Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Don’t try and pass them by: Trilinguals’ cognitive control abilities pattern differently from bilinguals’
Poster B36 in Poster Session B, Tuesday, October 24, 3:30 - 5:15 pm CEST, Espace Vieux-Port
Ariane Senécal1, Laura Sabourin1; 1University of Ottawa
While many studies integrate multilinguals under the banner of bilingualism without fanfare, others explicitly exclude them from their samples. Does multilingualism warrant separate assessment, or is it justified for bilingualism to include multilinguals by default? Research overtly acknowledging multilingualism has habitually focused on processing, acquisition, and cross-linguistic influence (e.g., de Bruin et al., 2023), but studies comparing bilingual and tri/multilingual participants’ executive functions have observed differences (e.g., Guðmundsdóttir & Lesk, 2019; Hsu, 2014; Madrazo & Bernardo, 2018). Our own pilot study of French–English bilinguals vs. trilingual speakers of French–English and one other language found a monitoring advantage—a readiness to respond to all types of trials, revealed through faster reaction times—for trilinguals compared to bilinguals, as well as group-level language-specific asymmetries pointing to different management strategies (Senécal, Gosselin & Sabourin, 2022). The current follow-up study compared the cognitive control abilities of 21 Spanish–English bilingual and 30 Spanish–English–French trilingual adults. Participants completed an online study including flanker and Stroop tasks, the latter with single-language Spanish and English blocks as well as a mixed Spanish-English block, to assess whether a more challenging task may reveal further differences (Sabourin & Vīnerte, 2015). Participants’ cognitive control abilities were represented as global (i.e., overall reaction times), facilitation (i.e., difference in averaged reaction times for congruent and neutral trials), and inhibition (i.e., difference in averaged reaction times for incongruent and neutral trials) effects. Linear regression models and mixed models were used to evaluate the relation between participant characteristics (demographics and linguistic behaviour, including social diversity of language use (i.e., language entropy (Gullifer & Titone, 2020)) and cognitive control. Bilinguals and trilinguals’ results differed in both the domain-specific (linguistic Stroop) and domain-general (non-linguistic Flanker) tasks. Contrary to the pilot study, trilinguals did not appear to possess monitoring advantages relative to bilinguals, having overall slower average reaction times (β = 257.09, t = 3.78, p <.01 for flanker; β = 1022.99, t = 4.48, p < .01 for Stroop). Bilinguals also generally showed greater facilitation (e.g., overall facilitation in the mixed block: β = -225.80, t = 2.00, p = .05) and inhibitory control (e.g., inhibition of English items: β = -308.29, t = -1.98, p = .05) in the Stroop task. However, some divergent trends for interactions with language entropy and English proficiency pointed to significant distinctions between the groups, more often seen in the mixed-language Stroop block. In many cases, these interactions reversed the better performance seen for bilinguals: our most consistent result saw trilingualism showing a protective effect against age-related decline, as previously reported in Alzheimer research (Chertkow et al., 2010). Overall, this study supports the view that multilinguals are not just bilinguals with bonus languages, but rather speakers with distinct characteristics that may impact their cognitive control abilities. We assert that though their exclusion from studies of bilingualism may not be justified, it would be prudent to consider multilinguals’ data separately at the onset, should they pattern differently from bilinguals.
Topic Areas: Multilingualism, Control, Selection, and Executive Processes