Synchronized U-turns of Zebra Fish pair separated by a transparent barrier

chikeung chan^1*

¹Institute of Physics, Academia Sinica, Taipei, Taiwan
²IoP, Academia Sinica, Taipei, Taiwan

* Presenter:chikeung chan, email:ckchan@gate.sinica.edu.tw

Zebrafish (Danio rerio) are social animals that communicate through visual and hydrodynamic interactions. It is well established that animals can anticipate the motion of others in their group; however, the specific dynamics of such interactions in zebrafish remain less understood. In this study, we investigate the visual interactions between two zebrafish separated by a transparent barrier to eliminate hydrodynamic influence. Our findings reveal that the pair exhibits strong interactions when in close proximity. Interestingly, one fish tends to maintain interest and attempts to engage, even when separated by distance, while the other may lose interest as the distance increases. When the pair interacts strongly, their motions become synchronized, particularly during U-turns. This synchronization suggests that the zebrafish retina generates a strong visual signal when a U-turn is detected in the other fish. It is likely that the observing fish utilizes these signals to adjust its movement, resulting in a delayed synchronization rather than anticipatory dynamics. We have also used machine learning to detect the direction of information flow between the two fish. Our findings indicate that zebrafish rely heavily on real-time visual cues rather than predicting future movements of their conspecifics, highlighting the importance of visual processing in their social interactions. These results contribute to our understanding of the neural and sensory basis of social behavior in zebrafish.

Keywords: zebra fish, anticipation, social interaction, information transfer, machine learning