Table 1 MMN studies of schizophrenia with stimulation beyond duration and frequency deviants

Baldeweg et al. 2002

14 schizophrenia patients and 14 healthy controls

Roving paradigm (changing number of standards between deviants: 2, 6, 18 or 38)

Frequency (12 frequencies from 700 to 1250 Hz) and duration deviants (50 ms)

Attenuation of MMN amplitude over frontal sites but not over mastoids was found in schizophrenia.

Isolated frontal MMN generator impairment in schizophrenia was suggested.

Baldeweg et al. 2004

28 schizophrenia patients; 20 healthy controls

Roving paradigm (changing number of standards between deviants: 2, 4, 8, 18 or 36)

Frequency (12 different frequencies ranging from 700 to 1250 Hz) and duration deviants (50 ms)

“MMN slope” (mean value of a relative gain of MMN amplitudes observed each time the number of standards was increased) was negatively correlated with cognitive tests ratings.

The dysfunction in memory trace formation seem to contribute to attenuation of MMN. For the first time reduction of MMN was associated with worse cognitive performance in schizophrenia.

Baldeweg and Hirsch 2014

49 schizophrenia, 25 bipolar, 15 dementia patients; 49 healthy controls

Roving paradigm (changing number of standards between deviants: 2, 6 or 36)

Frequency (12 different frequencies ranging from 700 to 1250 Hz) and duration deviants (75 ms)

Only for schizophrenia patients a significant attenuation of MMN, when compared to healthy controls was found.

This study provided evidence for specificity of reduction of MMN in schizophrenia.

Fisher et al. 2008b

24 schizophrenia patients and 12 healthy controls

3 different oddballs (stimulation replicated from Kasai et al. (2002))

Tone duration deviant, phenome duration deviant, across-phoneme deviant

Perceived clarity of AVH (auditory verbal hallucinations) correlated with reduction of MMN elicited by phenome paradigms but not pure tone stimulation.

This study opposes general notion that MMN is not influenced by symptoms experienced by patients.

Fisher et al. 2008a

24 schizophrenia patients and 12 healthy controls

Multi-feature paradigm developed by Näätänen et al. (2004)

Duration, frequency, location, gap and intensity deviants

In the hallucinating patients group duration, frequency and intensity MMN amplitude was significantly reduced, when compared to healthy controls and non-hallucinating patients.

It is another study from the same research group providing evidence that forming MMN might be influenced by the individuals clinical status.

Fisher et al. 2012b

12 schizophrenia patients and 15 healthy controls

Multi-feature paradigm developed by Näätänen et al. (2004)

Duration, frequency, location, gap and intensity deviants

Correlations of AVH features measured with PSYRATS (Psychotic Symptom Rating Scale) scale with attenuated gap and location MMN and negative correlation of chlorpromazine equivalents and location MMN amplitude.

Applying deviants other than frequency and duration was informative for analysing auditory stimuli processing in schizophrenia.

Fisher et al. 2012a

12 schizophrenia patients

Optimized multi-feature paradigm developed by Fisher et al. (2011b

Duration, frequency and intensity deviants

An administration of a nicotine significantly shortened latency of intensity MMN. Changes in intensity MMN were correlated with increased AVH clarity.

Stimulating nicotine action seems to have more significant impact on intensity than on duration or frequency MMN.

Fisher et al. 2011a

12 schizophrenia patients and 12 healthy controls

Optimized multi-feature paradigm developed by Fisher et al. (2011b)

Duration, frequency and intensity deviants

Reduction of duration and intensity MMN amplitude correlated with positive symptom ratings. Attenuation of duration MMN was more significant in patients with more severe AVH.

This study extends research on neurophysiological background of positive symptoms in schizophrenia.

Fisher et al. 2014

10 schizophrenia patients and 13 healthy controls

Oddball

Pure tone frequency deviant and environmental sound deviant (baby cry)

Amplitude and latency of MMN to novel environmental deviant was significantly reduced in schizophrenia. Amplitude of MMN was positively correlated with positive symptoms and AVH ratings.

This study explores applicability of a novel stimulation in eliciting ERP’s in schizophrenia.

Friedman et al. 2012

26 schizophrenia patients and 19 healthy controls

Modified oddball; all stimuli were harmonics (500 Hz, 1000 Hz and 1500 Hz for standards)

Intensity, frequency and duration deviants

Poorer premorbid educational status, longer illness duration and worse scoring on the independent living scale correlated with reduced duration and intensity MMN.

Results of the study confirm importance for evaluation of different deviant features in MMN research in schizophrenia.

Gjini et al. 2010

12 schizophrenia patients and 12 healthy controls

Oddball with three deviants and oddball with single abstract deviant

Frequency, duration, white-noise and abstract feature deviants (standards were paired tones of ascending frequency and deviants paired tones of descending frequency)

Negative correlation between P50 gating and abstract MMN amplitude was weaker in schizophrenia patients than in healthy controls. Only white-noise MMN was significantly reduced in the schizophrenia group.

Introducing novel deviants enabled more insight into auditory information processing in schizophrenia. No significant results were found for frequency and duration MMN.

(Kawakubo et al. 2007)

13 schizophrenia patients

3 different oddballs (stimulation replicated from Kasai et al. (2002))

Tone duration deviant, phenome duration deviant, across-phoneme deviant

Significant correlation was found between larger MMN to across-phoneme deviant and better improvement in social scores after 3 month social skills training.

Phoneme oddball was better than tone oddball in predicting improvement in social skills.

Kreitschmann-Andermahr et al. 1999

15 schizophrenia patients and 16 healthy controls

Oddball

Two different frequency deviants (1050 Hz and 5000 Hz) and tone omission deviant

A reduction of MMNm (MMN magnetic counterpart) for all deviants show dysfunction of temporal MMN generators.

The reduction of an omission MMNm suggests that an error prediction impairment contributes to schizophrenia deficits.

Matthews et al. 2007

18 schizophrenia patients and 19 healthy controls

3 oddballs (each with different deviant type)

Three types of deviants with delayed arrival-time or different phase or loudness of a stimuli for the left and right ear

Schizophrenia group showed dysfunction in processing spatial (arrival-time or phase difference) but not loudness deviants when compared to healthy controls.

The results suggest specific impairment in exact discrimination of time cues in schizophrenia.

Rissling et al. 2013

20 schizophrenia patients and 20 healthy controls

Oddball

Duration and white-noise deviant

Attentional demand (both auditory and visual) modulated white-noise MMN and P3a in schizophrenia patients and healthy controls. Attentional modulation of auditory change detection seem to be intact in schizophrenia.

MMN to white-noise deviant seems to be more sensitive to changes in attention load and direction (visual vs auditory) than duration MMN.

Thönnessen et al. 2008

12 schizophrenia patients and 12 healthy controls

Pure-tone oddball and multi-feature paradigm designed by Näätänen et al. (2004)

Duration and intensity deviants in the oddball; duration, frequency, intensity, gap and location deviants in the multi-feature paradigm;

The MMNm was significantly reduced in the schizophrenia group over left auditory cortex for all deviants (only gap and location over right auditory cortex) presented in multi-feature paradigm. No significant difference between groups were found for oddball.

This study suggest that MMNm elicited by multi-feature is more accurate than oddball paradigm in identifying deficits in forming MMN in schizophrenia. For both paradigms, no significant reduction of electrical MMN in schizophrenia was observed.

Todd et al. 2008

43 schizophrenia patients and 42 healthy controls

Oddball

Intensity, frequency and duration

A reduction in duration and intensity MMN in short-illness patients with schizophrenia. In a long-illness group only frequency MMN were significantly reduced.

This study revealed potential for more variable stimulation to create patterns for some groups of patients and healthy controls.

Todd et al. 2014

33 schizophrenia patients and 58 healthy controls

Oddball composed of two stimuli sequences one with a pseudorandom and one with fixed order of stimuli

Duration, frequency, intensity and ascending frequency deviant; in fixed order sequence all frequency deviants were preceded with duration and all ascending frequency preceded intensity ones.

Intensity, duration and ascending frequency but not frequency MMN were smaller in schizophrenia. When the deviants were paired MMN for the second deviant was equally reduced in schizophrenia patients and healthy controls.

Method used for a computation of MMN (traditional vs feature-controlled) significantly influenced results. Introduced oddball design failed to reveal additional impairment in stimuli processing in schizophrenia.