Wernicke's aphasia and normal language processing: A case study in cognitive neuropsychology

Abstract

We report the results of a cognitive investigation of the language deficits of a single Wernicke's aphasic patient. The patient, R.D., showed poor speech comprehension but good reading comprehension. His spontaneous speech and his attempts at reading aloud contained many neologisms and some verbal paraphasias. Following Butterworth (1979) we interpret the neologisms as due to problems with retrieval of the phonological specifications of words from a speech output lexicon, and we present evidence showing that success in lexical retrieval was affected by word frequency and not by any syntactic distinction between content (open-class) and function (closed-class) words. R.D.'s spelling was better preserved than his spoken naming and he could spell many words he was unable to say correctly. His spelling errors appeared to be attempts at a target word based on retrieval of partial orthographic information (just as his neologisms seem to be based on partial retrieval of phonological information). We argue that normal subjects may make similar speech and writing errors under certain circumstances. Garrett's (1982) model of speech production is presented and we discuss how neologistic jargonaphasia and other forms of Wernicke's aphasia may be explained in terms of it. We also argue that when R.D.'s deficits are analyzed and compared with those of other patients in the literature a number of implications emerge for theories of normal language processing. These include 1) that the comprehension and production of familiar written words do not involve obligatoruy phonological mediation, 2) that there are distinct phonological and orthographic lexicons, 3) that morphemes are seperately represented in the phonological lexicon, 4) that ease and speed of retrieval of items from the phonological lexicon is affected by frequency of usage, and 5) that retrieval from both lexicons is not all-or-nothing.