PATTERNS OF COOCCURRENCE AMONG THE ASSOCIATIVE AGNOSIAS - IMPLICATIONS FOR VISUAL OBJECT REPRESENTATION

The patterns of co-occurrence among associative agnosia for faces, words and other objects are analysed and found to be consistent with the existence of two, rather than three, underlying visual recognition capacities. Different degrees and combinations of damage to these two capacities can account for the five different combinations of word, face, and object agnosia that are found, as well as for the apparent absence of two particular combinations. A tentative interpretation of the two inferred visual recognition capacities is offered, in terms of the idea of structural representations.