The scan-copier mechanism and the positional level of language production: Evidence from phonemic paraphasia

Abstract

In this paper I argue that phonemic paraphasia in the fluent aphasias is best characterized as a computational disruption in the processes that map the Functional Level onto the Positional Level of the model of sentence production developed by Merrill Garrett. I claim that a scan-copying mechanism of the sort proposed by Shattuck-Hufnagel operates among the computations leading to the Positional Level and that mechanism, when it derails, gives rise to the production of phonemic level errors in the language of fluent aphasies with posterior left cerebral lesions. I then argue that phonetic level errors in non-fluent, verbal apraxia cannot be handled by the scan-copier, nor can they be accounted for at the Positional Level.