Project Grant 2214017

This National Science Foundation project grant of $550,000 supports research to evaluate computational models of infant language acquisition against experimental data on English- and Spanish-learning infants. Funded under the Social, Behavioral, and Economic Sciences program (CFDA 47.075), the award will enable researchers at the University of California, Los Angeles to directly compare the output of different computational word segmentation models to infant behavioral data on the acquisition of sound sequencing restrictions in their native language. Over the four-year period from September 1, 2022 to August 31, 2026, this will provide insight into the developmental mechanisms infants use to learn language and a better understanding of the timeline for acquiring these phonological rules. It will also serve as a template for future studies benchmarking computational models against infant behavior, with the aim of distinguishing between alternative hypotheses of language learning in development.