Australian researchers were the first to demonstrate reduced amplitude in patients of the brain event-related potential, mismatch negativity (MMN), generated in response to a deviant sound in an otherwise constant background (Shelley et al, 1991). Since that first report, in excess of 70 papers have been published replicating this finding. It is one of the most robust findings in the schizophrenia literature giving an overall effect size in a recent meta-analysis of approximately 1.00. This first report fortuitously used sound duration as the deviant attribute, and therefore tapped into the temporal processing capacities of the auditory system. Subsequent research, again initiated by Australian researchers, suggests that temporal processing as reflected in MMN amplitude may be differentially impaired in patients and possibly in first degree relatives. In this symposium, we (i) summarise past data implicating a differential deficit in temporal processing, (ii) present new data that support this contention and show that it is related to global impairments in everyday functioning and that its neurobiological basis probably resides in a combination of abnormal activation of temporal and frontal lobe generators, and (iii) suggest some new approaches to the use of MMN as an index of temporal processing capacities of the auditory system in the future.
O:17 Evidence from MMN of temporal processing deficits in schizophrenia
P.T. Michie
School of Psychology, University of Newcastle, Australia. Neuroscience Institute of Schizophrenia and Allied Disorders (NISAD), Sydney
Aims/Background The objective is to provide a brief overview of MMN research demonstrating that duration MMN (generated in response to a sound deviant in duration) may be particularly sensitive to auditory system abnormalities in schizophrenia.
Results/Conclusions One of the first indications that duration MMN may be particularly sensitive to auditory system dysfunction in schizophrenia came from a study in which MMN to duration deviants was compared to sounds that were deviant in frequency (pitch). A subsequent meta-analysis confirmed this finding by showing a significantly larger effect size for duration than for frequency. More recently we have shown that duration MMN is reduced in patients with a short illness duration (mean duration 2.6 years) relative to age and sex matched controls whereas frequency MMN is not. The opposite pattern was evident in patients with a longer illness duration (mean duration 18.9 years): frequency MMN is reduced but duration MMN is not. This puzzling result is due a decline in duration MMN in the older controls (aged and sex matched to the longer duration illness patients). The findings indicate that patients in early stages of the disorder produce duration MMN amplitudes equivalent to healthy controls 25 years older and given that it is so marked early in the disorder, may be reflective of a trait vulnerability, consistent with data showing similar patterns in first degree relatives. Frequency MMN on the other hand, seems to be sensitive to illness course and possibly to neuropathological changes that occur with the progression of the disorder.
O:18 Temporal processing in schizophrenia – evidence for impairment across a range of time scales
N.L. Matthews∗, J. Todd, G. Cooper, S. Finnigan, S. Catts, P.T. Michie
School of Psychology, University of Newcastle, Australia. Neuroscience Institute of Schizophrenia and Allied Disorders (NISAD), Australia.
Aims/Background There is increasing evidence that individuals with schizophrenia exhibit impairments in the encoding of the temporal characteristics of sounds. The present study was designed to further explore this ability in individuals with schizophrenia in order to determine if impairment can be demonstrated on a range of temporal processing tasks: encoding of the duration (or length) or sounds, gap detection and interaural differences in timing (ITD).
Method The integrity of cortical representations of the temporal characteristic of sounds in each task was investigated using the mismatch negativity (MMN) component of the auditory-event-related-potential in individuals with schizophrenia (n = 20) and controls (n = 19). In the Duration MMN condition, sound sequences consisted of 90% standards of 50 ms duration, and 10% deviants of 100 ms duration. For the Gap MMN condition, 88% of the sounds were 100 ms white noise bursts as standards, and 12% deviants which contained a 18 ms gap in the center of the sound. For the ITD MMN condition, participants were presented with 80% standards with no ITD and 20% deviant tones that had an ITD of 700 μsec, 10% of these had a right ear leading ITD and 10% had a left ear leading ITD.
Results Individuals with schizophrenia had reduced MMN amplitudes in all conditions.
Conclusion This finding provides evidence for a pervasive impairment in the encoding of the temporal properties of sounds across a range of time scales in individuals with schizophrenia.
O:19 Insensitivity to temporal context in schizophrenia
A.L. Richards, J. Todd∗, P.T. Michie
School of Psychology, The University of Newcastle, Australia. Neuroscience Institute of Schizophrenia and Allied Disorders (NISAD), Australia
Aims/Background In this study we explored how temporal context affects the mismatch negativity (MMN) produced by healthy controls and individuals with schizophrenia. Elicitation of MMN occurs when the auditory system detects a mismatch between a current sound and the memory formed of a repetitious sequence. MMN can be conceptualised as occurring when a sound violates the context established by preceding events. Individuals with schizophrenia are known to show impairments in the ability to utilise context when processing the relevance of an event. In this study we examined whether this impairment in utilising context would result in an insensitivity to the effects of temporal context on MMN.
Method Eighteen patients and age and sex-matched controls were presented with two sound sequences over headphones with attention directed towards a silent movie with subtitles. In the Fixed SOA sequence MMN was measured to a 100 ms duration deviant amongst 50 ms standard tones at a regular 500 ms stimulus onset asynchrony (SOA). The same tones were used in a Jittered SOA sequence with irregular SOAs ranging from 250–750 ms.
Results In controls, the MMN elicited by duration deviants with jittered SOA was significantly reduced compared to the same deviant occurring in a sequence with fixed SOA. However, changing temporal context/regularity had no significant impact on the MMN produced by the schizophrenia group.
Conclusions The results indicate that individuals with schizophrenia have difficulty encoding and forming a model of the temporal context in which a sound occurs.
O:20 Mismatch negativity (MMN), executive function and symptom severity in 1st episode and chronic schizophrenia
P.B. Ward∗, L. Meyer, E. Stone, E. Connaughton, K. Metcalf, U. Schall, P.T. Michie, P. Johnston, J. Todd
School of Psychiatry, University of New South Wales & Schizophrenia Research Unit, Sydney South West Area Health Service; Centre for Mental Health Studies, and School of Behavioural Sciences, University of Newcastle
Aims/Background As part of a multi-site collaborative study, we examined the clinical and neuropsychological correlates of mismatch negativity (MMN), an ERP index of auditory sensory memory in patients with schizophrenia with either a short duration of illness, or a more chronic illness. We tested subjects using both the classic ‘oddball’ paradigm, and a novel paradigm in which the number of standards preceding a deviant varied systematically.
Methods We tested patients meeting DSM-IV criteria for schizophrenia with a short duration of illness (DOI) (<2 years) and with a long DOI (>5 years) and age and gender matched healthy volunteers. Assessment of executive functions included the Hayling (Sentence Completion) and Brixton (Spatial Anticipation) Battery. EEG data were recorded from 64 scalp electrodes whilst subjects watched a silent movie, and tones were presented via headphones. The paradigm consisted of standard duration (50 ms) tones interspersed with deviant duration tones (100 ms) at a rate of 8%. MMN amplitudes at F3, Fz and F4 (fronto-central left, midline and right) were analysed using RMANCOVA, (within-subject factors of deviant probability and electrode, between-subject factor of group [1st episode schizophrenia, chronic schizophrenia, and healthy control]. Relationships between MMN amplitude, neuropsychological test results and symptom severity scores were analysed using Spearman's correlations.
Results There was a significant group effect for the MMN amplitude (p < 0.05) indicating lower MMN amplitudes for the long DOI patient group. A significant group by electrode interaction (p < 0.05) reflected that patients with long DOI, but not short DOI, showed smaller MMN amplitudes in the left hemisphere whereas in healthy controls amplitudes were greater in the right hemisphere. In the short DOI patient group, reduced frontal MMN amplitude was significantly correlated with lower GAF scores (rs = .77, p < 0.05), and greater attentional impairment (rs = .90, p < 0.01). In the long DOI patients, reduced frontal MMN amplitudes were significantly correlated with greater affective flattening (rs = .80, p < 0.01) on the SANS.
Conclusions MNN amplitude reduction differs according to duration of illness in patients with schizophrenia, and there is evidence of different patterns of correlation with clinical ratings.
O:21 Spatiotemporal activation of MMN generators in schizophrenia
W.R. Fulham, U. Schall∗, P.B. Ward, P.T. Michie, P. Johnston, P. Thompson
Centre for Mental Health Studies, The University of Newcastle, Australia. Neuroscience Institute of Schizophrenia and Allied Disorders, Sydney, Australia
Aims/Background Little is known about which cortical areas and generators contribute to the scalp-recorded MMN reduction in schizophrenia. In order to determine differences in the spatiotemporal activation of MMN generators, EEG, structural and functional MRI were obtained from chronic and first episode patients, and age-gender matched controls.
Methods A Duration MMN paradigm was recorded at 64 sites (standards 92%, 50 ms; deviants 8%, 100 ms). Equivalent dipole and current source density analysis was performed using Curry V4.6. Structural MRIs were used to construct Realistic Head Models and cortical surfaces for each individual for source analysis. Using the LONI procedures, cortical gyri were identified as landmarks permitting the results to be mapped into a common brain space. Event-related fMRI contrasts between deviant and standard tones were also obtained.
Results A reduction in MMN amplitude was seen in the younger patients but not the older cohort. Cortically constrained LORETA analysis of the early phase of MMN (<200 ms) suggested that the major cortical generator lies in the Superior Temporal Gyrus (STG), particularly in young subjects, while bilateral Rolandic Operculum sources accounted for the later MMN component (>200 ms). Event-related fMRI analysis in the same subjects also showed corresponding activation in the STG, while Inferior Frontal Orbital activation accounted for BOLD response-correlation with MMN scalp amplitude. Further BOLD activation in the Superior Frontal Cortex was significantly reduced in schizophrenia.
Conclusions Depending on brain imaging methodology, differential aspects in relation to age (i.e., STG) and illness (i.e., prefrontal cortex) were found for auditory MMN.
O:22 Between channel gap detection as a measure of left hemisphere advantage for temporal processing in the auditory system
J. Todd∗, R. Barker
School of Psychology, University of Newcastle, Australia
Aims/Background The mismatch negativity (MMN) component of the auditory event-related potential is significantly reduced in schizophrenia (SZ). There is growing recognition that the MMN elicited by a change in the temporal dimension of sound (namely duration MMN), is a particularly sensitive measure of this deficit. We present the first in a series of studies designed to test whether duration MMN indexes a core deficit in temporal processing in SZ. The design was based upon evidence that the left auditory cortex is specialized for processing faster rates of change in sound properties than the right. We reviewed psychophysical literature to determine the optimal sound parameters to reveal a left hemisphere/right ear advantage (LH/REA) in the MMN to a short gap presented within a sound.
Methods 23 volunteers (aged 18–35) were presented with a sequence of sounds to the left and right ear. The sequence comprised repeating standard no-gap sounds and rare deviant sounds containing a 40 ms gap after a leading marker length of 15 ms, 65 ms or 115 ms (6%) each. The event-related potential to the standard was subtracted from that to each deviant to extract the MMN elicited to gap sound.
Results Results revealed a significantly larger MMN response to deviants presented to the right vs left ear (i.e., a significant LH/REA for detecting the temporal change).
Conclusions This is the first study to reveal a LH/REA in MMN measures of temporal processing. The planned application to SZ and aging studies of temporal processing deficits will be discussed.
