Spatial attention and perception: seeing without paint

Abstract

Covert spatial attention alters the way things look. Objects situated at attended locations appear bigger, closer, if striped, stripier than qualitatively indiscernible counterparts whose locations are unattended. These results cannot be easily explained in terms of the number and kind of perceived properties of objects. Nor do they appear to be cases of visual illusions. Ned Block has argued that these results are best accounted for by invoking what he calls 'mental paint'. In this paper I argue, instead, for an account of these phenomena in terms of the perception of action scaled affordances concerning saccadic eye movement. As part of the argument I draw connections with the empirical literature on the way in which performance efficiency alters visual appearance. Attention alters the way things look. There is incontrovertible empirical evidence showing that objects situated at attended locations appear bigger, closer, if striped, stripier than qualitatively indiscernible counterparts whose locations are unattended. The colours of their surfaces look more saturated. In experimental conditions the onset of a dot in an attended portion of a screen seemed earlier than a simultaneous dot situated at a less attended location. Similarly the flicker rate of flickering objects seems faster when we attend to their locations. 1 At first sight these are not instances