Characterize neuronal responses to natural movies in the mouse superior colliculus

While artificial stimuli have been widely used in visual neuroscience and have significantly advanced our understanding of visual processing, they differ dramatically from the natural scenes that animals encounter in the wild. How natural stimuli are encoded in the superior colliculus (SC) and how neuronal responses to artificial and natural stimuli are related remain poorly understood. Here I applied two-photon calcium imaging to record neuronal activity in the mouse superficial SC in response to natural movies. An unsupervised learning algorithm grouped recorded neurons into 16 clusters based on their response patterns. Each cluster exhibited distinct temporal profiles, which arose from differences in both receptive field coverage and how neurons encode local visual features. Interestingly, I found a strong correlation between neuronal responses to natural movies and functional properties previously characterized using artificial stimuli. This suggests that the SC maintains a stable neural representation of visual information that is largely independent of the types of visual stimuli. Furthermore, neuronal responses to natural movies varied with depth within the superficial SC and across genetically defined neuronal types. These findings bridge the gap between our understanding of responses to artificial and natural stimuli, providing new insights into visual processing in the SC.