Short term animal movements reveal learning and individual traits
Representing short-term movement to reveal animal identity and learning progress
Machine LearningArtificial Intelligence
Summary
It is often hard to tell how animals learn over time using just obvious behaviors. The authors show that small snippets of an animal's movement, even just one second long, carry information about who the animal is and how much it has learned. Their method learns these movement patterns automatically without needing experts to label behaviors. They find that movements related to the animal’s own body parts predict identity well, while overall movement in the environment signals learning progress. This approach could help study learning and individual differences in animals in a new way.
What this means in practice
- •For animal behavior researchers: Use short movement patterns to identify individual animals and assess their learning stages without manual behavior labeling.
- •For wildlife monitoring teams: Detect individuality and learning signals in free-ranging animals through brief movement tracking for better monitoring and conservation.
Tested on one dataset.
Authors
Ping Zhang, Quang-Huy Nguyen, Yingtie Lei, Sameera Shridhar, Zachary Leveroni, Ellie Quattrocchi, Jai Y Yu, Wei-Lun Chao
Abstract
How learning progresses over time and varies across individuals is a longstanding question in cognitive and behavioral science. However, a major challenge is identifying observable behavioral measures that faithfully capture learning progression. Existing approaches commonly rely on predefined behavioral features or task-level performance, while fine-scale posture and movement in freely behaving animals remain less explored. In this paper, we ask: can short-term movement trajectories reveal learning progress and individual identity? We introduce a self-supervised representation learning framework built on a simple hypothesis: behavioral characteristics associated with identity and learning remain relatively stable over nearby time intervals. Using temporal proximity as supervision, our framework learns compact representations directly from movement trajectories without manually defining behavioral states or kinematic features. Using behavior data from freely behaving rats, we show that the learned representations not only improve prediction of identity and learning-related readouts over existing methods, but also provide a tool for investigating broader scientific questions about when and how these signals appear in behavior. Specifically, trajectories as short as one second contain informative learning and identity signals. Egocentric pose dynamics are more predictive of identity, whereas allocentric motion is more predictive of learning-related variation, and both signals are strongest at decision junctions in the alternation cohort. We further use a representation-conditioned diffusion model to inspect movement patterns associated with identity and learning progress. Our findings establish short-term movement as an informative behavioral "fingerprint" for studying learning and individual differences. Our code is available at https://github.com/pizhn/TRAIL.