Feature selectivity can explain mismatch signals in mouse visual cortex.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Sensory Processing

Background:

• Sensory experiences are influenced by self-generated actions and self-motion.
• Predictive coding theories suggest actions are regulated by prediction errors (sensory experience vs. expectation).
• Previous studies linked prediction error signals in mouse visual cortex (V1) to self-motion perturbations.

Purpose of the Study:

• To investigate if prediction error signals in V1 can be triggered by visual stimuli independent of self-motion.
• To explore the role of motor and sensory signals in generating these neural responses.

Main Methods:

• Recorded activity of V1 neurons in mice.
• Presented drifting gratings that unexpectedly stopped, both during running and at rest.
• Analyzed neuronal responses to these visual perturbations.

Main Results:

• Visual perturbations elicited strong responses in V1 neurons, even when uncoupled from running.
• These responses were enhanced during running.
• Responses were strongest at a neuron's preferred orientation and more prevalent in neurons preferring slow visual speeds.

Conclusions:

• Prediction error signals in V1 can be explained by the convergence of established motor and sensory signals.
• This provides a purely sensory and motor-based explanation for observed neural mismatch signals.
• The findings refine our understanding of predictive coding mechanisms in sensory processing.