Modulation of the error-related negativity by induction of short-term negative affect
Introduction
Affect influences how we perceive and act upon our environment. It has been shown that positive affect increases problem solving (Isen, 2001, Isen et al., 1987), memory performance (Lee, & Sternthal, 1999), executive attention (Ashby, Valentin, & Turken, 2002), decision making (Isen, 2001) and a variety of other cognitive tasks, while the influence of negative affect is more diffuse and difficult to predict (Ashby, Isen, & Turken, 1999; Mitchell, & Phillips, 2007). The present investigation is concerned with the affective modulation of action monitoring. In particular, we are interested in the affective modulation of a specific component in the event-related brain potential (ERP), the error-related negativity (ERN or Ne) (Falkenstein, Hohnsbein, Hoormann, & Blanke, 1991; Gehring, Goss, Coles, Meyer, & Donchin, 1993). It is observed with a frontocentral distribution and a peak-latency between 50 and 100 ms after the commission of an error. Electrophysiological (e.g., van Veen and Carter, 2002a, van Veen and Carter, 2002b) and brain imaging techniques (Marco-Pallares, Müller, & Münte, 2007; Ridderinkhof, Ullsperger, Crone, & Nieuwenhuis, 2004) agree that the ERN emerges in the anterior cingulate cortex (ACC).
Long-lasting emotional and motivational factors have also been shown previously to modulate ERN amplitude: It is increased in subjects suffering from increased anxiety and negative affect (Hajcak et al., 2003a, Hajcak et al., 2004). In an early study Luu, Collins, and Tucker (2000) similarly found that college students who were high on negative affect and negative emotionality displayed larger ERN amplitudes compared to participants who displayed low negative affect and emotionality. This effect decreased, however, over the course of the experiment, which was interpreted as signifying disengagement of the high negative affect participants from the task. More globally, the results were interpreted as evidence for an interaction between affect and associated behavioral patterns with frontal lobe executive functions. In patients suffering from obsessive-compulsive disorder who likewise display a high negative affect, several studies have reported an increased ERN amplitude (Gehring, Himle, & Nisenson, 2000; Hajcak, & Simons, 2002; Johannes et al., 2001a, Johannes et al., 2001b; Münte et al., 2008). There are also some indirect indications that long-lasting positive affect modulates ERN amplitude: Alcohol, which induces pleasant feelings, and oxazepam, a benzodiazepine derivative with anxiolytic properties reduce ERN amplitude (Johannes et al., 2001a, Johannes et al., 2001b, Riba et al., 2005a, Riba et al., 2005b; Ridderinkhof et al., 2002). These pharmacological studies have to be interpreted with caution, however, as a direct action of these agents (e.g., by enhancing the activity of cortical interneurons) on the ERP amplitudes rather than an indirect action via their influence on affective factors is a possibility. To summarize, stable and replicable influences of negative and – to a lesser degree – positive affect on action monitoring indexed by the ERN have been shown.
The influence of short lasting affective modulations has been less intensively studied and has led to contradictory and counterintuitive results. Larson, Perlstein, Stigge Kaufman, Kelly, and Dotson (2006) presented a flanker stimulus superimposed on neutral, unpleasant or pleasant pictures taken from the International Affective Picture System (IAPS) and, in contrast to what has been observed with longer lasting affective states (see above), observed an increase in ERN amplitude for pleasant compared to neutral and unpleasant background pictures. The authors interpreted this finding as possibly indicating a mismatch between the positive affect induced by the background picture and the negative event constituted by the error response. Because of the long presentation of the IAPS pictures and the fact that the flanker stimuli were presented during the presentation of background pictures, an alternative explanation might be that the unpleasant background pictures might have drawn attention away from the flanker stimulus and thus might have precluded an increase of ERN amplitude. Anxiety induced by a tarantula spider close to the phobic subject did not alter ERN amplitude (Moser, Hajcak, & Simons, 2005). However, if negative affect is induced by external stimuli, subjects allocate attentional resources towards the threatening cue. This might decrease error significance and counteract fear-induced effects.
In the present study we reexamined this issue by presenting pleasant, unpleasant and neutral IAPS pictures immediately prior to flanker stimuli for which the subjects had to perform a choice reaction time task1.
In light of the previous studies on the influence of longer lasting negative (Hajcak et al., 2003a, Hajcak et al., 2004, Luu et al., 2000) and positive (Johannes et al., 2001a, Johannes et al., 2001b, Riba et al., 2005a, Riba et al., 2005b, Ridderinkhof et al., 2002) affect on ERN and action monitoring, we formulated two hypotheses:
The first hypothesis stated that ERN amplitude to performance errors should be increased following the presentation of unpleasant IAPS pictures. The second hypothesis stated that ERN amplitude should be decreased after the presentation of pleasant IAPS pictures.
The primary focus of the experiment is thus on the (possible) influence of the valence of the preceding IAPS picture on the amplitude of the ERN. A second prominent effect in the flanker task is the so-called N2 component present in stimulus-locked averages that differentiates incongruent (HHSHH, SSHSS) from congruent (SSSSS, HHHHH) stimuli and has been interpreted as an index of response conflict (van Veen and Carter, 2002a, van Veen and Carter, 2002b; but see Wendt, Heldmann, Münte, & Kluwe, 2007). Like the ERN, this component has been attributed to the medial prefrontal cortex. Therefore it was of interest whether any valence-induced changes in the ERN might also be found for the N2 component.
Since the response to emotional stimuli has been shown to vary as a function of participants’ gender and to be weaker in men (Campbell et al., 2002, Kemp et al., 2004; Wrase et al., 2003), only women were enrolled in the present experiment.
Section snippets
Participants
Twenty-two women (mean age 24 years, range from 19 to 36) contributed data to the behavioral and ERP analyses. Two additional subjects were excluded because they committed not enough errors to generate reliable ERPs for erroneous responses. Four further subjects were excluded due to excessive and uncorrectable artifacts. Consent was given and subjects were reimbursed (7 € per hour).
Stimuli and procedure
A trial comprised the following sequence (timing is provided in brackets): fixation cross (varying duration
Behavioral data
Reaction time data are illustrated in Fig. 1 (upper panel). Erroneous responses were faster than correct responses in both experimental halves (correctness F (1,21) = 94.5, p < 0.001; correctness × half n.s.). Participants were faster in the first compared the second error half (F (1,21) = 31.5, p < 0.001).
There was a small, but significant main effect of valence: participants responded somewhat slower after seeing unpleasant pictures (valence: (F (2,42) = 5.20, p < 0.01), especially when responses were
Discussion
The main result of the present investigation is that affective information delivered by IAPS pictures modulated ERN amplitude especially during the first part of the experiment. This result extends previous findings that longer lasting negative affect can interact with executive processes, in particular action monitoring (Hajcak, & Simons, 2002; Hajcak et al., 2003a, Hajcak et al., 2004, Luu et al., 2000) by showing that such interactions might take place on a very short time scale and even on
Acknowledgements
We would like to thank Nadine Strien and Peggy Tausche for help in data recording and analysis. Research of JR, TG and TFM is financially supported by various grants from the German Research Agency (DFG); TFM is also supported by the Volkswagen-Stiftung.
References (61)
The central autonomic network: Functional organization, dysfunction, and perspective
Mayo Clinic Proceedings
(1993)- et al.
The locus coeruleus-noradrenergic system: Modulation of behavioral state and state-dependent cognitive processes
Brain Research Reviews
(2003) - et al.
The classification of ‘fear’ from faces is associated with face recognition skill in women
Neuropsychologia
(2002) - et al.
Repetitive picture processing: Autonomic and cortical correlates
Brain Research
(2006) - et al.
Anxiety and error monitoring: Increased error sensitivity or altered expectations?
Brain and Cognition
(2007) - et al.
Brain potentials in affective picture processing: Covariation with autonomic arousal and affective report
Biological Psychology
(2000) - et al.
Effects of crossmodal divided attention on late ERP components. II. Error processing in choice reaction tasks
Electroencephalography and Clinical Neurophysiology
(1991) - et al.
ERP components on reaction errors and their functional significance: A tutorial
Biological Psychology
(2000) - et al.
Error-related brain activity in obsessive–compulsive undergraduates
Psychiatry Research
(2002) - et al.
Anxiety and error-related brain activity
Biological Psychology
(2003)