Sensory Plasticity in the Auditory Striatum as an Impetus for Action Control

Project Summary The ability to translate sensory experiences into action is essential for our survival. Despite its importance in health and disease, we know remarkably little about how we assign meaning to and use sensory stimuli to guide behavior. The dorsal striatum is thought to be particularly important for the formation of sensorimotor associations during reinforcement learning due to the dopaminergic inputs it receives, as well as a diverse array of cortical and subcortical inputs. There are two cell types that make up the two output pathways of the dorsal striatum, direct pathway striatal projection neurons (dSPNs) and indirect pathway striatal neurons (iSPNs). While much work has focused on how these pathways might function to initiate movements, very little is known about how sensory learning influences the neuronal activity of these neurons and what effect this has on behavior. The experiments that make up this proposal provide a framework for understanding how sensory input shapes the activity of dSPNs and iSPNs in the dorsal striatum. In this proposal, we focus on a specific part of the dorsal striatum known as the auditory striatum (AudStr), that receives dense auditory inputs. We hypothesize that auditory sensorimotor learning enables the formation of cue-specific ensembles that correlate with and predict motor output. We expect that these changes will be primarily driven by synaptic plasticity of inputs that converge onto SPNs, rather than changes to their intrinsic excitability. We will perform two independent, inter-related aims to test this hypothesis. We will train mice on a task that requires them to associate a 'go' cue with a specified action to receive a reward, and to suppress this action in response to a 'no-go' cue. In Aim 1, we will employ longitudinal calcium imaging of AudStr neurons to characterize the outputs of these neurons to cues before and after learning. In Aim 2, we will explore the cellular mechanisms underlying anticipated changes in population activity that results from learning, and aim to demonstrate the importance of synaptic plasticity in the AudStr to this process. In both aims we will employ methods that enable us to identify neurons as dSPNs and iSPNs. This is crucial because a major outstanding question in this field is whether these cell types play opposing or complementary roles in producing appropriate motor responses. Overall, this work will lay the groundwork for a new conceptual model for understanding how sensory learning influences striatal activity to control behavior.