Influence of contextual constraint on lexical ambiguity resolution in normal aging and Broca's aphasia

Disordered sentence comprehension in aphasia may be due in part to deficient lexical-semantic processes. There are conflicting accounts as to whether these difficulties occur in initial word meaning activation, or in later meaning integration processes. Any such deficits in aphasia, however, must be interpreted with reference to lexical access processes in normal comprehension.; Despite a wealth of research, the mechanisms involved in the retrieval of word meanings in sentence contexts are the subject of vigorous debate. Essentially, two competing psycholinguistic models of initial meaning activation can be distinguished: an interactive account, postulating a constraining influence of context on meaning activation; and a modular account, claiming that meaning activation is autonomous and independent of context.; To contrast predictions of these accounts, this study used a cross-modal priming task to investigate activation of alternate meanings of ambiguous words in sentence contexts biased towards specific semantic features of those meanings. Seven adults with Broca-type aphasia and ten age-matched control subjects made speeded lexical decisions to visually-presented probe words. Probes were either unrelated to, or semantically related to, the dominant or subordinate meanings of ambiguous words embedded in auditorily presented sentences. For dominant-biased contexts, a constraint-based account predicts selective meaning activation, expressed in faster response times to probes reflecting the contextually appropriate meaning of the ambiguity than to other probes. A modular account predicts activation of all meanings, despite biasing context. Thus, response times to both the contextually appropriate and inappropriate probes should be faster than those to unrelated probes.; Neither group evidenced a response time advantage for either type of related probe when compared to unrelated probes, even after adjusting for unequal probe frequency. This unexpected result is not congruent with predictions from either of the models of normal lexical access. The findings for aphasic participants accord with one of the three contrasting accounts for lexical processing deficits in individuals with Broca's aphasia. Due to the uninterpretable findings for the control subjects, however, no conclusions are possible for the participants with aphasia.; The discussion addresses potential methodological confounds in this and previous studies examining word meaning activation in a cross-modal lexical decision task.