Mohan Matthen holds that visual perceptual content is divided into descriptive and referential elements. Descriptive content is our awareness of sensory features belonging to objects located in the visual field. Matthen conceives of this in terms of an image. The referential element is a demonstrative form of content, by which we pick out those objects as particulars and assert their physical presence. Matthen terms this ‘the feeling of presence’. Together, they make up the ‘assembled message’ that visual states present to (...) the perceiver in perceptual experience. I argue that the two elements cannot play together as Matthen envisages. Beginning with the referential element, I show that the feeling of presence is unable to both provide demonstrative reference and assert that the visual state reflects how things are. Turning to the descriptive element, I argue that Matthen is committed to a view on which descriptive (or image) content is a mental entity, a reified image. This is a common element across perceptual experience, episodic memory and visual imagination. The mental entity proves a defective ingredient in assembling perceptual content that purports to present reality. There are problems integrating the depicted viewpoint with the subject’s actual viewpoint, and problems maintaining a coherent notion of assertion. I argue that the project of assembling perceptual experience from descriptive and referential elements is unworkable. (shrink)

It is widely held that the content of perceptual experience is propositional in nature. However, in a well-known article, “Is Perception a Propositional Attitude?”, Crane has argued against this thesis. He therein assumes that experience has intentional content and indirectly argues that experience has non-propositional content by showing that from what he considers to be the main reasons in favour of “the propositional-attitude thesis”, it does not really follow that experience has propositional content. In this paper I shall discuss Crane’s (...) arguments against the propositional-attitude thesis and will try to show, in contrast, that they are unconvincing. My conclusion will be that, despite all that Crane claims, perceptual content could after all be propositional in nature. (shrink)

Evidence from cognitive science supports the claim that humans and other animals see the world as divided into objects. Although this claim is widely accepted, it remains unclear whether the mechanisms of visual reference have representational content or are directly instantiated in the functional architecture. I put forward a version of the former approach that construes object files as icons for objects. This view is consistent with the evidence that motivates the architectural account, can respond to the key arguments against (...) representational accounts, and has explanatory advantages. I draw general lessons for the philosophy of perception and the naturalization of intentionality. (shrink)

Malebranche holds that sensory experience represents the world from the body’s point of view. I argue that Malebranche gives a systematic analysis of this bodily perspective in terms of the claim that the five familiar external senses and bodily awareness represent nothing but relations to the body.

A compositional theory of perceptual representations would explain how the accuracy conditions of a given type of perceptual state depend on the contents of constituent perceptual representations and the way those constituents are structurally related. Such a theory would offer a basic framework for understanding the nature, grounds, and epistemic significance of perception. But an adequate semantics of perceptual representations must accommodate the holistic nature of perception. In particular, perception is replete with context effects, in which the way one perceptually (...) represents one aspect of a scene (including the position, size, orientation, shape, color, motion, or even unity of an object) normally depends on how one represents many other aspects of the scene. The ability of existing accounts of the semantics of perception to analyze context effects is at best unclear. Context effects have even been thought to call into question the very feasibility of a systematic semantics of perception. After outlining a compositional semantics for a rudimentary set of percepts, I draw on empirical models from perceptual psychology to show how such a theory must be modified to analyze context effects. Context effects arise from substantive constraints on how perceptual representations can combine and from the different semantic roles that perceptual representations can have. I suggest that context effects are closely tied to the objectivity of perception. They arise from a perceptual grammar that functions to facilitate the composition of reliably accurate representations in an uncertain but structured world. (shrink)

From the first-person point of view, seeing a red square is very different from thinking about a red square, hearing an alarm sound is very different from thinking that an alarm is sounding, and smelling freshly-roasted coffee is very different from thinking that there is freshly-roasted coffee in one’s vicinity. How might the familiar contrast between representing a fact in thought and representing it in perception be captured? One influential idea is that perceptual states are phenomenally conscious whereas thoughts are (...) not. However, those theorists who hold that thoughts have a distinctive kind of phenomenal character – often known as “cognitive phenomenology” – cannot account for the contrast between thought and perception in this manner. This paper examines the various options that are available to advocates of cognitive phenomenology for capturing the experiential contrast between thought and perception, and argues that each of them faces serious challenges. (shrink)