Neural Mechanisms of Behavioral Variability and Strategy Selection in Larval Zebrafish - Abstract Survival in dynamic environments requires the ability to switch between exploiting an established action and exploring potentially better alternatives. This strategy switch consists of multiple processes: one that recognizes when the current behavior is no longer successful and abandons it in favor of exploration in action space, and another that promotes the maintenance of successful behavioral programs. A critical component of action exploration is behavioral variability. While broad brain regions have been identified as being involved in (1) the behavioral variability component, for example the cortical region LMAN in juvenile songbirds, and (2) the exploratory versus exploitatory strategy tradeoff, for example the midbrain dopaminergic system and the locus coeruleus noradrenergic network in rodents, the generative algorithms and circuit-level implementation of computations underlying the control of behavioral variability and the switch to exploratory behavior are not known. Identifying such motifs in mammals and birds is challenging because of their behavioral complexity and their large brain size. Larval zebrafish are an ideal system for developing mechanistic understanding of behavioral variability and exploration versus exploitation of motor programs because of the simplicity of their behavioral repertoire and because they are small, transparent, and genetically tractable. Therefore, I will use larval zebrafish, to understand the neural mechanisms underlying behavioral variability and the exploration versus exploitation of motor programs. In Aim 1, I will optimize an existing virtual reality behavioral paradigm for larval zebrafish based on the optomotor response to trigger exploration of motor strategies. In Aim 2, I will design an experimental protocol to induce larval zebrafish to switch from an exploratory to an exploitatory strategy. Detailed analyses of swim bout kinematics in combination with whole-brain functional imaging during both experimental preparations will enable the characterization of neural circuits responsible for increased behavioral variability in exploration and the exploratory versus exploitatory tradeoff. Completion of this project will yield a comprehensive and quantitative model of the neural basis of the complex task of action selection in different contexts. Because identification of optimal motor strategies requires exploration of any organism’s motor repertoire, these principles may suggest generalizable motifs across the animal kingdom.
