A shared multi-feature population code for sensory reliability across mouse visual cortex.

Sensory reliability in mouse visual cortex is encoded by coherence-dependent coordinated changes in response amplitude, selectivity, and signal-to-noise ratio across over 150,000 neurons.

• Why it matters: It is unclear how neural populations represent reliability and whether the same code is shared across cortical areas during sensory uncertainty.
• What they did: Using two-photon calcium imaging in mouse visual cortices, the authors measured population responses to coherent motion stimuli and tested stimulus-based normalization models against observed coherence-dependent patterns.
• The result: The patterns are not fully explained by normalization models, track the distribution of motion energy across directions while staying relatively invariant to total motion energy, and are consistent across visual areas, supporting a shared encoding scheme.

AI summary of the abstract · may contain errors · not medical advice · read the paper.