In search for common patterns in aphasia: The case of Spanish wh-dependencies

Abstract

BackgroundResearch on comprehension of wh-dependencies in Broca’s aphasia has resulted in discrepant findings, suggesting that structurally different languages may involve different comprehension patterns. AimsIn the present study, we investigated whether Spanish speakers with Broca’s aphasia rely on a language-specific cue that might facilitate comprehension of object wh-dependencies – the preposition a. Since the preposition a in Spanish indicates animate objects, reliance on this cue would facilitate comprehension of more demanding object wh-dependencies, cancelling out processing differences between structures with subject and object extractions. MethodsWe studied comprehension of subject vs. object quién (“who”) and qué (“which”) direct and embedded questions, and relative clauses introduced by que in seven speakers with aphasia (SWA): six with Broca’s and one with mixed aphasia. Control data were obtained from ten neurologically intact elderly Spanish speakers. ResultsMann-Whitney test revealed that compared to healthy controls, SWA had preserved comprehension of both subject and object who direct questions, object which direct questions, and object relative clauses. However, they exhibited considerably worse comprehension of subject which direct questions, all embedded questions regardless of the extraction site (subject, object) or type of wh-word (who, which), and subject relative clauses. Individual comprehension profiles obtained in a modified t-test and revised standardized difference test, while containing some variability, largely support lack of subject-object asymmetry and objects-better patterns. Furthermore, in direct questions, comprehension of who questions was better than the comprehension of which questions, while the opposite was found in embedded questions. ConclusionsThe lack of significant subject-object asymmetry in most conditions, together with better comprehension of object than subject extractions in direct questions, indicates that Spanish SWA rely on the preposition a in the comprehension of object wh-dependencies. However, this strategy fails in more complex conditions (embedded questions), because the processing costs outweigh the facilitating effects of this cue, unless additional cues to the object are present (object relatives). The present findings have implications for the accounts of wh-dependencies in Broca’s aphasia involving Discourse-linking, the Intervener hypothesis, and the Competition model.