Hierarchical Compression Reveals Sub-Second to Day-Long Structure in Larval Zebrafish Behavior

Animal behavior is dynamic, evolving over multiple timescales from milliseconds to days and even across a lifetime. To understand the mechanisms governing these dynamics, it is necessary to capture multi-timescale structure from behavioral data. Here, we develop computational tools and study the behavior of hundreds of larval zebrafish tracked continuously across multiple 24-h day/night cycles. We extracted millions of movements and pauses, termed bouts, and used unsupervised learning to reduce each larva's behavior to an alternating sequence of active and inactive bout types, termed modules. Through hierarchical compression, we identified recurrent behavioral patterns, termed motifs. Module and motif usage varied across the day/night cycle, revealing structure at sub-second to day-long timescales. We further demonstrate that module and motif analysis can uncover novel pharmacological and genetic mutant phenotypes. Overall, our work reveals the organization of larval zebrafish behavior at multiple timescales and provides tools to identify structure from large-scale behavioral datasets.Significance Statement Behavior is dynamic and not only can change from 1 s to the next but also can unfold over many hours or even days. Understanding how behavior is organized on these different timescales is a critical task in neuroscience, because the constraints on and patterns of behavior offer important clues about the underlying computations being performed in the brain. The analysis tools we develop in this manuscript and apply from sub-second to day-long larval zebrafish behavior expands our understanding of how behavioral patterns change at multiple timescales. The computational metrics we describe can now be used to understand the behavioral consequences of psychotropic drugs or genetic lesions associated with neurodevelopmental and neuropsychiatric disorders.