Journal article
SA68. Dysfunctional Prediction Error Coding in Schizophrenia: Test–Retest Reliability of Auditory Mismatch Negativity and Repetition Positivity
Schizophrenia bulletin, Vol.43(Suppl 1), pp.S137-S138
03/2017
DOI: 10.1093/schbul/sbx023.067
PMCID: PMC5475964
Abstract
Background:
The mismatch negativity (MMN) event-related potential (ERP) reflects predictive coding as well as short-term neuroplasticity. MMN is elicited when a deviant tone violates the prediction of an expected, frequently presented standard tone. ERP amplitudes to standard tones become more positive (repetition positivity, RP) and amplitudes to deviant tones become more negative (MMN) as the number of standard tone repetitions increases. The RP reflects the strength of the memory trace to standard stimuli, while the MMN reflects detection of prediction errors. To examine the temporal stability of these responses, we examined MMN and RP in patients with schizophrenia (SZ) and healthy controls (HC) at baseline and at a two-week follow-up.
Methods:
SZ (n=44) and HC (n=30) participated in an auditory MMN roving standard paradigm using double-deviant stimuli (deviants change in both pitch and duration). We examined ERPs to the 3rd, 8th, and 33rd standard tones and their immediately subsequent deviant tones as well as the deviant minus standard MMN difference wave. Intraclass correlation coefficients (ICC) were calculated to assess test–retest reliability of RP and MMN amplitudes.
Results:
For standard tones, there was a significant main effect of repetitions,
F
(2, 118) = 29.08,
P
< .001, indicating that amplitudes became more positive as the number of repetitions increased. There was no main effect of group or Group × Repetition interaction. For deviant tones, there was a significant main effect of group,
F
(1, 59) = 7.00,
P
< .01; repetitions,
F
(2, 118) = 12.75,
P
< .001; and a significant quadratic Group × Repetition interaction,
F
(1, 59) = 4.12,
P
< .05. The interaction was due to HC showing a larger increase in negativity than SZ, as the number of standard repetitions preceding the deviant increased. For MMN, there was a significant main effect of group,
F
(1, 59) = 6.58,
P
< .02; repetitions,
F
(2, 118) = 29.26,
P
< .001; and a significant quadratic Group × Repetition interaction,
F
(1, 59) = 4.58,
P
< .04. The interaction was due to HC showing a larger increase in MMN amplitudes than SZ as the number of standard repetitions preceding the deviant increased. Reliability of standard and deviant amplitudes was poor to good (ICCs, .21–.61) in both groups. MMN amplitude ICCs were poor in both groups after 3 repetitions and fair after 8 and 33 (ICCs, 0.31–0.57) repetitions.
Conclusion:
Our results suggest that SZ have an intact memory trace mechanism to repeated tones (RP) but a dysfunctional prediction error detection mechanism to deviant tones (MMN). Reliability of ERPs to standard, deviant, and difference wave ERPs was poor to good but generally showed better reliability as the number of standard repetitions increased. These results indicate that SZ exhibit deficits in short-term auditory neuroplasticity and that ERP measurements of this phenomenon are relatively stable over time.
Details
- Title: Subtitle
- SA68. Dysfunctional Prediction Error Coding in Schizophrenia: Test–Retest Reliability of Auditory Mismatch Negativity and Repetition Positivity
- Creators
- Jonathan WynnAmanda McCleeryWarren SzewczykEugene KutasevichBrian J RoachDaniel MathalonMichael F Green
- Resource Type
- Journal article
- Publication Details
- Schizophrenia bulletin, Vol.43(Suppl 1), pp.S137-S138
- DOI
- 10.1093/schbul/sbx023.067
- PMCID
- PMC5475964
- ISSN
- 0586-7614
- eISSN
- 1745-1701
- Publisher
- Oxford University Press
- Language
- English
- Date published
- 03/2017
- Academic Unit
- Psychiatry; Psychological and Brain Sciences; Iowa Neuroscience Institute
- Record Identifier
- 9984214755102771
Metrics
24 Record Views