Attention and mismatch negativity

The mismatch negativity (MMN) component of the auditory event-related potential (ERP) is elicited by infrequent, physically deviant stimuli in a sequence of frequent homogeneous stimuli (standards). It has been suggested that the MMN is generated by an automatic (attention-independent) neural mismatch process with a memory trace that encodes the physical features of the standard stimulus. The proposed MMN independence of attention was addressed in the present study. Standard stimuli and two types of deviant stimuli, differing from standards either in frequency or intensity, were dichotically presented in random order and at a very rapid rate. The subject attended either to left-or right-ear stimuli, counting the number of a designated type of deviants in that ear. In the present conditions of very strongly focused attention, the MMN was elicited even by frequency change in the ignored input stream, and its amplitude was very similar to that of the MMN elicited by equivalent deviant stimuli (targets) in the attended input stream. In contrast, the MMN to intensity deviation was clearly attenuated in the absence of attention. This effect is, however, probably due to the attention effect on the MMN generator itself rather than the antecedent sensory-analysis and -storing functions.     

Long-term habituation of the speech-elicited mismatch negativity

A significant issue in the use of the mismatch negativity evoked potential (MMN) concerns its low signal-to-noise ratio (SNR). One can improve the noise level by increasing the number of samples included in the averaged response. However, improvement achieved in this way assumes that the signal, the MMN, remains stable for extended test times, an assumption which has not been tested. If the MMN is not stable, or exhibits habituation over the test session, then SNR would be adversely affected. MMN response magnitude was measured in 5-min intervals over the course of a test session in response to various speech syllable contrasts. Significant long-term habituation of MMN was observed for all three subject populations tested: young adults, school-age children, and guinea pigs. The time course of the habituation and the stimulus conditions under which it occurs have important implications for research and clinical applications of the MMN. Recording procedures that minimize habituation effects may be used to advantage to improve the signal-to-noise ratio of the MMN.     

Effects of auditory hallucinations on the mismatch negativity (MMN) in schizophrenia as measured by a modified ‘optimal’ multi-feature paradigm

The recently developed Optimal-3 multi-feature MMN paradigm, a shortened version of the ‘optimal’ multi-feature MMN paradigm, allows for the focused recording of the most widely reported MMN deviants (frequency, duration, intensity) within an efficient and time-saving paradigm. The objective of this study was to examine MMN acoustic change detection in schizophrenia (SZ), and elucidate its association with auditory verbal hallucinations (AH), using the Optimal-3.

Methods

MMN to duration, frequency and intensity deviants were recorded in 12 SZ outpatients (SZs) with persistent AHs and 12 matched healthy controls (HC). Electrical activity was recorded from 32 scalp electrodes; MMN amplitudes and latencies for each deviant were compared between groups and were correlated with trait (PSYRATS) and state measures of AH severity and Positive and Negative Syndrome Scale (PANSS) ratings in SZs.

Results

SZs showed a significantly smaller duration MMN compared to HCs. Furthermore, in SZs attenuated duration MMN amplitudes were correlated with increased PSYRATS scores, as well as increased PANSS positive symptom, hallucination item and general psychoticism ratings, while attenuated intensity MMN amplitudes were correlated with increased PSYRATS scores.

Conclusions

This is the first study to examine MMN in SZ within the modified (Optimal-3) multi-feature MMN paradigm. This study corroborates previous research reporting a robust duration MMN deficit in SZ and supports previous findings suggesting that AHs may contribute to MMN deficits in SZ.     

Memory-based or afferent processes in mismatch negativity (MMN): a review of the evidence

The mismatch negativity (MMN) is an electromagnetic response to any discriminable change in regular auditory input. This response is usually interpreted as being generated by an automatic cortical change‐detection process in which a difference is found between the current input and the representation of the regular aspects of the preceding auditory input. Recently, this interpretation was questioned by Jääskeläinen et al. (2004) who proposed that the MMN is a product of an N1 (N1a) difference wave emerging in the subtraction procedure used to visualize and quantify the MMN. We now evaluate this “adaptation hypothesis” of the MMN in the light of the available data. It is shown that the MMN cannot be accounted for by differential activation of the afferent N1 transient detectors by repetitive (“standard”) stimuli and deviant (“novel”) stimuli and that the presence of a memory representation of the standard is required for the elicitation of MMN.     

Visual mismatch negativity elicited by facial expressions under non-attentional condition

A modified "cross-modal delayed response" paradigm was used to investigate putative processing of face expression in the absence of focused attention to the face. Neutral, happy and sad faces were presented during intervals occurring between a tone and a response imperative signal (a faint click), while subjects were instructed to discriminate the location of the tone as quickly and accurately as possible and to ignore the faces. A neutral face was presented in 80% of the trials whereas the happy and sad faces were presented in the remaining trials - 10%, respectively. Expression mismatch negativity (EMMN) was obtained by subtracting the ERP elicited by neutral faces from that elicited by sad faces or happy faces. The EMMN started from around 120 ms (sad) and 110 ms (happy) lasting up to 430 ms (sad) and 360 ms (happy) post-stimulus. The EMMN elicited by sad faces was more negative than that elicited by happy faces. Both EMMNs distributed over posterior areas and covered larger areas in the right than in the left hemisphere sites (especially for happy EMMN).     

Variable cerebral responses to equally distinct deviance in four auditory dimensions: A mismatch negativity study

In this study we compared the amplitude, latency, and spatial distribution of the mismatch negativity (MMN), elicited by tones deviating in either frequency, intensity, stimulus-onset asynchrony (SOA) or location, and assessed inter- and intra-subject variability. Cross-dimensional comparisons were facilitated by adjusting the magnitude of deviance individually, so that the detection rates were similar, both across dimensions (within subject) and across subjects (within dimension). Despite similar detection rates, the MMN elicited by frequency deviance was larger, and the MMN elicited by SOA deviance was earlier than the other types of MMN. The reliability of frequency-related MMN was better than that of the other types of MMN. The results highlight the problems of comparing MMN across dimensions, especially in clinical populations.     

Event-Related Brain Potentials Reflecting Processing of Relevant and Irrelevant Stimuli During Selective Listening

Event-related brain potentials were recorded from the human scalp during selective listening to tone pips differing in location and/or pitch from irrelevant tones. The subjects' task was to discriminate infrequent deviant tones of lower intensity appearing among designated (relevant) tones. A large processing negativity was observed in the event-related potentials to relevant tones differing from the irrelevant tones in location even when both tones randomly varied in pitch. Similarly, a large processing negativity was elicited by the relevant tones differing from the irrelevant tones in pitch even when the location of both tones varied randomly. The results support the theory that the processing negativity to relevant stimuli reflects a match of these stimuli with an “attentional trace,' an actively maintained neuronal representation of the physical feature(s) of relevant stimuli that distinguish these stimuli from the irrelevant stimuli. Furthermore, the infrequent lower-intensity tones appearing among irrelevant tones elicited a mismatch negativity similar to the mismatch negativity elicited by target tones, equivalent lower-intensity tones appearing among relevant tones. This indicates that these infrequent stimulus changes were automatically discriminated by the generator mechanism associated with mismatch negativity.     

Comparison between mismatch negativity amplitude and magnetic mismatch field strength in normal adults

The auditory mismatch negativity (MMN) or its magnetic counterpart (magnetic mismatch field, MMF) has been widely used to assess the ability of stimulus-driven change detection process in humans. The authors evaluated the similarity of inter-individual variation of the response strength between MMN and MMF recordings. Three types of MMN or MMF were recorded in ten healthy subjects: change in duration of pure-tone stimuli, change in duration of the Japanese vowel /a/, and difference between the Japanese vowels /a/ and /o/. There was no significant correlation between MMN amplitude and MMF strength under any condition and in either hemisphere. These results suggest that widely used indices of MMN in the two technologies, i.e., EEG-amplitude and MEG-ECD may not be proportional in an individual. To further clarify the differential significance of recording MMN/MMF may be important to establish MMN/MMF as clinical indices of individual ability of preattentive stage of auditory processing.     

Impulsivity affects mismatch negativity (MMN) measures of preattentive auditory processing

The current study addresses the relation between impulsive personality and mismatch negativity (MMN) as measure of auditory processing. In a sample of 33 normals, MMN was measured and related to self-reported measures of impulsive behavior. Participants were tested using a passive auditory oddball paradigm while EEG recordings were obtained. It was found that the negative MMN amplitude at the left lateral cluster was negatively correlated with self-reported impulsivity, with high impulsive individuals showing larger MMN amplitudes than low impulsive individuals. This indicates that impulsivity is related to preattentive auditory processing, a process that is not dependent on response execution. In addition to the well-known involvement of top–down control in impulsivity, these findings implicate that the cognitive bottom–up processing of incoming stimuli is associated with individual differences in impulsivity. Since the source of the MMN is thought to be located in the auditory cortex, this suggests involvement of “lower order” temporal lobe processes in impulsive behaviors.     

Preattentive processing of spectral, temporal, and structural characteristics of acoustic regularities: a mismatch negativity study

In the present study we investigated the relationship between detecting violations of structural regularities of tone sequences and detecting deviations from temporal regularities or repetitive spectral auditory stimulus features. Twelve subjects were presented with randomized sequences of two tones (differing both in frequency and intensity) delivered alternately to the left and right ears at a constant stimulus onset asynchrony (SOA). In separate blocks, occasional deviant stimuli broke one, two, or three of the following regularities: spatial alternation, the constancy of SOA, or the dominant frequency-intensity conjunctions. Unlike the mismatch negativity (MMN) elicited by alternation-plus-SOA deviants, the MMN elicited by alternation-plus-conjunction deviants was approximately equal to the sum of the two corresponding single-deviant MMNs. These results suggest that the preattentive change-detection system processes infrequent violations of the structural regularities of sound sequences together with changes in temporal regularities, but separately from changes in repetitive spectral sound features. The MMN elicited by the triple-deviant stimuli corroborated these conclusions.     

Mismatch negativity in dichotic listening: Evidence for interhemispheric differences and multiple generators

The characteristics of mismatch negativity (MMN) elicited by dichotic stimulation were examined using frequency-deviant stimuli presented to the right, to the left, or to both sides. The experiment was run twice, once using earphones and once using loudspeakers in free field. With both modes of stimulation, deviants presented in the left, right, or both ears, or tones that were switched between ears, elicited comparable MMNs, with a peak latency of about 180 ms. With earphones, the amplitude of the MMN was bigger at the frontal-lateral right hemisphere sites than at the homologous left-hemisphere sites for all deviance conditions. Scalp current density analysis revealed that deviance in the right side elicited bilaterally equivalent frontal current sinks and a trend towards stronger contralateral current sources at the mastoid sites. In contrast, left side deviance elicited frontal sinks and temporal current sources stronger over the right hemiscalp. These results are compatible with the multiple-generator model of MMN. The attention-related role of the MMN is discussed, suggesting comparable attention mechanisms for vision and audition.     

Small Pitch Separation and the Selective-Attention Effect on the ERP

In conditions demanding selective attention, the stimuli to be attended elicit an event-related brain potential (ERP) component called the processing negativity (PN). However, some previous results might be interpreted as suggesting that even the stimuli to be ignored elicit some PN, at least if they physically resemble those to be attended. This “generalization’ of PN was studied by presenting three equiprobable tones differing in pitch in random order, one of the tones at a time being the target to be counted. The small pitch separation between the three tones was of different magnitude in different blocks. A PN was observed in the ERPs to the counted stimuli but was even elicited, though with a smaller amplitude, by the nontargets, at least by those which physically resembled the targets. The latter, “generalized’ PN was larger in amplitude with a smaller pitch separation between the nontarget and target stimulus. These results support the interpretation that PN reflects a cerebral matching process between the sensory input and a voluntarily maintained neuronal representation of the stimulus to be attended, the “attentional trace.’ Moreover, these results indicate rather high discrimination accuracy of this stimulus-selection mechanism underlying PN.     

Separability of Different Negative Components of the Event-Related Potential Associated with Auditory Stimulus Processing

The lateral scalp distribution of the early negative selective-attention effect on auditory eventrelated potentials (ERPs) was compared to the distribution of the midline-negative N1 component inverting in polarity at the scalp area below the Sylvian fissure. No such polarity inversion was observed for the selective-attention effect at the N1 latency. This result supports the proposal that another component, the processing negativity (PN), overlaps (summates to) the exogenous N1 in the ERPs to attended stimuli. A subsequent experiment demonstrated that the mismatch negativity (MMN: peaking at 150–200 ms), an ERP component automatically elicited by deviant stimuli presented among homogenous repetitive stimuli, had opposite polarities above and below the Sylvian fissure. Thus, the MMN at least partly emanates from a cerebral generator which considerably overlaps that of the N1. The results are discussed in terms of separability of the brain mechanisms involved in the voluntary and involuntary processing of auditory stimuli reflected by the PN and MMN, respectively.     

Mismatch negativity in children and adults, and effects of an attended task

Attention has been shown to modulate the amplitude of the mismatch negativity (MMN) elicited by a small deviation in auditory stimuli in adults. The present study examined the effects of attention and deviant size on MMN amplitude in children. Children and adults were presented with sequences of tones containing standards (1000 Hz) and three deviants varying in degree of deviance from the standard (1050, 1200, and 1500 Hz). Tones were presented in three conditions: (1) while participants ignored them; (2) while participants listened to them and responded to all three deviants; and (3) while participants again ignored them. We found that the MMNs elicited by the hard deviant (1050 Hz) were larger when the children were actively discriminating the stimuli than when they were ignoring them. However, the MMNs elicited by the easy and medium deviants (1500 and 1200 Hz, respectively) in the children and by all three deviants in the adults were not affected by attention.     

Loudness summation and the mismatch negativity event-related brain potential in humans

Abstract Infrequently omitting a sound from a repetitive sequence elicits the mismatch negativity (MMN) ERP response when the stimulus onset asynchrony (SOA) is less than 200 ms. We contrasted two alternative explanations of omission MMN. (1) Each sound starts a separate temporal integration process. Omissions violate the constancy of the temporal structure within the integration window. (2) Sounds preceding an omission are perceived to be louder than those followed by a sound within the integration period, because omissions allow the full stimulus aftereffect to be included in perceived loudness. We varied the SOA between 117 and 217 ms. For this case, the temporal structure explanation predicts that no MMN will be elicited, whereas the loudness summation explanation predicts that MMN will be elicited. MMN was elicited by tone omissions with random SOA, suggesting that loudness summation plays an important role in the elicitation of omission MMN.     

Effects of physiological aging on mismatch negativity: A meta-analysis

Mismatch negativity (MMN) is a promising window on how the functional integrity of auditory sensory memory and change discrimination is modulated by age and relevant clinical conditions. However, the effects of aging on MMN have remained somewhat elusive, particularly at short interstimulus intervals (ISIs). We performed a meta-analysis of peer-reviewed MMN studies that had targeted both young and elderly adults to estimate the mean effect size. Nine studies, consisting of 29 individual investigations, were included and the final total study population consisted of 182 young and 165 elderly subjects. The effects of different deviant types and duration of ISIs on the effect size were assessed. The overall mean effect size was 0.63 (95% CI at 0.43–0.82). The effect sizes for long ISI (> 2 s, effect size 0.68, 95% CI at 0.31–1.06) and short ISI (< 2 s, effect size 0.61, 95% CI at 0.39–0.84) were both considered moderate. A further analysis showed a prominent aging-related decrease in MMN responses to duration and frequency changes at short ISIs. It was also interesting to note that the effect size was about 25% larger for duration deviant condition compared to the frequency deviant condition. In conclusion, a reduced MMN response to duration and frequency deviants is a robust feature among the aged adults, which suggests that there has been a decline in the functional integrity of central auditory processing in this population.     

Magnetic mismatch fields elicited by vowels and consonants

 Auditory-evoked mismatch fields (MMFs) elicited by vowel contrasts and plosive stop consonant place-of-articulation contrasts were recorded over the left hemisphere of neurologically and audiologically normal subjects. Two experiments were conducted: vowels were presented in isolation in experiment 1 and embedded in consonant-vowel syllables in experiment 2. Best-fit equivalent MMF sources were obtained using the model of a single, spatiotemporal current dipole in a sphere. In both experiments, MMF sources activated by place-of-articulation contrasts were later in latency and smaller in dipole moment amplitude than MMF sources excited by vowel contrasts. There was evidence, albeit not unambiguous, for the vowel-contrast MMF sources being located more posteriorly than the consonant-contrast MMF sources in experiment 1 and more laterally in experiment 2. In both experiments, the MMF source excited by the contrast between /da/ and /ga/ was more anterior than the MMF source excited by the contrast between /da/ and /ba/. The effects on latency and dipole moment may be interpreted to mirror differences in perceptual discriminability and auditory memory decay between consonantal place-of-articulation contrasts and vowel contrasts. Similarly, the effects on location may be interpreted to reflect featural specificity of the mismatch response. Interestingly, the dipole source analysis results show a correspondence to the pattern of preservation and loss of the mismatch response to vowel and consonant place-of-articulation contrasts recently observed in Wernicke’s aphasia. Received: 14 June 1996 / Accepted: 6 February 1997     

Feature versus gestalt representation of stimuli in the mismatch negativity system of 7- to 9-year-old children

We examined preattentive auditory change detection in 7- to 9-year-old children. The question of interest was whether the preattentive comparison of stimuli indexed by the scalp-recorded mismatch negativity (MMN) was performed on representations of individual stimulus features or on gestalt representations of their combined attributes. The design of the study, based on a work by D. Deacon, J. Nousak, M. Pilotti, W. Ritter, and C. Yang (Psychophysiology, 1998), was such that both feature and gestalt representations could have been available to the comparator mechanism generating the MMN. The data indicated that for the majority of the children-those that exhibited an inverse relationship between the amplitude of the MMN and the probability of the deviant-the MMN was based on feature-specific information. This study also provides a method to obtain MMNs to deviants in three different features in the time usually required to obtain an MMN to a single acoustic feature.     

Mismatch negativity reflects numbers of tones of specific frequencies in humans

The human brain can automatically quantify objects and events. The mismatch negativity (MMN) of the event-related potential (ERP) was recently found to reflect one such ability in the auditory modality. The present study aimed to further validate the finding. ERPs were recorded in adult humans who passively listened to a series of 300-ms sequences of tones. The sequences occurred at 300-ms silent intervals. They comprised a total of six tones that each was either 440 or 660 Hz in frequency. MMN was found for a rare 4:2 ratio interspersed with a frequent 3:3 ratio between the tones of the different frequencies in a sequence, suggesting the ability of the human brain to preattentively discriminate numbers of sounds of specific frequencies at least at a 3:2 ratio.     

Scalp potentials to pitch change in rapid tone sequences. A correlate of sequential stream segregation

The object of the study was to look for a neurophysiological substrate of sequential auditory stream segregation. When a sequence of tones alternates rapidly between pitches separated by more than a few semitones, there is a tendency for it to be perceived as two independent "streams". We examined the scalp potentials evoked when the pitch interval abruptly changes, to see whether there are response parameters which might be correlated with sudden stream segregation and/or integration. For 3 s a continuous synthesized tone of "clarinet" timbre oscillated between pitches of F4 and F#4 (one semitone higher) at 16 notes/s, perceived as an integrated stream. The upper note was then raised to E5 (11 semitones above F4, perceived as segregated streams) for a further 3 s and the cycle was repeated 40 times. In a second condition also starting with oscillation between F4 and F#4, the upper note was lowered to E4 (one semitone below F4, still perceived as a single stream). Further conditions examined the changes between oscillations of 1 and 11 semitones down from E5, 1 and 23 semitones up from F4, and 10 and 11 semitones up from F4. Virtually no potentials were detectable during the periods of unchanging oscillation, but an N1/P2 complex was evoked on each change in the pitch interval. The N1 was termed "MN1" on account of its arguable relatedness to the mismatch negativity, recorded in a separate experiment using discontinuous tones at a much slower rate. The mean peak latency of the MN1 varied between 96 and 123 ms, the shortest latencies being recorded, not to the largest changes of pitch interval but to the widest pitch intervals between the new tone and the immediately preceding one. Therefore, although a causal relationship with streaming cannot necessarily be inferred, the MN1 latency appears to mark the degree of pitch contrast between consecutive tones, in correlation with the streaming effect. Electronic Publication