BAMBI: A Ca²⁺ imaging-based brain-computer interface for longitudinal neuronal tracking in freely behaving mice

Brain-computer interfaces (BCIs) are powerful tools for investigating neuronal dynamics underlying cognitive processes. However, BCIs commonly rely on electrophysiological techniques, which are limited in their ability to track populations of the same neurons over long periods, constraining their utility for studying long-term neuronal processes.

To address this, we developed a BCI system, BAMBI (BCI Applying Mini-microscopes and Behavioral Imaging), which provides feedback based on real-time Ca²⁺ imaging that tracks neuronal activity over extended periods in freely behaving mice.

By maintaining a stable alignment of the same field of view over multiple sessions, BAMBI can reliably track the same neurons over multiple days without requiring retraining. We validated BAMBI using distinct behavioral paradigms in head-fixed and freely moving conditions and demonstrated that mice can reliably modulate specific hippocampal neurons to obtain rewards over multiple days.

Previous Ca²⁺ imaging based BCI applications have been largely limited to head-fixed conditions or lacked in-vivo proof-of-concept and have not demonstrated stable BCI performance across days.

Our results establish Ca²⁺ imaging-based BCI as a tool for longitudinal studies of neuronal population dynamics over extended timescales.