Engaging proactive control: Influences of diverse language experiences using insights from machine learning

Area of Science:

• Cognitive Neuroscience
• Psycholinguistics
• Computational Linguistics

Background:

• The relationship between bilingualism and executive functions remains debated.
• Previous research suggests potential links, but effect sizes and methodologies vary.

Purpose of the Study:

• To investigate the association between multidimensional bilingual experience and proactive executive control.
• To determine if machine learning models can robustly predict executive control from bilingualism metrics.

Main Methods:

• Recruited over 400 young adult bilinguals.
• Assessed proactive executive control using reaction time on the AX continuous performance task (AX-CPT).
• Quantified bilingual experience across age of acquisition, language entropy, and second language exposure; employed linear mixed-effects and LASSO regression with cross-validation.

Main Results:

• Significant associations were found between bilingual experience and proactive control.
• Machine learning models demonstrated robustness in predicting data from novel participants.
• Simpler models using only task performance (AX-CPT trial type) achieved similar predictive power, indicating small effects of bilingualism.

Conclusions:

• The impact of bilingual language experience on proactive executive control in young adults is likely small.
• Future research requires larger or distinct samples (e.g., older adults, children) and refined measures of language experience to reveal predictive effects.