Context and word class influences on lexical retrieval in aphasia

Background: Perceived differences in word retrieval between naming and connected speech contexts have been of clinical interest to clinicians and researchers for the last 50 years. However, empirical studies of contextual influences on retrieval have surfaced only in the last two decades. Aims: The purpose of this exploratory investigation was to study lexical retrieval in both connected speech and naming tasks in a group of participants with mild aphasia. As previous studies have outlined possible relationships between word class (noun vs. verb) and elicitation context, retrieval of both nouns and verbs were investigated in naming and connected speech tasks. Methods & Procedures: Twenty individuals with mild aphasia presenting primarily as residual anomia and 10 age- and education-matched controls participated in this study. Accuracy in lexical retrieval attempts was compared across two contexts - confrontation naming and video narration, for targets in two word classes - nouns and verbs. Lexical targets in these two word classes had been matched as groups on an a priori basis for frequency and familiarity Outcomes & Results: Results indicated that all participants experienced greater word finding difficulty in confrontation naming than in video narration. Individuals with aphasia also demonstrated greater word finding difficulty for nouns than verbs, although subsequent analyses suggest that these findings may be explained by joint effects of word length and word frequency on of target stimuli. Conclusions: Context influenced lexical retrieval of both aphasic and normal aging subjects, although "word class'' effects may have been attributable to combined characteristics of lexical targets. Although results appear to have implications for both assessment and treatment of lexical retrieval deficits, discussion emphasizes ways in which word finding abilities in connected speech may require different methods of study than those traditionally employed in the analysis of naming responses.