pubmed.ncbi.nlm.nih.gov

Not all errors are alike: theta and alpha EEG dynamics relate to differences in error-processing dynamics - PubMed

  • ️Sun Jan 01 2012

Not all errors are alike: theta and alpha EEG dynamics relate to differences in error-processing dynamics

Joram van Driel et al. J Neurosci. 2012.

Abstract

Performance errors in conflict tasks often result from inappropriate action impulses, and are thought to signal the need for increased control over the motor system. However, errors may also result from lapses in sustained attention, which may require different monitoring and adaptation mechanisms. Distinguishing between the mechanisms of adaptation is important as both error types may occur intermixed. To this end, we measured EEG of healthy human subjects while they performed three variants of the Simon task in which errors were more likely to occur due to attentional lapses, failures of motor control, or both. Behavioral results showed that subjects exhibited less conflict effects and less impulsive errors in sustained attention compared with the other Simon conditions. Time-frequency analyses of EEG data showed that the sustained attention Simon condition, compared with the motor control Simon condition, was characterized by: (1) less error-related MFC theta (4-8 Hz) power and an absence of error-related MFC-DLPFC theta phase synchronization; (2) stronger error-related suppression of parieto-occipital alpha (8-12 Hz) power and stronger parieto-occipital-frontal alpha synchronization. A control experiment, using SART (the Sustained Attention to Response Test), confirmed that adaptation after attentional lapses involved posterior alpha power suppression, in addition to inter-regional frontal theta activity. Together, these results suggest that at least two cortical mechanisms exist for performance monitoring, and that different tasks and task-settings can recruit these mechanisms in a different way. Post-error brain dynamics thus consist of heterogeneous activity from multiple neurocognitive processes.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.

Task design and trial sequence. A, Simon task, Experiments 1 and 2. Colors used: red–green and blue–yellow, mapped to either left or right thumb (counterbalanced across subjects), and presented either left or right from fixation. In one third of trials position on the screen and required response was incongruent. These settings were consistent for the different Simon conditions. B, SART Experiment 2. Trials contained digits 1–9 presented in lower left of visual field. One ninth of trials were NoGo (digit 5). Other settings were similar to the Simon task.

Figure 2.
Figure 2.

Behavioral results from Experiment 1. A, RT effects according to condition and accuracy on the current and previous trials, reflecting error speeding (A1) and posterror slowing (A2). B, Conditional accuracy functions showing accuracy as a function of quintiles of the RT-distribution, for congruent (dashed line, open circles) and incongruent (closed line, filled squares) trials of the different Simon conditions. Error bars denote SEM. GE, General; MC, Motor Control; SA, Sustained Attention Simon condition.

Figure 3.
Figure 3.

Behavioral results from Experiment 2. A, RT effects according to condition and accuracy on current and previous trial, reflecting error speeding (A1) and posterror slowing (A2). B, Conditional accuracy functions showing accuracy as a function of quintiles of the RT-distribution, for Simon congruent (CG; black dashed line, open circles), Simon incongruent (IC; black closed line, filled squares), and SART (gray closed line, filled squares). Error bars denote SEM. SI, Simon task; SA, SART.

Figure 4.
Figure 4.

Response locked oscillatory power averaged over the different Simon conditions in Experiment 1, for correct and error trials, and error-related activity (cE–cCc). A, Theta (4–8 Hz; left) and alpha (8–12 Hz; right) power over space, from top to bottom at 0 ms, 250 ms, and 500 ms after response. Black-white disks denote electrodes plotted in B. B, Power in dB relative to baseline for electrode FCz (left) and for pooled electrodes PO3/PO4 (right). Black line boxes in bottom row show time-frequency windows of interest used for ANOVA.

Figure 5.
Figure 5.

Error-related (cE–cCc trials) oscillatory power shows differential theta and alpha dynamics for the different Simon conditions in Experiment 1. A1, Time-frequency power plots for electrode FCz for the different conditions (rows). Black lines enclose regions of contiguous pixels that were significantly different from baseline at p < 0.0001, for at least 200 ms and three consecutive frequencies. B1, Same as in A1, for pooled electrodes PO3/PO4, with p < 0.001. A2, Topographical power plots for the theta band averaged over a postresponse time window of 50–300 ms. Black-white disks denote electrodes plotted in A1. B2, Same as in A2, for alpha 150–500 ms. A3, Line plots of FCz theta activity over time for the different conditions. B3, Same as in A3, for PO3/PO4 alpha.

Figure 6.
Figure 6.

Error-related (cE–cCc trials) interchannel phase synchronization shows differential long-range functional connectivity dynamics for the different Simon conditions in Experiment 1. A1, Topographical difference maps of FCz-seeded theta synchronization, for Motor Control more than Sustained Attention, over a 50–300 ms postresponse time window. Black-white disks denote target electrodes plotted in A2. B1, As in A1, PO3/PO4-seeded alpha synchronization for Sustained Attention more than Motor Control, 400–600 ms. A2, Line plots of error-related theta ICPS over time for the different conditions. B2, As in A2, for alpha ICPS.

Figure 7.
Figure 7.

Cross-subject rank correlations for the different Simon conditions. A1, Error-related (cE–cCc) interchannel theta phase synchrony between FCz and F5/F6 correlates with error-related theta power at FCz only in the Motor Control condition. A2, Same as in A1, for FCz–POz ICPS. B, FCz theta power at error trials predicts reaction time on subsequent correct trials only in the General and Sustained Attention condition (lower RT rank means faster response).

Figure 8.
Figure 8.

Error-related (cE–cCc trials) oscillatory power shows similar theta but strongly differing alpha dynamics for Simon versus SART in Experiment 2. A1, Topographical power plots for the theta band averaged over a postresponse window of 50–300 ms, for Simon (top) and SART (bottom). Black-white disks denote electrodes plotted in A2. B1, Same as in A1, for alpha 150–500 ms. A2, Time-frequency power plots for electrode FCz for Simon (top) and SART (bottom). Black lines enclose regions of contiguous pixels that were significantly different from baseline at p < 0.0001. B2, Same as in A2, for pooled electrodes PO3/PO4. A3, Line plots of FCz theta activity over time for Simon (gray) and SART (black). B3, Same as in A3, for PO3/PO4 alpha.

Figure 9.
Figure 9.

Error-related (cE–cCc trials) interchannel phase synchronization shows differential long-range functional connectivity dynamics for Simon versus SART in Experiment 2. A1, Topographical difference maps of FCz-seeded theta synchronization, for SART more than Simon, over a 50–300 ms postresponse time window. Black-white disks denote target electrodes plotted in A2. B1, As in A1, PO3/PO4-seeded alpha synchronization for SART more than Simon, 150–500 ms. A2, Line plots of error-related theta ICPS over time for Simon (gray) and SART (black). B2, As in A2, for alpha ICPS.

Figure 10.
Figure 10.

Cross-subject rank correlations between theta power and theta ICPS for the two tasks in Experiment 2. Error-related (cE–cCc) interchannel theta phase synchronization between FCz and F5/6 correlates with error-related FCz theta power in the SART (right; black) but not in the Simon task (left; gray).

Similar articles

Cited by

References

    1. Alexander WH, Brown JW. Medial prefrontal cortex as an action-outcome predictor. Nat Neurosci. 2011;14:1338–1344. - PMC - PubMed
    1. Bengson JJ, Mangun GR, Mazaheri A. The neural markers of an imminent failure of response inhibition. Neuroimage. 2012;59:1534–1539. - PubMed
    1. Berens P. CircStat: A MATLAB toolbox for circular statistics. J Stat Softw. 2009;31:1–21.
    1. Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD. Conflict monitoring and cognitive control. Psychol Rev. 2001;108:624–652. - PubMed
    1. Burle B, van den Wildenberg WPM, Ridderinkhof KR. Dynamics of facilitation and interference in cue-priming and Simon tasks. Eur J Cogn Psychol. 2005;17:619–641.

Publication types

MeSH terms

LinkOut - more resources