Jérôme Dokic

This essay is about our perceptual ability to recognize familiar persons. The question is whether and to what extent our ordinary recognition judgments rely on perceptual experience as opposed to background beliefs. It argues that in order to give a proper answer to this question, we need to introduce a third character into the picture, namely the feeling of presence. Ordinary person recognition involves qualitative recognitional abilities, which (in the visual case) enable us to see that a particular person looks just like a familiar person, but also an independent causal mechanism which generates the feeling that the familiar person herself is present. When we recognize someone, we identify her (general) visual appearance, but we also feel her presence. Feelings of presence are essential contributions to the phenomenology of person recognition, but they do not supervene on perceptual content proper.

Perceptual experiences have been construed either as representational mental states—Representationalism—or as direct mental relations to the external world—Disjunctivism. Both conceptions are critical reactions to the so-called ‘Argument from Hallucination’, according to which perceptions cannot be about the external world, since they are subjectively indiscriminable from other, hallucinatory experiences, which are about sense-data ormind-dependent entities. Representationalism agrees that perceptions and hallucinations share their most specific mental kind, but accounts for hallucinations as misrepresentations of the external world. According to Disjunctivism, the phenomenal character of perceptions is exhausted by worldly objects and features, and thus must be different from the phenomenal character of hallucinations. Disjunctivism claims that subjective indiscriminability is not the result of a common experiential ground, but is because of our inability to discriminate, from the inside, hallucinations from perceptions. At first sight, Representationalism is more congenial to the way cognitive science deals with perception. However, empirically oriented revisions of Disjunctivism could be developed and tested by giving a metacognitive account of hallucinations. Two versions of this account can be formulated, depending on whether metacognition is understood as explicit metarepresentation or as implicit monitoring of first-order informational states. The first version faces serious objections, but the second is more promising, as it embodies a more realistic view of perceptual phenomenology as having both sensory and affective aspects. Affectbased phenomenology is constituted by various metacognitive feelings, such as the feeling of being perceptually confronted with the world itself, rather than with pictures or mere representations.