Conceptual Structure and the Structure of Concepts: A Distributed Account of Category-Specific Deficits

Abstract

We present a new account of the fine-grained structure of semantic categories derived from neuropsychological, behavioral, and developmental data. The account places theoretical emphasis on the functions of the referents of concepts. We claim (i) that the distinctiveness of functional features correlated with perceptual features varies across semantic domains; and (ii) that category structure emerges from the complex interaction of these variables. The representational assumptions that follow from these claims make strong predictions about what types of semantic information are preserved in patients showing category-specific deficits following brain damage. These claims are illustrated with a connectionist simulation which, when damaged, shows patterns of preservation of distinctive and shared functional and perceptual information which varies across semantic domains. The data model both dissociations between knowledge for artifacts and for living things and recent neuropsychological evidence concerning the robustness of functional information in the representation of concepts.

Distinctiveness and Correlation in Conceptual Structure: Behavioral and Computational Studies.

“…In all the analyses there was a significant difference between living and nonliving things for the distinctive feature conditions but little or no difference for the shared features. This pattern of results is precisely as predicted by the conceptual structure account Tyler et al, 2000).…”

Progressive knowledge loss: A longitudinal case study

“…At least in the case of our patient, living things seem to crucially rely on perceptual processing, although there are differences due to task difficulties, as pointed out by Tyler et al (2000). However, no theory seems to account for the generalized decline of functional knowledge of living things that we observed during our second assessment of VZ.…”

“…For the mild-tomoderate-degeneration stages, Tyler et al (2000) predicted that only the ability to determine the category to which living stimuli belong would be spared. Keeping in mind VZ's general loss of knowledge for items of both categories, note that (1) she performs best with the help of functional features of nonliving things, and that (2) her ability to identify objects by referring to actions that can be performed with them (by producing verbs in her attempts to answer) is still intact (see also Bird et al, 2000;Bak & Hodges, 2003;Silveri et al, 2003).…”

“…More often, the deficits show fractionations, involving small categories instead (e.g., Hart et al, 1985;Hillis & Caramazza, 1991), such as, a deficit for, say, knowledge of animals that spares knowledge of fruits and vegetables. Devlin et al (1998) and Tyler et al (2000) took a different perspective, and proposed correlational models that assume a unitary semantic memory system. The model by Devlin and colleagues is based on the finding of Gonnerman et al (1997) that in the early stages of Alzheimer's disease (AD) patients show a category-specific deficit for nonliving things, whereas in later stages this effect reverses, with a greater deficit for living things instead.…”

“…The model proposed by Tyler et al (2000) focuses, on the one hand, on the close interdependency between living and nonliving categories, which is based on a complex relation between two dimensions of features: (a) their nature (i.e., types), being either perceptual or functional; and (b) their distribution across the network being either shared (a feature is common to many exemplars of a semantic category) or distinctive (a feature is typical of a specific exemplar or of very few exemplars). Different combinations of these two dimensions (i.e., perceptual0distinctive; perceptual0shared; functional0distinctive; functional0shared) characterize semantic features.…”