Research report
Error processing and impulsiveness in normals: evidence from event-related potentials

https://doi.org/10.1016/j.cogbrainres.2005.02.003Get rights and content

Abstract

Electrophysiological correlates of impulsiveness were investigated in thirty-two healthy subjects using event-related potentials (ERP). Impulsiveness was determined by calculating individual reaction times (as a function of general response speed) in order to split the entire group into two subgroups with a more controlled (n = 16) and less controlled (n = 16) response style. Participants performed a Go/Nogo task while a 64 channel EEG was recorded. Artifact-free EEG segments were used to compute ERPs on correct Go trials and incorrect Nogo trials, separately. Three ERP components were of special interest: the error-related negativity (ERN)/error negativity (Ne) and the “early” error positivity (Pe) reflecting automatic error processing and the “late” error positivity (Pe) which is thought to mirror the awareness of erroneous responses. Subjects with higher impulsiveness showed smaller amplitudes than subjects with lower impulsiveness for the ERN/Ne component and the “early” Pe component. With regard to the “late” Pe groups did not differ. Hence, ERP measures appear suitable for detailed analyses of impulsiveness in healthy participants. Moreover, present results argue for the necessity of careful control of impulsiveness when including normal comparison groups in the context of clinical studies.

Introduction

The concept of impulsiveness has a long tradition in psychiatry [3]. Impulsiveness is a core feature of a variety of psychiatric diseases like attention-deficit/hyperactivity disorder, drug intoxication, borderline personality disorder, and antisocial personality disorder [24], [31]. Pharmacological studies found close connections between impulsive personality traits and a dysfunction of the serotonergic and noradrenergic system [26]. This finding was confirmed by several positron emission tomography (PET) studies which showed hypometabolism in prefrontal cortical areas [8], [27] reflecting a diminished serotonergic turnover and consecutively an impaired regulation of impulsive behavior [17].

Event-related brain potentials (ERPs) are a useful tool to investigate impulsiveness because they permit tracking the time course of fast cognitive processing on-line with a time resolution in the range of milliseconds. During the last years, special attention was paid to the error negativity (Ne; [10]) or error-related negativity (ERN; [13]), an ERP component which mirrors erroneous responding in forced choice reaction time paradigms like the Eriksen flanker task [9], [19], [23]. The ERN/Ne is a negative ERP deflection peaking between 100 and 150 ms after the onset of an erroneous response [25]. Larger amplitudes of the ERN/Ne were found when task instructions emphasize accuracy over speed (speed accuracy trade-off; [14]). Experimental evidence from dipole solutions of the ERN/Ne with brain electric source analysis (BESA) and from several fMRI studies (e. g. [5]) pointed to neural generators in the anterior cingulate cortex (ACC).

Originally, the ERN/Ne was considered in the context of error detection resulting from a mismatch between the representation of the correct response and the representation of the actual (false) response [10], [14]. Alternative accounts view the ERN/Ne as a brain potential reflecting the response evaluation process itself rather than the outcome of this process [29]. Rather contrary to these interpretations, Cohen and coworkers interpret the ERN/Ne to be associated with the detection of response conflict [4], [5].

Several studies showed that variability in the amplitude of the ERN/Ne depends on mood and personality variables. Luu, Collins, and Tucker [19] found large ERN/Ne amplitudes in college students who were high on negative affect (NA) and negative emotionality (NEM) in the beginning of an Eriksen flanker task. Moreover, a shift on response patterns was found during the experiment. By means of a post-task questionnaire parts of the subjects were reported to have been bored and dissatisfied with their performance resulting in motivational problems and disengagement from the task. When EEG data were re-analyzed for members of the high-NA and high-NEM groups with motivational problems, the amplitude of the ERN/Ne decreased. This pattern of results was strikingly different from results of participants who were low on NA and NEM. Similarly, Dikman and Allen [9] demonstrated that individuals low on socialization exhibit smaller ERN/Ne amplitudes during tasks which penalize error responses. In the same vein, Pailing and coworkers [23] found smaller ERN/Ne peak amplitudes in subjects with a tendency towards impulsive responding. Impulsivity was rated based on linear regression from correct individual reaction times on reaction times from erroneous responses. Mean RT residual scores were defined as mean difference of observed RTs (Ŷ) minus predicted RTs (Ŷ) for error trials (Σ(YŶ)/n). Less negative mean residual RTs were regarded as indicating a more cautious (controlled) response strategy whereas more negative residuals were interpreted to indicate a less controlled (i.e., more impulsive) response style. Furthermore, in their study, ERN/Ne latencies were positively related with percentage of errors, suggesting that individuals with shorter ERN/Ne latencies should have more opportunity to control for erroneous response tendencies [23].

Another ERP component discussed in the context of error processing is the error positivity (Pe), first described by Falkenstein and coworkers [10], [11]. The Pe is a slow positive wave with centroparietal distribution which usually follows the ERN/Ne in a time window between 300 ms and 500 ms after erroneous responses. The Pe has been differentiated from the P300 by some authors [12], whereas others interpret the Pe as a P300 on the erroneous response [7]. A source localization analysis using BESA revealed that the Pe consists of two components: an “early” Pe component with probable generators in an area around the caudal ACC and a “late” Pe component with probable generators in an area around the rostral ACC. The “early” Pe has been regarded as functionally belonging to the ERN/Ne [28], whereas the “late” Pe component was associated with awareness of erroneous responses and was more pronounced for perceived than for unperceived errors [20].

In the present study, we investigated ERPs related to errors of commission (i.e., pressing a button when one is not supposed to do in a Go/Nogo task) and correct responses (i.e., pressing a button when one is supposed to do so). Errors due to delayed response (“faster” as feedback) were excluded from ERP analysis. We analyzed the relationship between amplitudes and latencies of the three error-related ERP components (ERN/Ne, “early” Pe, and “late” Pe) and two behavioral indices of response control (RT residual values and error rates). Similar to the method originally introduced by Pailing et al. [23], subjects were split into a high (henceforth: HI) and low (henceforth: LI) impulsiveness group based on individual mean RT residuals. We reasoned that individuals with high impulsiveness (more negative RT residual values) should demonstrate smaller ERN/Ne amplitudes (less negative) and smaller “early” Pe amplitudes (less positive) than individuals with a more controlled response strategy (less negative RT residual values). Besides that, we expected longer ERN/Ne latencies and “early” Pe latencies in less controlled subjects. From their finding of a positive correlation between error rates and ERN/Ne latencies, Pailing et al. [23] reasoned that subjects with faster ERN/Ne's have a more controlled response strategy as they have more opportunity to catch erroneous intentions (see also [32]). With regard to the “late” Pe component, group differences on this component should indicate differences in the awareness of errors between HI and LI subjects as has been suggested by Nieuwenhuis and coworkers [20].

Section snippets

Participants

Thirty-two right-handed [21] healthy subjects (eleven males) with no history of neurological or psychiatric disorders took part in the study. After complete description of the study to the subjects, written informed consent was obtained. The study was approved by the local ethical committee and was in accordance with the Declaration of Helsinki. The entire group had a mean (SD) of 29.4 (10.9) years of age (range, 20–65) and a mean of 12.2 (1.7) years of education (range, 8–13). We calculated

Behavioral data

Given the task, only false positive responses on Nogo trials were of interest (commission errors). Consequently, error rates were individually calculated as number of false positive reactions during Nogo trials. For HI subjects, mean number of errors was 54.9 (SD: 24.4), corresponding to an error rate of 18.3%. LI subjects demonstrated a mean number of 40.8 errors (SD: 24.4), corresponding to an error rate of 13.6%. An ANOVA on the mean number of correct and incorrect trials including the

Discussion

In the present study, we used a Go/Nogo paradigm to investigate neurophysiological correlates of impulsiveness in healthy controls. Following a proposal by Pailing and coworkers [23], we calculated individual mean reaction time residuals as kind of scores in order to determine response control in participants. Using these RT residuals, the entire group (n = 32) was split into two subgroups with high (n = 16) and low impulsiveness (n = 16), respectively. Comparing performance data of HI and LI

References (32)

  • J.H. Woodcock

    A neuropsychiatric approach to impulse disorders

    Psychiatr. Clin. North Am.

    (1986)
  • G.E. Berrios et al.

    Abulia and impulsiveness revisited: a conceptual history

    Acta Psychiatr. Scand.

    (1995)
  • M.M. Botvinick et al.

    Conflict monitoring and cognitive control

    Psychol. Rev.

    (2001)
  • C. Carter et al.

    Anterior cingulate cortex, error detection, and the online monitoring of performance

    Science

    (1998)
  • P.T. Costa et al.

    Personality disorders and the five-factor model

    J. Pers. Disord.

    (1990)
  • Z. Dikman et al.

    Error monitoring during reward and avoidance learning in high- and low-socialized individuals

    Psychophysiology

    (2000)
  • Cited by (96)

    • Electrophysiology of goal-directed versus habitual control during outcome devaluation

      2019, Cortex
      Citation Excerpt :

      However, the ERPs elicited during learning could not predict the dN2 from the slips-of-action test, once again underscoring how N2 and ERN probably reflect different aspects of performance monitoring (Falkenstein et al., 1999; Larson, Clayson, & Clawson, 2014; Yeung & Cohen, 2006). Psychopathology and personality research suggests that variations in ERN amplitude are state-independent and rather related to trait differences (Boksem, Tops, Kostermans, & De Cremer, 2008; Boksem, Tops, Wester, Meijman, & Lorist, 2006; Hajcak et al., 2004; Hajcak, Franklin, Foa, & Simons, 2008; Hajcak, McDonald, & Simons, 2003; Hall, Bernat, & Patrick, 2007; Luu, Collins, & Tucker, 2000; Moser, Hajcak, & Simons, 2005; Potts, George, Martin, & Barratt, 2006; Ruchsow, Spitzer, Grön, Grothe, & Kiefer, 2005). This is in line with our ERN findings, which show that dERN amplitudes remain stable across tasks in the fabulous fruit game (see also Riesel, Weinberg, Endrass, Meyer, & Hajcak, 2013), and are suggestive of an overarching cognitive control trait.

    View all citing articles on Scopus
    View full text