Federal Project Grant Award Summary

New York University received $435,974 in funding from the National Institute of Mental Health (NIMH) under the Mental Health Research Grants program (CFDA 93.242) for a project running from May 1, 2026 through April 30, 2028. The project develops a novel computational framework utilizing very small recurrent neural networks (RNNs) and dynamical systems theory to model individual learning and decision-making processes in a flexible and interpretable manner. The research focuses on reward learning mechanisms and aims to capture behavioral idiosyncrasies in individual subjects, enabling more accurate prediction of individual choices compared to classical modeling approaches.

The primary deliverables include computational models and methodologies that characterize individual differences in task performance and predict how behavior in psychological tasks relates to psychiatric symptoms. By combining neural network-based cognitive models with interpretable analytical techniques, the project generates research products including peer-reviewed publications, validated computational tools, and mechanistic insights into healthy and dysfunctional decision-making processes. These outputs advance the field's understanding of cognitive processes underlying mental health conditions and support NIMH's mission to transform the understanding and treatment of mental illnesses through rigorous basic and clinical research.