ABSTRACT
On this hypothesis, semantic representations of objects are distributed across modality-specific semantic stores that are defined in terms of the types of features they encode (Farah & McClelland, 1991; Hart & Gordon, 1992; Humphreys & Forde, 2001; Shallice, 1988; Warrington & Shallice, 1984; Warrington & McCarthy, 1983). One variant of the hypothesis is illustrated in Figure 6.2(b), which contains separate semantic stores for sensory or perceptual properties of objects, and for functional information relating to object use. Selective impairments for living and nonliving things are assumed to derive from an asymmetry in the representation of living and nonliving things across the underlying modality-specific stores. Specifically, it is assumed that the semantic representations of living things are more heavily weighted in terms of visual sensory features than functional features (Labov, 1973). In contrast, representations of nonliving things are assumed to be more heavily weighted in terms of functional than visual features. Thus, selective damage to the modality-specific store that encodes visual sensory features is more likely to impair representations of living things than nonliving things. In contrast, damage to the modality-specific store that encodes functional features will result in a disproportionate impairment for nonliving things. The hypothesis also provides a possible explanation for the apparent associations among impaired and preserved categories found in some patients. For example, impairments with musical instruments might be more likely to co-occur with deficits for living than nonliving things, because musical instruments, like objects from animate categories, arguably tend to be distinguished from one another largely on the basis of sensory attributes, such as their visual appearance and the sound that they make. In contrast, body parts, like many inanimate objects, may tend to be more heavily weighted in terms of functional properties than other biological kinds (Farah & McClelland, 1991; Warringon & McCarthy, 1983), and therefore more vulnerable to impairments affecting functional knowledge.
