If you sit in a quiet room with a regular series of identical tones — beep, beep, beep — and an occasional slightly different tone is slipped in, your brain notices, automatically, even if you are not paying attention. About 100 to 250 milliseconds after the unexpected sound, an electrical response appears over your frontal scalp called mismatch negativity, or MMN. It happens whether you are reading, watching TV, or asleep.
MMN has become one of the most studied physiological measurements in schizophrenia research. It is consistently reduced in people with the disorder, present at illness onset, and possibly even before — a rare degree of replicability for a psychiatric biomarker.
Mismatch negativity is a brain response to unexpected sounds that is reliably reduced in schizophrenia, may reflect NMDA receptor function, and is one of the leading candidate biomarkers in the field.
How MMN is measured
MMN is recorded with electroencephalography (EEG) — small electrodes placed on the scalp that pick up the summed electrical activity of cortical neurons. The classic experiment is called an oddball paradigm: subjects listen to a long series of standard tones, with rare deviant tones (different in pitch, duration, or intensity) inserted unpredictably. The brain's response to the standards is subtracted from its response to the deviants, leaving the MMN waveform.
The peak appears around 150–200 ms after the unexpected sound and is largest over the frontal-central scalp. It is generated in the auditory cortex and supplemented by a frontal contribution that may signal a "switch attention" component.
What MMN reflects
MMN is widely understood as the electrical signature of an automatic process called predictive coding. The brain continually builds a model of what should come next; when reality violates that model, an error signal is generated. MMN is essentially that error signal made visible.
Two reasons this matters for schizophrenia:
- Many influential theories propose that schizophrenia involves a failure of predictive coding — the brain becomes less able to distinguish what it expects from what is actually arriving.
- MMN generation depends critically on NMDA receptor function. NMDA receptor blockade (with ketamine) reduces MMN amplitude in healthy volunteers in patterns that resemble what is seen in schizophrenia, supporting the link between MMN deficits and the NMDA hypofunction hypothesis.
What the schizophrenia research shows
- MMN amplitude is consistently reduced in chronic schizophrenia, with effect sizes among the largest in psychiatric biomarker research.
- The reduction is present in first-episode psychosis, although somewhat smaller than in chronic illness.
- It is also present, in attenuated form, in people at clinical high risk for psychosis, and may help predict who will transition to a full psychotic disorder.
- MMN reduction correlates with cognitive impairment and with poorer day-to-day functioning, suggesting it taps something behaviourally meaningful.
- MMN deficits are present in unmedicated patients, ruling out antipsychotics as the cause.
The duration-deviant MMN (using deviants that differ in tone length) tends to be the most robust marker; pitch-deviant MMN is more variable.
MMN as a biomarker — what it could be used for
- Risk stratification. In clinical-high-risk populations, MMN may identify those most likely to progress to psychosis.
- Outcome prediction. MMN amplitude correlates with functional outcome in early psychosis and may inform prognosis.
- Drug development biomarker. Because MMN is reliably altered by NMDA modulation, it can serve as a "go/no-go" test for whether a candidate drug is engaging the right system.
- Translational bridge. MMN can be measured in similar form in animals, healthy humans, high-risk groups, and patients, allowing findings to be tested across these populations.
What it is not (yet) used for
MMN is not a clinical diagnostic test. The reasons:
- While the average difference between groups is large, there is significant overlap between individuals with schizophrenia and healthy controls — MMN cannot reliably classify a single person.
- MMN is reduced in other conditions too, including some forms of dementia and other psychiatric disorders.
- EEG measurement of MMN requires careful equipment and analysis; results vary across labs and paradigms.
- No standardised normative reference values exist for routine clinical use.
For now, MMN sits where many useful biomarkers in psychiatry sit: very useful for research, not yet ready for individual diagnosis or treatment selection.
The bigger picture
MMN is one of a small number of endophenotypes in schizophrenia — measurable biological traits that lie between genes and overt symptoms. Other candidates in the same family include P50 sensory gating, prepulse inhibition, antisaccade eye movements, and certain working memory tasks. Each is reliably altered in schizophrenia, present in some unaffected first-degree relatives, and tied to specific neurobiological systems.
None of them has yet become a routine clinical test. But together they are reshaping how researchers think about schizophrenia: not as a single disease with a single cause, but as a constellation of measurable circuit-level abnormalities, of which MMN is one of the most reliable.
What this means for patients
Practically, MMN does not affect anyone's treatment plan today. Its real importance is conceptual. It demonstrates that schizophrenia produces measurable, replicable changes in basic brain function — quietly, before anyone speaks a word, in response to a sound the listener may not consciously notice. That fact alone, when it sinks in, helps push back against the lingering stigma that schizophrenia is a failure of will or character. It is a brain condition, and brains leave traces.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.