Event:

Activity in the primary visual cortex is influenced not only by sensory stimuli but also by the animal’s behavioural state. In mice, it is well established that sensory response gain is systematically modulated by locomotion speed. However, the intricacies of the interactions between locomotion and sensory coding are not yet fully understood. In this thesis project, we systematically characterised the influence of sensory input and behavioural state on neural activity in the primary visual cortex (V1). To this end, we analysed calcium imaging data from the Allen Brain Observatory. Focusing on the relationship between visual stimulus direction and running speed, we developed a statistical model to capture the joint orientation-velocity tuning of neurons. The model decomposes tuning preferences into direction-selective subunits that are independently modulated by velocity. Our analysis revealed a rich diversity of behavioural modulations, including non-monotonic gain changes as well as more complex effects, such as shifts in preferred stimulus direction and transitions between orientation and direction selectivity as a function of running speed. An information-theoretic analysis reveals that visual-stimulus-selective V1 neurons consistently convey information about locomotion speed. Our observations suggest that V1 not only encodes sensory input modulated by behavioural variables but may also instantiate a joint sensory-behavioural representation.