Federal Grant Award Summary

Georgia TECH Research Corp received a $2,596,724 Project Grant from the National Institute of Mental Health (NIMH) under the Mental Health Research Grants program (CFDA 93.242), effective April 1, 2026 through March 31, 2030. The award funds research on multiscale models for understanding multi-animal interactions and intent, an investigation into the cognitive and neural mechanisms underlying social behaviors and decision-making processes. The research employs a novel framework inspired by linguistic analysis, breaking down complex multi-animal behaviors into fundamental units (behavioral syllables), examining their combinations and grammatical-like rules (syntax), and ultimately understanding behavioral intent. This approach addresses limitations in existing automation methods for behavioral analysis by capturing the complexity and context of social interactions among multiple animals.

The funded research progresses across three primary objectives: developing machine learning methods to identify latent behavioral representations and segment multi-animal behavioral time series into social syllables (Aim 1); extracting longer interaction sequences called motifs using transformer-based models to reveal behavioral syntax (Aim 2); and determining behavioral intent through compositional analysis (Aim 3). Findings from this project will advance understanding of social bonding, aggression, cooperation, and collective behaviors at the neural and cognitive level, with potential applications to fundamental neuroscience and behavioral research methodologies.