Counter-regulation triggered by emotions: Positive/negative affective states elicit opposite valence biases in affective processing

Abstract

The present study investigated whether counter-regulation in affective processing is triggered by emotions. Automatic attention allocation to valent stimuli was measured in the context of positive and negative affective states. Valence biases were assessed by comparing the detection of positive versus negative words in a visual search task (Experiment 1) or by comparing interference effects of positive and negative distractor words in an emotional Stroop task (Experiment 2). Imagining a hypothetical emotional situation (Experiment 1) or watching romantic versus depressing movie clips (Experiment 2) increased attention allocation to stimuli that were opposite in valence to the current emotional state. Counter-regulation is assumed to reflect a basic mechanism underlying implicit emotion regulation.

Keywords:

• Counter-regulation
• Affective processing
• Valence biases
• Automatic attention allocation
• Emotional Stroop task
• Implicit emotion regulation

Acknowledgments

The research reported in this article was supported by a grant of the Deutsche Forschungsgemeinschaft to Klaus Rothermund and Dirk Wentura (DFG RO 1272/2-3).

Notes

¹Methodologically, the F-test for the interaction is equivalent to a t-test with emotion condition as the independent variable and the difference between error rates for positive and negative target trials as the dependent one. Thus, given our specific predictions, a one-tailed test is recommended in order to increase the power of the test (Maxwell & Delaney, 1990, p. 144).

²To ensure that that the effects of emotion induction on affective processing were independent of specific semantic relations, we conducted a control analysis in which all trials of the search task were discarded in which the presented word was semantically related to the specific future event that had been described by the individual participant (e.g., the target word “success” would not just be affectively but also semantically related to the event of receiving a good grade in an exam; on average, this was the case for 8% of all target trials). Eliminating these trials from the analyses did not change the results; in particular, the significant two-way interaction of Target Valence and Emotion Condition was significant again for the error data, F(1, 47)=4.33, p=.043, .

³One elderly participant (62 years) and one non-native speaker claimed to have difficulties in understanding and implementing the instructions. Another participant admitted that she had had a severe emotional experience this day and failed to follow the emotional movie clip. The pattern of findings is highly similar, however, if these participants are included in the analyses.

⁴An additional analysis revealed that this finding did not interact with the group factor that determined the type of affect measure that was applied for the manipulation check analyses (PANAS vs. IPANAT), F<1.

⁵One might raise the concern that the obtained results simply reflect mere mismatch-induced salience effects. In a recent study, Rothermund et al. (2011) disentangled effects of functional counter-regulation and mere salience. In their study, incongruency effects in affective processing were only evident after inducing an affective-motivational focus, but not if positive or negative information had to be maintained in working memory, which speaks in favour of a motivational explanation of the incongruency effect and rules out an interpretation in terms of mismatch-induced salience. We did not include a non-motivational baseline condition in the present study, but the previous findings by Rothermund et al. (2011) support our hypothesis that incongruent effects of emotion/motivation on attention reflect functional regulation processes rather than merely perceptual salience effects.

⁶Another important difference between some of the previous studies and our experiments relates to the assessment of biases in affective processing: Wadlinger and Isaacowitz (2006) assessed eye movements rather than measuring attentional capture and interference effects directly. Although eye movements are closely related to visual attention, recent research has shown that the motor programming of eye movements is independent from attentional engagement and disengagement effects that were assessed in the search task and emotional Stroop task of our study: Processes of endogenously cued attention allocation can take place in the absence of saccade programming (Belopolsky & Theeuwes, 2012; Smith, Schenk, & Rorden, in press). In addition, attention and saccade generation represent distinct functions that are located in separate brain areas (Lee, Ahn, & Keller, 2012; Wardak, Olivier, & Duhamel, 2011). Becker and Leinenger (2011) adapted a variant of the inattentional blindness (IB) paradigm to assess attention for valent information. Detecting or not detecting an unanticipated object in a visual scene in the IB paradigm is heavily influenced by top-down processes, that is, the unexpected object is detected much more easily if it shares a feature with those objects that are actively monitored during the task (e.g., Most, Scholl, Clifford, & Simons, 2005). Differences in detection rates in the IB paradigm thus probably reflect strategic search, which might be more prone to congruency effects, rather than indicating bottom-up effects of automatic attentional capture, which are assumed to reveal incongruency effects due to automatic counter-regulation. In addition, the paradigm that was employed in the study by Becker and Leinenger (2011) did not allow for an assessment of attentional biases within participants because only stimuli of one valence were presented to each participant.