Errors can elicit an error positivity in the absence of an error negativity: Evidence for independent systems of human error monitoring

Highlights

• Two components (Ne/ERN and Pe) of error-related brain activity are investigated.

• Error processing in both components relies on different types of information.

• The Ne/ERN but not the Pe requires a representation of the correct response.

• A Pe can emerge in the absence of an Ne/ERN.

• Pe and Ne/ERN reflect independent monitoring processes.

Abstract

Errors in human behavior elicit a cascade of brain activity related to performance monitoring and error detection. Whereas the early error-related negativity (Ne/ERN) has been assumed to reflect a fast mismatch or prediction error signal in the medial frontal cortex, the later error positivity (Pe) is viewed as a correlate of conscious error processing. A still open question is whether these components represent two independent systems of error monitoring that rely on different types of information to detect an error. Here, we investigated the prediction that the Ne/ERN but not the Pe requires a representation of the correct response to emerge. To this end, we created a condition in which no information about the correct response was available while error detection was still possible. We hypothesized that a Pe, but no Ne/ERN should be obtained in this case. Participants had to classify targets but ignore flankers that were always associated with an incorrect response. Targets but not flankers were masked with varying target-masking intervals. Crucially, on some trials no target at all was presented, thus preventing the representation of a correct response and the emergence of an Ne/ERN. However, because flankers were easily visible and responses to the flankers were always incorrect, detection of these flanker errors was still possible. In line with predictions of a multiple-systems account, we observed a robust Pe in the absence of an Ne/ERN for these errors. Moreover, this Pe relied on the same neural activity as that on trials with a visible target, as revealed by multivariate pattern analysis. These findings demonstrate that the mechanisms reflected by the two components use different types of information to detect errors, providing evidence for independent systems of human error monitoring.

Introduction

Error monitoring is crucial for achieving optimal goal-directed performance. Scalp EEG methods provide important evidence about the neural mechanisms underlying this ability. The error negativity (Ne; Falkenstein et al., 1990) or error-related negativity (ERN; Gehring et al., 1993) is a negative deflection over frontocentral electrodes, which peaks around 50 ms after an erroneous response, and is generated in the medial-frontal cortex (MFC; Ullsperger and von Cramon, 2001, Debener et al., 2005, Iannaccone et al., 2014). Whereas early theories suggested the Ne/ERN to reflect a mismatch between correct and actual responses (Scheffers and Coles, 2000), it has later been proposed to represent a post-response conflict (Yeung et al., 2004) or a prediction error (Holroyd and Coles, 2002). All these accounts have in common that the Ne/ERN is based on the detection of a discrepancy between the expected correct response and the actual response at an early stage of error processing. The Ne/ERN is followed by the error positivity (Pe), which is a parietal positivity occurring at about 200–500 ms after an error (Falkenstein et al., 1991, Overbeek et al., 2005). In tasks in which participants had to signal whether their response was correct or incorrect, the Pe was larger (e.g. Nieuwenhuis et al., 2001, Endrass et al., 2007) or appeared only (Murphy et al., 2012) when participants were aware of their errors. Therefore, the Pe is believed to be a correlate of error awareness (Steinhauser and Yeung, 2010, Ullsperger et al., 2014) or confidence about response correctness (Boldt and Yeung, 2015) representing a later stage of error processing.

Despite extensive research on the functional significance of the Ne/ERN and Pe, fundamental questions on their relationship remain. In particular, there has been little research investigating whether both components are causally related. One possibility is that the earlier Ne/ERN provides the basis for the later emergence of the Pe and error awareness (Scheffers and Coles, 2000, Yeung et al., 2004, Ullsperger et al., 2014), which implies a stage-like architecture of error monitoring. The other possibility is that Pe and error awareness can emerge independently of the Ne/ERN, which implies that different error detection mechanisms exist and may proceed independently of each other (Charles et al., 2013, Falkenstein et al., 2000, Maier et al., 2015). So far, little attempt has been made to directly test between these two accounts. Several studies investigated the relationship between Ne/ERN and error awareness (for a review, see Wessel, 2012), but these studies yielded mixed results. Some found that Ne/ERN amplitudes were positively correlated with error awareness (Scheffers and Coles, 2000, Wessel et al., 2011), while others reported no relationship (Endrass et al., 2007, Hughes and Yeung, 2011, Nieuwenhuis et al., 2001) or even a negative correlation (Di Gregorio et al., 2016, Maier et al., 2008). These inconclusive results might reflect that correlations between Ne/ERN and Pe can be moderated or even mediated by third variables which affect both components, such as stimulus perceptibility, attention or response conflict (for a discussion, see Di Gregorio et al., 2016).

The goal of the present study was to directly test the idea that Ne/ERN and Pe reflect independent mechanisms of error monitoring by asking whether the two mechanisms rely on different types of information to detect an error. More specifically, we hypothesized that only the mechanism underlying the Ne/ERN but not the mechanism related to the Pe requires a representation of the correct response to detect an error. This prediction can be derived from current theoretical accounts of both components. The Ne/ERN is assumed to result from the detection of a mismatch or post-response conflict between correct and actual response (Scheffers and Coles, 2000, Yeung et al., 2004) and thus necessarily requires a representation of the correct response. In contrast, error detection reflected by the Pe might be based on an evidence accumulation process (Steinhauser and Yeung, 2010, Ullsperger et al., 2010) which receives input from various sources, such as cognitive, autonomous, and sensory processing (Wessel et al., 2012, Wessel et al., 2011), and hence, does not necessarily require a representation of the correct response. Our approach was to construct a paradigm in which error detection was possible even if information about the correct response was fully omitted. In such a condition, we would expect that a Pe is still observed (because the error is detectable) while an Ne/ERN is absent (because no representation of the correct response is available). While such a pattern would demonstrate that the mechanisms underlying the Ne/ERN and Pe operate on different types of information, it would also provide direct evidence for the independence of the Ne/ERN and Pe. If the two components were causally related, no Pe should be observable in a condition in which no Ne/ERN was generated.

In the present experiment, participants performed a flanker task (Eriksen and Eriksen, 1974), in which they had to respond to the central target while ignoring the lateral flankers. The target but not the flankers was masked with varying stimulus-masking intervals. Crucially, on some trials, a zero interval was used and no target appeared at all. Because the Ne/ERN is assumed to require a representation of the correct response, we predicted the Ne/ERN to be absent on these trials. However, because we used a three-choice task and flankers were always associated with a different response than the target, participants knew that responses associated with the unmasked flankers must have been errors, thus making the detection of these errors easily possible. This allowed us to directly study whether a Pe can emerge in the absence of an Ne/ERN, and thus, whether Ne/ERN and Pe are based on dissociable and independent processes.