2:00–3:00 pm KPTC 206
Xiaowen Chen, École Normale Supérieure
Collective behavior from social interactions to population coding
Collective behavior is ubiquitous in living systems. Interacting animals form social networks, and interacting neurons give rise to cognitive activities. Despite their difference, both social interactions and collective neuronal activities deal with behavior encoding collectively and dynamically. In this talk, I will discuss two examples where we learn directly from data the interaction rules. In the first example, using statistical physics models, we unravel the interaction structures among a group of social mice from their co-localization patterns housed in semi-naturalistic environments. To capture both the equal-time correlation and the long-tailed waiting time distribution in the data, we developed a novel inference method termed the generalized Glauber dynamics that can tune the dynamics while keeping the steady state distribution fixed. The inferred interaction strength can characterize sociability for different mice strains. In the second example, we studied information flow among neurons in the larval zebrafish hindbrain. By adapting the method of Granger causality to single cell calcium transient data, we were able to detect both a global information flow from sensory to behavior, as well as how the Mesencephalic Locomotor Region is recruited in locomotion. Towards the end, I will briefly discuss future directions bridging the social and the neuronal scales through model-predicted perturbation.
