and demonstrated using exam ples of texture segmentation and figure-ground segregation. The model is al so the first that performs texture or region segmentation in exactly the sa me neural circuit that solves the dual problem of the enhancement of contou rs, as is suggested by experimental observations. The computational framewo rk in this model is simpler than previous approaches, making it implementab le by V1 mechanisms, though higher-level visual mechanisms are needed to re fine its output. However, it easily handles a class of segmentation problem s that are known to be tricky. its behaviour is compared with psyche-physic al and physiological data on segmentation, contour enhancement, contextual influences and other phenomena such as asymmetry in visual search.