Computation of abstract context in the midbrain reticular nucleus during perceptual decision-making

Perceptual decision-making is crucial for survival, requiring the integration of dynamic sensory evidence from the environment with internal state variables (e.g. context, priors, value) to choose an appropriate action plan. Neural correlates of perceptual decisions are more distributed than once thought, having been found within the cortex, striatum, and superior colliculus (SC). Recent work has suggested a potentially novel node in this network - the midbrain reticular nucleus (MRN) - which was one of the few regions that encoded visual and action selection signals in a brain-wide survey of 2-alternative forced choice (2AFC) task-related activity in mice. To characterize the functional properties of MRN, we have conducted extensive recordings throughout MRN while both task-naive and trained mice passively view a variety of task stimuli. Recordings performed thus far have revealed topographic organization of sensory/motor coding in MRN and increased visual responses after task training. Additionally, we sought to assess whether the MRN, in addition to containing action selection signals, contains internal state signals required to compute the abstract decision. We have successfully designed and trained a novel reverse contingency task for mice wherein the computation of context can be measured. Identical visual stimuli require opposing motor reports for reward depending on the task block. In preliminary Neuropixels 2.0 recordings during task performance, we found contextual modulation of visual and action-related responses in canonical decision-making regions including secondary motor cortex (MOs), caudoputamen, and SC as well as in the MRN. Analyzing the population activity structure, we find evidence for a computation in which context alters the integration of evidence by shifting the attractor landscape. Together, these findings suggest that the MRN may be a novel node in the distributed network underlying the computation of abstract contextual information related to perceptual decisions.