Theory-driven approaches to identify minimal circuits for susceptibility and resilience

Individuals vary widely in the types of behavioral transitions they undergo in response to adverse environmental conditions. While some individuals respond to adversity by developing persistent negative behavioral states such as hopelessness or anxiety, others are more resilient. Individual differences in neural dynamics likely underlie the likelihood of these transitions, yet we lack a unified framework to detect and quantify them. This is a challenging problem, since networks that encode these transitions are broadly distributed throughout the brain, and changes may unfold across long periods of time. Using a principled theoretical approach (Rajan lab), combined with neural recording performed at multiple temporal and spatial scales (Falkner and Witten labs), we hope to identify multi-region circuit motifs that predict susceptibility and resilience, and to detect when changes in these networks bias individuals toward specific behavioral states.

We will achieve this by developing a new class of flexible and robust ‘networks of networks’: multi-region recurrent neural network (RNN) models constrained directly by both neural and behavioral data. Neural data will be collected using a novel ‘macrocircuit’ recording approach, which uses longitudinal multisite photometry to capture networkwide dynamics, and from targeted recordings with cellular resolution in key areas, as animals undergo a multiday chronic defeat paradigm. In addition, we will collect behavioral data from freely moving animals tracked by pose detection to serve as readouts for the RNN and to correlate with dynamical motifs found in the neural data. To identify minimal circuits for susceptibility and resilience, we will first take a networkwide approach to modeling macrocircuit data that are spatially sparse but temporally extended. Next, we will validate and further constrain these minimal circuit models by recording at single-cell resolution in key areas. Finally, we will use the model to reveal bifurcations toward susceptibility and resilience, and compare this to experimental outcomes after a second adverse environmental challenge.

This collaborative project leverages significant theoretical and technical expertise from the Rajan, Falkner and Witten labs. The Rajan lab is a pioneer in the development of multi-region RNNs and has successfully applied these tools to discover behavioral transitions in the dynamics of larval zebrafish. To apply these computational tools to individuals with a richer behavioral repertoire, the Falkner and Witten labs will perform multi-region (Falkner) and cellular-resolution (Witten) recordings in tandem with behavioral quantification. Together, these experiments will reveal foundational principles about networkwide communication, and how changing network dynamics predict susceptibility and resilience.