Head direction and orienting-suppression signals in primary visual cortex of freely-moving rats


February 13, 2019 - 12:00pm - 1:15pm
Northwest Building, Room 243
About the Speaker
Greg Guitchounts
Speaker Title: 
Graduate Student
Speaker Affiliation: 
Cox Lab

Recent studies have brought into question the idea that information in sensory cortex is only sensory. Signals related to locomotion or navigation have been observed in several sensory cortical brain regions. Neurons in primary visual cortex (V1), for example, modulate their firing rates based on the animal’s running speed, even in the absence of visual input. V1 activity has recently also been shown to be modulated by perceived position in an environment and passive head turning. Most experiments of this nature have been performed in head-fixed mice whose range of motion is limited, leaving open the question of how V1 is modulated by the variety of movements free animals make. We recorded neuronal activity continuously (24/7) for up to three weeks in rat V1 using tetrodes while the animals lived in their home cages. Coupling the recordings with measurements of behavior using a head-mounted accelerometer/gyroscope/magnetometer allowed us to probe the relationship between V1 activity and natural movement, while the continuous nature of the recordings allowed us to examine differences in activity at the two phases of the light cycle (light and dark). We find that V1 activity linearly encodes the animal’s 3D head direction (yaw, roll, and pitch), even in complete darkness. Further, V1 dynamics are strongly modulated by turning (left, right, up, down, clockwise, counterclockwise). While activity is generally suppressed during turns in the dark, it is increased at the end of turns in the light. A linear classifier can distinguish different turn types. Bilateral lesions of secondary motor cortex (M2), a major source of cortical feedback to V1, greatly diminish turn-related activity, both in the light and in the dark, while preserving responses to visual stimulation and head direction encoding, suggesting that the latter signals originate elsewhere (e.g. retrosplenial cortex, another major input to V1). These results support predictive coding theories of cortical function and suggest that corollary discharge signals in mammalian brains contain detailed movement and head direction information.