Abstract

The distinction between "object vision" and "spatial vision" was made by Ungerleider and Mishkin (1982) on the evidence of behavioural and neuroanatomical studies, largely with monkeys. It holds that separate cortical systems are involved, the occipito-temporal in object vision, the occipito-parietal in spatial vision. This distinction has been unquestioned; it appears as though fact in influential textbooks; but has never been subjected to critical scrutiny. Theoretically, there are substantial difficulties, for instance because during perception objects do not form a special category, apart from their features; and because shape, pattern and size are spatial features of objects. Empirically, for the monkey the behavioural distinction does not hold (because parietal cortex is not the sole cortical area involved in "spatial vision"); and in man because parietal lesions have long been known to lead to impaired perception of incomplete pictures. Moreover, although Ungerleider and Mishkin do not generalise their distinction to touch, this sense deserves consideration, given that Mishkin had earlier argued that vision and touch have a similar organisation. In touch there are direct anatomical connections between the areas serving "object touch" and "spatial touch", and both in man and in the monkey it seems established that the same region processes spatial performance in vision and in touch. Further work is needed particularly on the spatial disorder after parietal lesions: in the monkey, research has frequently been confined to one non-specific test; in man, important differences related to the laterality of the lesion, have only recently emerged. Then "spatial performance" may be found to be more closely linked to motor output than to sensory input.