Abstract:The visual cortex predicts incoming sensory stimuli through internal models of the world. Unexpected stimuli that violate these predictions update internal models and drive adaptation. Cortical inhibitory neurons, particularly VIP (vasoactive intestinal peptide) interneurons, are suggested to play a key role in representing unexpected stimuli, given their robust firing following unexpected omissions of familiar images. Importantly, this response is stimulus non-specific, raising an important question about what information it conveys. Given their unique connectivity with other cell types and brain areas, we hypothesized that during unexpected events, VIP neurons encode contextual information, defined as neuronal activity that is not driven by the stimulus itself. To test this hypothesis, we analyzed the Allen Institute Visual Behavior dataset, in which mice viewed repeated familiar images and unexpected omissions of these images while two-photon calcium imaging data from different cell types and visual areas were recorded. Using dimensionality reduction techniques, we found that omissions trigger contextual signaling in VIP neurons in the primary visual cortex (V1) and the lateral medial (LM) visual area, particularly in the superficial layers. Similarly, contextual coding was enhanced in excitatory neurons following omissions. This contrasted sharply with the excitatory response to expected images, during which contextual information was substantially suppressed. Our results suggest that unexpected events activate VIP neurons, which subsequently propagate contextual information across the cortical network. This potentially facilitates the integration of context within the cortical network, and leads to updated predictions about our dynamic environment.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to explore how VIP neurons in the visual cortex encode contextual information in the face of unexpected events. Specifically, the researchers are concerned with whether and how VIP neurons represent non - specific "surprise" signals when mice encounter the unexpected absence of familiar images in visual behavior tasks, and whether such signals contain task - irrelevant contextual information. By analyzing two - photon calcium imaging data in the Allen Institute Visual Behavior dataset, the researchers hope to verify whether VIP neurons are activated when unexpected events occur, and whether these activations are mainly influenced by contextual information rather than being specific to the missing stimulus itself. In addition, the study also explores the different encoding patterns of excitatory neurons for contextual information in the face of expected and unexpected events, and how these patterns affect information processing and adaptive updating in the cerebral cortex network.

Unexpected events modulate context signaling in VIP and excitatory cells of the visual cortex