Contrasting time and frequency domains: ERN and induced theta oscillations differentially predict post-error behavior

Abstract

The present study investigated the neural dynamics of error processing in both the time and frequency domains, as well as associated behavioral phenomena, at the single-trial level. We used a technique that enabled us to separately investigate the evoked and induced aspects of the EEG signal (Cohen & Donner, 2013, Journal of Neurophysiology, 110[12], 2752-2763). We found that at the single-trial level, while the (evoked) error-related negativity (ERN) predicted only post-error slowing (PES)and only when errors occurred on incongruent trials-induced frontal midline theta power served as a robust predictor of both PES and post-error accuracy (PEA) regardless of stimulus congruency. Mediation models of both electrophysiological indices demonstrated that although the relationship between theta and PEA was mediated by PES, there was not a relationship between the ERN and PEA. Our data suggest that although the ERN and frontal midline theta index functionally related underlying cognitive processes, they are not simply the same process manifested in different domains. In addition, our findings are consistent with the adaptive theory of post-error slowing, as PES was positively associated with post-error accuracy at the single-trial level. More generally, our study provides additional support for the inclusion of a time-frequency approach to better understand the role of medial frontal cortex in action monitoring.Keywords ERN . Frontal midline theta . Action monitoring . Induced vs. evoked Within simple forced-choice decision-making tasks, committing an error initiates a cascade of cognitive events within the action-monitoring system. The participant will briefly disengage from the ongoing task, attend to and identify the error, and make cognitive and behavioral adjustments in an effort to prevent the error from reoccurring once they reengage in the task (Wessel, 2017;Wessel & Aron, 2017). In particular, these behavioral adjustments include increases in response time on trials that follow error commission (post-error slowing [PES]; Laming, 1979), as well as either increases or decreases in subsequent trial accuracy (post-error accuracy [PEA]). A major topic of debate in the research community is whether this slowing is an adaptive or maladaptive consequence of an error. Findings in support of adaptive theories suggest that PES provides additional time to process subsequent trial task stimuli, consequently enabling a relative improvement in task performance (increased PEA; Botvinick, Braver, Barch, Carter,

Subthreshold error corrections predict adaptive post‐error compensations

“…For all analyses, post-error slowing was computed by subtracting current trial error RTs from next trial (post-error) RTs (Schroder et al, 2020). Post-error accuracy during averagedlevel subset analyses was computed as a percentage change between (1) the accuracy for trials following an error and (2) the accuracy for trials following a correct response (Beatty et al, 2018(Beatty et al, , 2020Buzzell et al, 2017), while post-error accuracy during single-trial analyses was computed as the accuracy on post-error trials. Analyses for post-error slowing and post-error accuracy were conducted using separate error-only 2 (Congruency: Congruent, Incongruent) × 2 (Correction: Corrected, Uncorrected) ANOVAs.…”

“…While many studies have investigated how the magnitude of error processing influences post-error compensations such as PES, they omit a critical step-the tendency to salvage the ongoing trial (see Figure 1). For example, while some studies explicitly investigate error corrections by experimentally modulating their likelihood (Crump & Logan, 2013;Fiehler et al, 2004Fiehler et al, , 2005Steinhauser, 2010;Ullsperger & von Cramon, 2006), other studies conclude that error corrections are potential confounds to the research question being proposed, and therefore, contend with them by either forbidding them altogether (Amengual et al, 2013;Buzzell et al, 2017;de Visser et al, 2018;van Meel et al, 2007) or removing them post hoc (Beatty et al, 2018(Beatty et al, , 2020. Thus, long-standing questions regarding the functional significance of error monitoring might be clarified by investigating corrections that occur between trial N and trial N + 1.…”

“…When investigating the relation between error processing and post‐error behavior, researchers often observe increases in post‐error response time (post‐error slowing; PES). Although there is considerable evidence suggesting that PES reflects increased caution, resulting in improved performance on the subsequent trial (Beatty et al., 2020; Botvinick et al., 2001; King et al., 2010; Maier et al., 2011; Marco‐Pallarés et al., 2008; Steinhauser & Andersen, 2019), there is also evidence suggesting that PES reflects distraction, resulting in a decline in performance on the subsequent trial (Beatty et al., 2018; Buzzell et al., 2017; Jentzsch & Dudschig, 2009; Notebaert et al., 2009; Steinhauser et al., 2018; Ullsperger & Danielmeier, 2016; Van der Borght et al., 2016). These disparate findings for adaptive and maladaptive PES can partially be explained by the duration of the response‐stimulus interval (Danielmeier & Ullsperger, 2011), as well as differences in the method chosen to compute PES (Schroder et al., 2020).…”

“…Although researchers have found that event‐related potentials such as the error‐related negativity (ERN), the correct response negativity (CRN), and the error positivity (Pe) could index error corrections, findings have been inconsistent (ERN: Crump & Logan, 2013; Fiehler et al., 2004; Meckler et al., 2017; Navarro‐Cebrian et al., 2016; Rodrı́guez‐Fornells et al., 2002; Ullsperger & von Cramon, 2006; Pe: Ficarella et al., 2019; Kieffaber et al., 2016) suggesting that these components may not be a reliable index of subthreshold error corrections. Recently, two studies (Beatty et al., 2020; Valadez & Simons, 2017) have suggested that induced, transient, rhythmic cortical activity, is more closely linked to post‐error compensations than evoked cortical activity such as event‐related potentials (ERPs). More specifically, while the ERN was predictive of post‐error slowing, frontal midline theta power (4–7 Hz) predicted both increases in post‐error slowing and post‐error accuracy (Beatty et al., 2020; Valadez & Simons, 2017).…”

“…Recently, two studies (Beatty et al., 2020; Valadez & Simons, 2017) have suggested that induced, transient, rhythmic cortical activity, is more closely linked to post‐error compensations than evoked cortical activity such as event‐related potentials (ERPs). More specifically, while the ERN was predictive of post‐error slowing, frontal midline theta power (4–7 Hz) predicted both increases in post‐error slowing and post‐error accuracy (Beatty et al., 2020; Valadez & Simons, 2017). We propose that, similar to the findings for post‐error adjustments that occur from one trial to the next, it is possible that induced rhythmic oscillatory activity would be able to predict subthreshold error corrections.…”