Estimating Fluctuations in Neural Representations of Uncertain Environments

Neural Coding analyses often reflect an assumption that neural populations respond and consistently to particular stimuli. For example, analyses of spatial in hippocampal populations often assume that each environment has unique representation and that remapping occurs over long time scales as animal traverses between distinct environments. However, as neuroscience begin to explore more naturalistic tasks and stimuli, and reflect more in neural representations, methods for analyzing population neural codes adapt to reflect these features. In this paper, we develop a new state-space framework to address two important issues related to remapping. First, may exhibit significant trial-to-trial or moment-to-moment variability in firing patterns used to represent a particular environment or stimulus. Second, ambiguous environments and tasks that involve cognitive uncertainty, neural may rapidly fluctuate between multiple representations. The statespace addresses these two issues by integrating an observation model, which for multiple representations of the same stimulus or environment, with a model, which characterizes the moment-by-moment probability of a shift the neural representation. These models allow us to compute instantaneous of the stimulus or environment currently represented by the population. demonstrate the application of this approach to the analysis of population in the CA1 region of hippocampus of a mouse moving through ambiguous environments. Our analyses demonstrate that many hippocampal cells significant trial-to-trial variability in their representations and that the representation can fluctuate rapidly between environments within a trial when spatial cues are most ambiguous.