Domain-Specific Knowledge Systems in the Brain: The Animate-Inanimate Distinction

Abstract

We claim that the animate and inanimate conceptual categories represent evolutionarily adapted domain-specific knowledge systems that are subserved by distinct neural mechanisms, thereby allowing for their selective impairment in conditions of brain damage. On this view, (some of) the category-specific deficits that have recently been reported in the cognitive neuropsychological literature - for example, the selective damage or sparing of knowledge about animals - are truly categorical effects. Here, we articulate and defend this thesis against the dominant, reductionist theory of category-specific deficits, which holds that the categorical nature of the deficits is the result of selective damage to noncategorically organized visual or functional semantic subsystems. On the latter view, the sensory/functional dimension provides the fundamental organizing principle of the semantic system. Since, according to the latter theory, sensory and functional properties are differentially important in determining the meaning of the members of different semantic categories, selective damage to the visual or the functional semantic subsystem will result in a category-like deficit. A review of the literature and the results of a new case of category-specific deficit will show that the domain-specific knowledge framework provides a better account of category-specific deficits than the sensory/functional dichotomy theory.

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

“…In the domain-specific account (Caramazza & Shelton, 1998), conceptual knowledge is organized by category, with evolutionary demands generating specialization for a small number of categories such as living things and plant life. On this account we might expect performance on living things to be facilitated (i.e., overall faster RTs and fewer errors) relative to performance on nonliving things.…”

“…However, it represents an advance over these earlier models in that it addresses one of the criticisms that has been leveled against them-they are so flexible that they can explain any pattern of deficit and are therefore theoretically unhelpful (e.g., Caramazza & Shelton, 1998). By developing specific claims about conceptual structure, based on well-supported claims in the psychological literature (e.g., Keil, 1986Keil, , 1989Malt & Smith, 1984), we have been able to constrain the power of the account and make falsifiable predictions, thus overcoming this kind of objection.…”

“…One type of account is that conceptual knowledge is explicitly organized into domain-specific subsystems, each of which may be selectively impaired by brain damage (e.g., Caramazza & Shelton, 1998;Santos & Caramazza, 2002). However, the breakdown of conceptual knowledge does not always respect domain boundaries.…”

Electrophysiological evidence for a natural/artifactual dissociation

“…The natural/artifactual dissociation has also been considered from an evolutionary point of view. Caramazza and Shelton (1998) proposed that this dissociation reflects the discrete organization in the brain of different domains of knowledge. At the neuroanatomical level, this distinction emerges as a function of the roles of natural and artifactual objects in the evolution of the human brain: living objects represent food and potential predators, whereas inanimate objects might be manipulated to obtain food or protect against predators.…”

“…This task has already been used in Italian in an fMRI study by Miceli et al (2002), who pointed out that the semantic content of a noun bears an arbitrary relationship to its grammatical gender, so that gender effects cannot be ascribed to the organization of knowledge in the brain. We predicted that if natural and artifactual domains constitute separate semantic subsystems with different neurological substrates (Warrington and Shallice, 1984;Caramazza and Shelton, 1998), the scalp topography of the ERPs corresponding to each of these domains would differ regardless of the task. Furthermore, according to the multiple subsystems theories, differences between domains would occur in the semantic processing phase and coinciding with the N400 component.…”