Catherine Diaz-Asper has received new funding from the NIH AIM-AHEAD Hub Specific Projects – Year 4 Grant to lead an innovative research project focused on improving early detection of cognitive decline in bilingual older adults.
The project, titled “Cognitive screening using automated speech analysis: the effect of preferred language in bilingual Spanish-English speakers,” will run from December 2025 through July 2026. Dr. Diaz-Asper will collaborate with co-investigators Nathan Green (School of Technology and Innovation), Sara Pappa (School of Health Sciences), and Mahederemariam Bayleyegn Dagne (Marymount Center for Optimal Aging).
Machine learning–based speech analysis has shown promise in detecting Alzheimer’s-related cognitive impairments by examining features such as word choice, sentence structure, pauses, and speech rhythm. However, most existing models are trained on monolingual English speakers, limiting their accuracy and applicability for bilingual populations.
This project aims to address that gap by developing and evaluating an AI-driven cognitive screening tool tailored for older bilingual Spanish-English speakers. Using accessible technology, the research team will assess how well automated speech analysis identifies mild cognitive impairment in both languages and whether model performance and linguistic features differ based on an individual’s preferred language.
By centering bilingualism in cognitive assessment, this work has the potential to reduce disparities in early detection of cognitive decline and support more equitable, inclusive approaches to brain health.
