Attention seems to raise a problem for pure representationalism, the view that phenomenal content supervenes on representational content. The problem is that shifts of attention sometimes seem to bring about a change in phenomenal content without a change in representational content. I argue that the representationalist can meet this challenge, but that doing so requires a new view of the representational content of perception. On this new view, the representational content of perception is always relative to a way of attending. (...) I call this the attention-indexed view of perceptual content. (shrink)

Students of perception have long known that perceptual constancy is an important aspect of our perceptual interaction with the world. Here is a simple example of the phenomenon concerning color perception: there is some ordinary sense in which an unpainted ceramic coffee cup made from a uniform material looks a uniform color when it is viewed under uneven illumination, even though the light reflected by the shaded regions to our eyes is quite different from the light reflected by the unshaded (...) regions to our eyes (see figure 1). Or consider this example concerning size perception: there is some ordinary sense in which two telephone poles look the same size when the first is viewed from 100 meters and when the second is viewed from 1 meter, even though the visual angle subtended by the two poles on our retinae is very different (see figure 2). Or consider this example concerning shape perception: there is some ordinary sense in which a penny looks round both when viewed head on and when viewed from an acute angle, even though the area projected by the penny onto our retinae under these two conditions is very different (see figure 3). Or, finally, consider this example concerning auditory volume perception (which I cannot depict graphically): there is some ordinary sense in which a speaker’s voice sounds the same volume when heard from across the room and when heard from a distance of 1 meter, even though the energy striking our ears under these two conditions is very different. (shrink)

There is nothing in this World constant but Inconstancy. [Swift 1711: 258] In this paper I argue that two standard characterizations of colour constancy are inadequate to the phenomenon. This inadequacy matters, since, I contend, philosophical appeals to colour constancy as a way of motivating illumination-independent conceptions of colour turn crucially on the shortcomings of these characterizations. After critically reviewing the standard characterizations, I provide a novel counterfactualist understanding of colour constancy, argue that it avoids difficulties of its traditional rivals, (...) and defend it from objections. Finally, I show why, on this improved understanding, colour constancy does not have the philosophical consequences that have been claimed for it in the literature. (shrink)

The subject of the present thesis is the phenomenon of simultaneous colour contrast: As is well known, the perceived colour of a given light stimulus depends almost as strongly on the stimulation of neighbouring regions of the visual field as on the local stimulus itself. Thus, the perceived colour of a target stimulus can be manipulated either by changing the colour co-ordinates of the target itself, or, alternatively, by changing the colour co-ordinates of the surround. Classical models of simultaneous contrast (...) are based on the implicit assumption that the changes in perceived colour brought about by these two kinds of manipulation are basically of the same kind, and thus may compensate each other. This ‘compensation assumption’ is equivalent to the widespread assumption that the space of perceived colours is three-dimensional. In principle, though, the well-established facts of trichromatic theory are compatible with a dimensionality of the space of perceived colours higher than three (namely three times the number of differently coloured regions in the visual field). One of the central results of the present investigations is that the compensation assumption seems to be unwarranted in the case of simple centre-surround stimuli. This instilled the expectation that classical methods for measuring simultaneous contrast effects, such as grey settings and asymmetric colour matching, may sometimes be inappropriate, and thus yield misleading results. In further experiments it is shown that this is indeed the case when uniformly coloured surrounds are used. In the case of variegated surrounds, however, matters appear to be much less complicated. This difference seems to be due to impressions of perceptual transparency being evoked by uniform surrounds, but not by variegated ones. (shrink)