Subjective versus objective language proficiency measures in the investigation of bilingual effects on cognitive control

Abstract

Purpose: Bilingual language experience is thought to confer non-linguistic benefits in general cognition including improved cognitive control. These bilingual effects are most often observed in samples of bilinguals who are highly proficient in both languages. However, across the majority of previous studies, assessments of language proficiency are exclusively subjective. While evidence supports that subjective and objective measures of language proficiency are correlated, no studies have explored whether the use of either measure impacts on model results when investigating bilingual effects on cognitive control. Methodology: Mandarin-English bilingual young adults completed both subjective and objective assessments of language proficiency and a Simon task to measure differences in cognitive control. Data and Analysis: Data were analyzed using linear mixed-effects models to best account for differences in linguistic and non-linguistic variables as well as the repeated-measures nature of the Simon task. Findings and Conclusions: We report no evidence in support of improved cognitive control associated with higher levels of language proficiency. Crucially, results did not differ when either subjective or objective measures were included in our models. Results support that both subjective and objective assessments of language proficiency may be equivalent when modeling bilingual effects on cognitive control. Originality: This study is the first direct investigation of the influence of the proficiency assessment method on model results. Implications: The findings of this study have implications for the assessment of language proficiency in future investigations of bilingual effects.