BRIDGING EXPERIMENTAL CONTROL AND ETHOLOGY: A MODULAR ARENA FOR QUANTITATIVE ANALYSIS OF COMPLEX MOUSE BEHAVIOR

Animals continuously sample their environment through a diverse set of specialized sensory systems. The information acquired shapes behavioral responses that are flexibly adjusted to a dynamically changing environment. In mammals, such as mice, behavior emerges from the integration of multiple environmental factors, for instance food and water availability, predator threats, and social interactions with conspecifics. Capturing this complexity in the laboratory remains a central challenge in behavioral neuroscience. Here, we present a semi-natural, highly modular experimental arena designed to study complex mouse behavior under controlled yet ethologically relevant conditions. The setup enables high-throughput experiments with minimal experimenter interference and supports continuous, high-quality video recordings of animal behavior. Combined with pose-estimation–based behavioral analysis using DeepLabCut, this approach allows for efficient and quantitative assessment of behavior. Importantly, the modular design permits rapid adaptation of the experimental paradigm to a wide range of research questions without the need to construct new behavioral arenas. We demonstrate system suitability in three proof-of-principle paradigms: first, by showing olfactory-driven behavioral changes in single animals; second, by implementing a three-chamber assay within the same arena to highlight its modularity; and third, by extending the setup to multi-animal experiments to capture complex social interactions. Together, this platform provides a robust and flexible tool for investigating mouse behavior in a controlled, yet naturalistic laboratory environment.