Anxiety not only increases, but also alters early error-monitoring functions

Abstract

Anxiety has profound influences on a wide range of cognitive processes, including action monitoring. Event-related brain potential (ERP) studies have shown that anxiety can boost early error detection mechanisms, as reflected by an enhanced error-related negativity (ERN) following errors in high-anxious, as compared with low-anxious, participants. This observation is consistent with the assumption of a gain control mechanism exerted by anxiety onto error-related brain responses within the dorsal anterior cingulate cortex (ACC). However, whether anxiety simply enhances or, rather, alters early error detection mechanisms remains unsolved. In this study, we compared the performance of low- versus high-trait-anxious participants during a go/no-go task while high-density EEG was recorded. The two groups showed comparable behavioral performance, although levels of state anxiety increased following the task for high-anxious participants only. ERP results confirmed that the ERN/Ne to errors was enhanced for high-anxious, relative to low-anxious, participants. However, complementary topographic analyses revealed that the scalp map of the ERN/Ne was not identical between the two groups, suggesting that anxiety did not merely increase early error detection mechanisms, but also led to a qualitative change in the early appraisal of errors. Inverse solution results confirmed a shift within the ACC for the localization of neural generators underlying the ERN/Ne scalp map in high-anxious participants, corroborating the assumption of an early effect of anxiety on early error-monitoring functions. These results shed new light on the dynamic interplay between anxiety and error-monitoring functions in the human brain

Evidence for the automatic evaluation of self-generated actions

“…The novel contribution of our study is to show that beyond these enhanced arousal or attention orienting effects following the detection of these adverse events, dedicated internal monitoring systems enable organisms to rapidly map specific affective values (either negative or positive) onto self-generated actions (either incorrect or correct). Indirect evidence for this idea was already provided by earlier ERP studies showing that the rapid processing and monitoring of response errors involved an emotional component that might be altered in (trait) anxiety or negative affect (Aarts & Pourtois, 2010;Hajcak et al, 2004). The results of the present study add to the literature that this mechanism appears to operate along a genuine valence dimension, which is not restricted to errors or a specific class of deviant outcomes (De Bruijn an abstract level of action representation, as opposed to being bound to a specific motor output or command.…”

“…More specifically, high anxious participants did not show a normal amplitude variation of the FRN (time-locked to the evaluative feedback) as a function of the perceived "correctness" of their actions (Aarts & Pourtois, 2012). Therefore, we predicted that trait anxiety might actually decrease the size of the evaluative priming effect, bearing in mind that this effect is assumed to depend upon an automatic and fast affective marking of self-generated actions (responses errors/bad vs. correct responses/good) and this specific process that may be impaired in high anxious participants (Aarts & Pourtois, 2010). Dutch speakers who did not have a history of neurological or psychiatric disease and had normal or corrected-to-normal vision.…”

“…Earlier event-related potential (ERP) studies have also provided evidence for an early emotional processing of errors following their onset (Luu, Collins, & Tucker, 2000), as well as for an early differential appraisal of errors compared to correct hits (Aarts & Pourtois, 2010;Hajcak, McDonald, & Simons, 2004). Moreover, it has been found that early brain responses related to error detection are typically enhanced in high trait anxious individuals.…”

“…It was increased to 600 ms in Experiment 3 and to 1000 ms in Experiment 4. If the evaluation of correct and incorrect actions is also a fast acting automatic process (see Pourtois et al, 2010, for converging neuroscientific evidence) then evaluative priming should be observed at short (300 ms) but not long (1000 ms) SOAs.…”

“…Actions in response to stimuli are usually deemed conducive or obstructive depending on their actual match with goals stored in long term memory (Scherer, 1984(Scherer, , 1988. Indirect evidence supporting this view comes from recent psychophysiology studies showing that unwanted response errors (i.e., goal obstructive events) yield larger skin conductance responses and greater heart rate deceleration than Evaluation of actions 4 correct decisions (Hajcak, McDonald, & Simons, 2003b), as well as a larger startle potentiation (Hajcak & Foti, 2008) and differential early activation in the amygdala (Pourtois et al, 2010). These results suggest that errors may be perceived as aversive events, and accordingly be associated with enhanced arousal within the autonomic nervous system.…”