Abstract

Abstract: Our brain needs to selectively process that sensory information that is most relevant for decision-making, given the constant bombardment of data on our senses and the limited capacity of the brain. However, the neural mechanisms that underlie this selection are not well understood. This talk will discuss how the mouse can be used as a model to study the selection of visual information for decision-making. Mice can learn within a few days a visual discrimination task, enabling us to study responses to the same visual features in neural circuits before and after learning. Mice also learn to flexibly switch between visual and olfactory discrimination tasks (while viewing the same visual features) within seconds, enabling us to compare neural responses when mice attend or ignore visual features. Long-term 2‐photon calcium imaging in mice during learning and attention reveals that already in the primary visual cortex (V1), neural responses become increasingly selective for relevant visual features after learning, and that responses are more selective when animals attend visual features compared to when they ignore visual features. The effects of learning and attentional switching on the response selectivity of the same cells are largely independent, and lead to a distinct profile of changes in the selectivity and co‐activation patterns across multiple cell classes, including both excitatory and inhibitory interneurons. These experimental results, and computational modelling of these effects, indicate there are separate mechanisms underlying the increased discriminability of relevant sensory stimuli across longer and shorter time scales to help improve visually-guided decisions.

Bio: Jasper Poort is a Wellcome Trust Sir Henry Dale Fellow and group leader in the Selective Vision Lab (www.psychol.cam.ac.uk/svl/) at the Department of Psychology in Cambridge, where he works on understanding the cortical mechanisms of visual learning and attention.