Slide Slam A16
Clustering and switching in verbal fluency across different language contexts: Evidence from bilingual aphasia
Erin Carpenter1, Claudia Peñaloza1, Leela Rao1, Swathi Kiran1; 1Boston University
Introduction. Understanding levels of control during language production in bilingual persons with aphasia (BPWA) remains a highly complex topic. In bilingual speakers, we can operationalize two levels of control which are critical for successful language production. First, language control, manages interference from the non-target language across different interactional contexts, a factor which is modulated by relative proficiencies in each language. Second, semantic executive control directs activation and inhibition of lexical candidates in a contextually appropriate manner. These two levels work in conjunction to ensure effective communication for bilingual speakers. In BPWA, damage to the language system results in deficits in language production, however it is not yet well understood if these deficits arise, in part, from the level of language control, semantic executive control, or a combination of the two. Verbal fluency tasks provide a useful opportunity to examine these two levels of control in BPWA. Aims. Therefore, this research examined clustering and switching performance on semantic category generation and letter fluency tasks across four language conditions for thirty-five Spanish-English BPWA and twenty-two Spanish-English healthy bilinguals (HB). Here, clustering refers to successively producing lexical items belonging to a semantic subcategory, or with overlapping phonemic features, while switching is a transition from one cluster to another. Methods. All tasks were completed in the first- (L1) and second-acquired (L2) language as identified by a language use questionnaire (LUQ). The four language conditions implemented in the semantic category generation task consisted of two No-Switch (NS-L1 and NS-L2) conditions where participants responded in only their L1 or L2, one Self-Switch (SS) condition, where participants switched between languages as desired, and one Forced-Switch (FS), which required participants to switch between languages after each response. Participants also completed a traditional letter fluency task in each language (LF-L1 and LF-L2). Results. Overall, we found that HB outperformed BPWA across all measures, reflective of damage to the language system for BPWA. In the semantic category generation task, clustering performance did not differ across conditions, but both groups demonstrated superior switching performance in SS and NS-L1 compared to both NS-L2 (p=.015 for SS and p=.001 for NS-L1) and FS (p=.002 for SS and p=.004 for NS-L1). This suggests that increased control demands originating from the level of language control (i.e., switching between languages in a controlled manner in FS or inhibiting the prepotent language in NS-L2) impedes participants’ ability to implement semantic executive control processes to systematically search within the lexicon, while automatic spreading activation engaged during clustering was not affected. In the letter fluency task, we found a significant Group × Condition interaction [F(1, 54)=5.900, p=.019] for clustering performance, indicating that HB produced larger cluster sizes in LF-L1 compared to LF-L2, but BPWA performed similarly across conditions, suggesting BPWA were sensitive to increased semantic executive control demands imposed by this task, resulting in reduced clustering performance in both languages. Conclusion. In sum, this study highlights that BPWA demonstrate semantic executive control deficits which are further impacted by increasing language control demands.