Progress in Neuro-Psychopharmacology and Biological Psychiatry
Impaired MMN/P3a complex in first-episode psychosis: Cognitive and psychosocial associations
Introduction
Event-related potentials (ERPs) are important neurophysiological biomarkers for research into schizophrenia (Javitt et al., 2008, Kasai et al., 2002). In particular, ERPs can be used to probe underlying regional and neurochemical hypotheses of schizophrenia as they reflect the integrity of cognitive pathways (Kasai et al., 2002). A review by Javitt et al. (2008) highlights the value of two ERP components in particular: the P3 (or P300) and mismatch negativity (MMN); both reduced in amplitude in schizophrenia, with moderate to large effect sizes across studies. Unlike P3, MMN deficits are relatively specific to schizophrenia (Catts et al., 1995, Salisbury et al., 2007, Umbricht et al., 2003) and correlate with the symptom severity and poor functional status (Baldeweg et al., 2004, Light and Braff, 2005a, Umbricht and Krljes, 2005). Overall, the findings suggest that MMN indexes a core deficit of psychosis (Javitt et al., 2008, Michie, 2001, Naatanen and Kahkonen, 2009, Umbricht and Krljes, 2005).
MMN reflects the brain's ability to extract relevant information from an irrelevant background; it indexes sensory memory and context-dependent information processing. Simply, MMN is a neurophysiological representation of an automatic response to deviant stimuli presented among an array of standard stimuli; it reflects the preliminary organization of subsequent controlled responses to a significant event (Oades et al., 2006). Typically, MMN tasks are auditory, passive and undemanding; the independence from active attention is considered to be an advantage (Michie, 2001). The peak of the MMN is observed about 200 ms after the presentation of a deviant (e.g. ‘frequency’ or ‘duration’) tone that differs (in pitch or time of exposure, respectively) from a series of more frequent standard tones. Notably, duration deviants have been found to be more sensitive than frequency deviants in detecting differences between schizophrenia and controls (Michie, 2001, Naatanen and Kahkonen, 2009). The negative deflection of the MMN is thought to reflect a neural mismatch between features of the current versus preceding stimuli, stored in short-term memory (Naatanen, 1990).
EEG recordings at the scalp find MMN amplitudes to be largest at fronto-central sites; however, source localisation and intracranial techniques have revealed generators within the auditory (temporal) cortex (Giard et al., 1995). Different underlying mechanisms of the MMN may operate; temporal lobe sources may be associated with pre-perceptual change detection, whereas frontal lobe sources may be related to involuntary attentional switching (Naatanen et al., 2007). Higher-order cognitive processes have been associated with each of these mechanisms; short-term verbal memory processes with the former and cognitive flexibility with the latter (Oades et al., 2006).
MMN amplitude deficits become established during the course of psychotic illness. While there are a few reports of reduced MMN in first presentation or early onset of schizophrenia (Javitt et al., 2000, Oades et al., 2006) there is a general consensus that more prominent MMN amplitude deficits tend to occur with illness progression (Javitt et al., 2000, Salisbury et al., 2002, Umbricht et al., 2006) and persist despite symptom improvement (Oades et al., 2006). Limited evidence that MMN is associated with auditory cortex grey matter loss in schizophrenia (Rasser et al., 2009) and in FEP (Salisbury et al., 2007) suggests that MMN indexes the progressive pathologic process in schizophrenia. Additionally, decreased temporal and frontal MMN correlates with deficits in verbal memory and attentional switching, respectively, in younger and older schizophrenia groups, suggesting the functional relevance of these changes (Baldeweg et al., 2004, Oades et al., 2006, Toyomaki et al., 2008, Umbricht and Krljes, 2005). There is, however, limited research in those with FEP, particularly with regard to the functional correlates of MMN (and associated measures).
Light and Braff (2005a) discovered that in 25 individuals with chronic schizophrenia, MMN deficits were associated with lower ratings on the Global Assessment of Functioning (GAF) and that they were highly predictive of the level of domestic independence. Similar associations (in similar samples) have been reported in other studies; GAF scores negatively correlate with phoneme-duration MMN (Kawakubo and Kasai, 2006), while decreased duration MMN has been shown to associate with lower GAF and poor performance in both proverb interpretation and verbal memory (Kiang et al., 2007). The predictive validity of MMN to social functioning was supported by the work of Kawakubo et al. (2007) who identified that in a group of 13 participants with schizophrenia, larger right frontal and temporal MMN values predicted the ability to acquire social skills.
In healthy normal subjects, Light et al. (2007) showed that the MMN is followed by a positive-going ERP component called ‘P3a’ at 250–300 ms post deviant stimuli. This P3a elicited during the passive ‘mismatch’ deviance detection task is unlike the canonical P300 or P3b elicited by tasks which require button pressing or counting of rare target stimuli (i.e. that reflect contextual updating). Rather, the P3a is largest at fronto-central sites and is thought to reflect automatic reorienting (Halgren et al., 1998, Keage et al., 2006, Polich, 2007). While not as frequently reported as the P3b, a number of studies show reduced P3a in schizophrenia (Mathalon et al., 2000, Turetsky et al., 2009, Turetsky et al., 1998). Light et al. (2007) extended the analysis of the traditional ‘mismatch’ paradigm by assessing the MMN/P3a complex with the view to explore the relationships between two markers of automatic sensory discrimination. They also examined the relationships of MMN and P3a with high-order cognitive measures and psychosocial functioning; following the logic that efficiency at elementary levels of information processing is required for successful cognitive and social functioning. Both MMN and P3a were associated with psychosocial functioning in normal subjects; with the P3a also showing a strong association with higher-order cognitive processes (Light et al., 2007).
Studies seeking to better characterise these relationships are required because cognitive deficits are apparent in the large majority of people with psychosis. Such deficits are present at first episode of psychosis (Bilder et al., 2000, Fitzgerald et al., 2004) and also predate the illness onset (Brewer et al., 2005). While many functions are affected, verbal memory and executive functions are the most characteristically impaired (Elvevag and Goldberg, 2000); and, even in the early stages of illness, they are related to structural and functional changes in the temporal and prefrontal cortex (Pantelis et al., 2003). More adequate delineation of these changes could help to determine the inter-relationships between dysfunctional auditory processing and cognitive and/or functional impairment in FEP. The goal, then, would be to elucidate the predictive capacity of associated biomarkers for longitudinal decline.
There is now good support for the role of these cognitive deficits in predicting social functioning (Addington and Addington, 2008, Green, 1996) and quality of life (Wegener et al., 2005) in psychosis. For example, limitations in working memory and long-term memory have been shown to predict poorer social skills (Reeder et al., 2006). However, cognitive deficits account for only a small (10–40%) amount of variance in social outcome (Couture et al., 2006). Therefore, there is a critical need to identify other factors that better predict cognitive and social outcomes particularly at the first episode of illness. A description of such factors in the early stages of illness will enhance the clarity regarding the implementation of preventative and treatment strategies (Yung and McGorry, 1996). The ultimate aim, then, is to prevent further functional decline (Simon et al., 2007) and improve quality of life. Only some studies have examined the relationships between various ERP components and cognition in psychosis. Those that have suggest that there may be a relationship between verbal memory dysfunction and defective processes at the early stage of encoding (Danion et al., 2007); some studies have suggested that people with schizophrenia are impaired in their ability to form and use transient memory traces (Javitt et al., 2000).
Few studies have specifically examined the relationship between MMN (and P3a), cognition and social functioning in psychosis. It remains unresolved, to which domains of higher cognitive functions, MMN/P3a are associated and whether there are significant relationships between MMN/P3a and quality of life. Thus, if MMN/P3a changes are apparent in psychosis and they also relate to key neurobiological changes, then they may be biomarkers for longitudinal progression (Banati and Hickie, 2009). The assessment of similar relationships (between deviance-related ERP markers, and cognitive and social functioning) in early stages of psychosis is the focus of this report.
Section snippets
Subjects
Seventeen first-episode psychosis (FEP) patients aged 18 to 29 were recruited from a specialised tertiary referral service for assessment and early intervention of mental health problems in young people (Scott et al., 2009). The FEP status of each patient was determined by a psychiatrist who made a primary diagnosis of schizophrenia (n = 1); schizoaffective disorder (n = 3); schizophreniform disorder (n = 5); bipolar disorder with psychotic features (n = 2); or major depressive disorder with psychotic
Demographic, clinical and social functioning findings
Matching for gender ratio and age was achieved (Table 1). As expected, there were significant between-group differences (at p < 0.01 and d > 0.8) across all of the clinical and social functioning measures. FEP showed significantly worse ratings for general psychiatric (BPRS) and current depression (HDRS) symptoms; FEP also showed increased levels of self-reported depression, anxiety and stress (DASS). Similarly, FEP patients were rated to have lower levels of Social and Occupational Functioning
Discussion
Young adults with first-episode psychosis showed deficits in two neurophysiological markers of deviance detection (MMN, P3a) with corresponding deficits in measures of cognitive and psychosocial functioning compared to matched healthy controls. The two neurophysiological markers, MMN and P3a, were distinguished by topographical differences and their associations with the ‘functional’ variables. The finding of reduced MMN amplitudes at fronto-central sites in FEP is consistent with a large body
Conclusion
This is the first study to examine the MMN/P3a complex, and associated cognitive and psychosocial functioning, in first-episode psychosis. While both MMN and P3a amplitudes were found to be significantly reduced in FEP, the P3a response recorded at the vertex (Cz) showed the most pronounced reductions and the strongest associations with psychosocial measures. The ERP components, MMN and P3, have been shown previously to be reliable biomarkers of chronic schizophrenia, typically via separate
Acknowledgements
This study was supported by a Faculty of Medicine, Early Career Researcher Grant Scheme at the University of Sydney. This study was also supported by an NHMRC Program Grant (No. 350241), Centres of Clinical Research Excellence Grant (No. 264611) and NHMRC Australia Fellowship (No. 511921). The authors would like to express their gratitude to the individuals that participated in this study. We also thank the anonymous reviewers for their excellent comments and suggestions on the manuscript.
References (67)
- et al.
Social and cognitive functioning in psychosis
Schizophr Res
(2008) - et al.
Mismatch negativity potentials and cognitive impairment in schizophrenia
Schizophr Res
(2004) - et al.
ERPs in schizophrenia: effects of antipsychotic medication
Biol Psychiatry
(1994) - et al.
Oxytocin enhances the encoding of positive social memories in humans
Biol Psychiatry
(2008) - et al.
Generators of the late cognitive potentials in auditory and visual oddball tasks
Electroencephalogr Clin Neurophysiol
(1998) - et al.
Neuroanatomy and neurophysiology in schizophrenia
Neurosci Res
(2002) - et al.
Support for an association between mismatch negativity and social functioning in schizophrenia
Prog Neuropsychopharmacol Biol Psychiatry
(2006) - et al.
The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories
Behav Res Ther
(1995) - et al.
Trait and state aspects of P300 amplitude reduction in schizophrenia: a retrospective longitudinal study
Biol Psychiatry
(2000) What has MMN revealed about the auditory system in schizophrenia?
Int J Psychophysiol
(2001)