Visual mismatch responses index surprise signalling but not expectation suppression

Area of Science:

• Cognitive Neuroscience
• Neuroscience
• Psychology

Background:

• Distinguishing expected from surprising stimuli is crucial for learning and behavior.
• Visual mismatch responses (VMRs) in electroencephalography (EEG) are used to study responses to surprising stimuli.
• Existing oddball designs confound stimulus repetition, expectation suppression, and surprise effects.

Purpose of the Study:

• To isolate and quantify the effects of expectation suppression and surprise.
• To adapt an oddball design using Fast Periodic Visual Stimulation (FPVS) to control for stimulus repetition.
• To investigate the nature of VMRs in response to varying probabilities of deviant stimuli.

Main Methods:

• EEG was recorded from 48 participants viewing periodic face stimuli.
• An adapted oddball paradigm presented one of two face identities as every 7th image.
• Stimulus probabilities varied (10-90%) to establish expectations, with 50% probability for neutral control.

Main Results:

• VMRs were found to index surprise responses, showing more negative-going event-related potentials (ERPs) for surprising versus neutral oddballs.
• No significant differences were observed between expected and neutral oddballs, indicating absent expectation suppression effects.
• ERPs for surprising oddballs were consistent across appearance probabilities from 10% to 40%.

Conclusions:

• VMRs, when controlling for repetition, primarily reflect an all-or-none surprise response.
• The findings challenge the notion that VMRs represent a minimization of prediction error associated with expectation suppression.
• Future research should consider VMRs as a direct index of surprise detection in the brain.