ABSTRACT
The issue of category specificity and its relevance for the study of brain-behavior relationships is not new. Indeed, it is well embedded in 19th-century and early 20th-century neurology. Consider Broadbent's (1884) introduction to a fascinating paper "On a Particular Form of Amnesia: Loss of Nouns." In his first sentence, he is clear about the neuropsychological objective that continues to inform our current approach: "On two previous occasions I have brought before the Society cases of affection of speech with post-mortem examinations, and have endeavoured by their means to elucidate the mechanism of speech and thought" (p. 249). He went on to describe the case histories of three patients who behaved rationally and cooperatively and who were able to produce fluent and appropriate phrases (e. g., "I am very much better today, thank you" (p. 250) and "I hope you will be able to do me good ... to take this away (p. 253» but were "scarcely if ever known to utter a noun substantive" (p. 250). He considered the anatomical esions in these cases and suggested that the loss of nouns may be due to a disconnection-"cutting off the naming centre from the motor speech apparatus" (p. 259). He also considered the functional implications of this particular naming problem:
Nearly a century later, Damasio and Damasio (1990b) refined this observation. On the basis of modem neuroimaging and detailed behavioral observation, they proposed that systems in the anterior sector of the left temporal cortices "probably constitute the neural basis for the reference lexicon (the collection of words that denotes concrete entities and actions"). They further specified that "damage to the left anterotemporal sector including the temporal pole (area 38) and the anterior part of the inferotemporal region (areas 21, 20, 37), causes a severe defect for naming of concrete entities" (p. 281). Patients possess the generic information about a given animal or object. They are aware of its visual and functional properties. But they cannot access the unique name label. The nominal problem can be further fractionated. With a lesion confined to area 38, the Damasios reported that "the defect is restricted to the retrieval of proper nouns, e.g., the names of persons or places" (p. 282). This intriguing hypothesis has received some experimental support (Semenza & Zettin, 1988) and presumably raises interesting questions for psycholinguists in relation to the organisation of the mental lexicon.
