Adaptive integration of self-motion and goals in posterior parietal cortex

Abstract

Animals engage in a variety of navigational behaviors that require different regimes of behavioral control. In the wild, rats readily switch between foraging and more complex behaviors such as chase, wherein they pursue other rats or small prey. These tasks require vastly different tracking of multiple behaviorally-significant variables including self-motion state. It is unknown whether changes in navigational context flexibly modulate the encoding of these variables. To explore this possibility, we compared self-motion processing in the multisensory posterior parietal cortex while rats performed alternating blocks of free foraging and visual target pursuit. Animals performed the pursuit task and demonstrated predictive processing by anticipating target trajectories and intercepting them. Relative to free exploration, pursuit sessions yielded greater proportions of parietal cortex neurons with reliable sensitivity to self-motion. Multiplicative gain modulation was observed during pursuit which increased the dynamic range of tuning and led to enhanced decoding accuracy of self-motion state. We found that self-motion sensitivity in parietal cortex was history-dependent regardless of behavioral context but that the temporal window of self-motion tracking was extended during target pursuit. Finally, many self-motion sensitive neurons conjunctively tracked the position of the visual target relative to the animal in egocentric coordinates, thus providing a potential coding mechanism for the observed gain changes to self-motion signals. We conclude that posterior parietal cortex dynamically integrates behaviorally-relevant information in response to ongoing task demands.