Limited Evidence for Sensory Prediction Error Responses in Visual Cortex of Macaques and Humans

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual Cortex Function

Background:

• Predictive coding theory posits that cortical neurons signal sensory prediction errors.
• These errors represent the difference between top-down predictions and bottom-up sensory input.

Purpose of the Study:

• To investigate the presence of prediction error signals in the visual cortex.
• To test predictions of the computational motif of sensory prediction error signaling.

Main Methods:

• Recorded neural activity (spiking and LFP) in macaque visual cortex (V1, V4).
• Utilized electroencephalography (EEG) in human participants.
• Presented predictable visual stimulus sequences with occasional violations.

Main Results:

• Spiking, LFP, and EEG responses to expected and pattern-violating stimuli were virtually indistinguishable.
• No significant evidence for prediction error responses was detected.

Conclusions:

• The findings challenge the assertion that sensory prediction error signaling is a fundamental computational motif in sensory cortex.
• This challenges predictive coding theory, particularly concerning temporal pattern-based predictions in visual processing.