Acting against your own interests: The tension between emotion regulation preference and efficacy and its implications for individuals with depressive symptoms

Rotem Vered; Shilat Haim-Nachum; Einat Levy-Gigi

doi:10.1371/journal.pone.0254213

Abstract

The aim of this study was twofold: first, to compare individuals’ strategy choices in low and high intensity conditions and the actual efficacy of these strategies; second, to assess whether and how perceived intensity levels of aversive situations moderate the relationship between depressive symptoms and a strategies’ efficacy. In Experiment 1A (N = 58), we replicated previous results, showing that individuals prefer distraction in high- and reappraisal in low-intensity conditions, irrespective of depressive symptom levels. Experiment 1B (N = 50) assessed the efficacy of distraction and reappraisal strategies in aversive conditions with low and high intensity. Contrary to our prediction, reappraisal was more effective than distraction, independent of the intensity of the aversive conditions. In Experiment 2 (N = 113), we tested the interactive relationship between perceived intensity levels and depression on the relative effectiveness of reappraisal and distraction. We found that while in perceived low-intensity situations the advantage of distraction over reappraisal increased as depressive symptoms increased, no such relationship was found in high-intensity situations. The results suggest that while all individuals prefer to apply reappraisal in both low- and high-intensity conditions, for those with high level of depressive symptoms, such a preference acts against their own interests. The study highlights the need to distinguish between emotion regulation preferences and their actual efficacy, while illuminating possible implications for individuals with depressive symptoms.

Citation: Vered R, Haim-Nachum S, Levy-Gigi E (2021) Acting against your own interests: The tension between emotion regulation preference and efficacy and its implications for individuals with depressive symptoms. PLoS ONE 16(7): e0254213. https://doi.org/10.1371/journal.pone.0254213

Editor: Anthony Charles Ruocco, University of Toronto at Scarborough, CANADA

Received: September 21, 2020; Accepted: June 22, 2021; Published: July 12, 2021

Copyright: © 2021 Vered et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information.

Funding: The study was supported by the Israel Science Foundation Grant# 1128_16 to ELG.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Do we always choose what is right for us? Modern society supports consumer sovereignty, free marriage and democratic elections, all of which reflect the fundamental assumption that individuals are able to choose what is best for them. However, empirical investigations suggest that this is not always the case; in many instances, individuals fail to accurately predict which option will provide them with the best result [1–4]. For example, when individuals were asked to predict the pleasure they would receive from eating ice cream, there was no correlation between their predictions and their actual experience [5]. However, studies on emotion regulation tend to assume that individuals choose emotion regulation strategies that serve them the best [6, 7]. The first aim of the present study was to assess both regulatory preferences and their actual efficacy in reducing distress. The second aim was to test the effectiveness of different regulatory strategies as a function of intensity level and depressive symptoms.

Emotion regulation is defined as a range of activities that allow individuals to monitor, evaluate, and modify the nature and course of emotional responses in order to react appropriately to environmental demands [8, 9]. Two types of regulatory strategies that have been extensively studied are disengagement strategies, such as distraction, and engagement strategies, such as cognitive reappraisal. Distraction involves directing attention away from emotional information by producing neutral thoughts before an initial appraisal has been made. Reappraisal involves attending to emotional information and reinterpreting its negative meaning in order to reduce distress after an initial appraisal has been made [8, 10, 11].

According to the traditional approach, engagement strategies have been considered more effective in downregulating distress compared to disengagement strategies [12–16]. However, a more recent approach suggests that different regulatory strategies are adaptive depending on the context in which they are applied [10, 17–22]. For example, in conditions of high levels of peer victimization, habitual reappraisal is adaptive, while in conditions of low levels of peer victimization, it predicts maladaptive physiological reactivity [23].

According to the process-specific timing hypothesis, the intensity of the regulated condition and the timing of the regulation attempt determine whether a certain strategy will be adaptive or not [24, 25]. A further examination reveals that distraction is more effective in high- compared to low-intensity conditions [6, 26], since it occurs early in the emotion regulation process and does not involve engaging with the aversive situation. Conversely, reappraisal is more effective in low- compared to high-intensity conditions, since it occurs after an emotion has gained strength and requires more cognitive resources, which are more available under low-intensity conditions [27–30].

The process-specific timing hypothesis is widely supported by studies that allow participants to choose between distraction and reappraisal in low- and high-intensity conditions [6, 29, 31, 32]. These studies reveal that when given a choice, participants prefer distraction under high-intensity situations and reappraisal under low-intensity situations. However, it is not yet clear whether these preferences reflect the effectiveness of these strategies. The aim of Experiment 1A was to test distraction and reappraisal preferences in low and high intensity conditions, whereas the aim of Experiment 1B was to test the actual efficacy of distraction and reappraisal in these conditions.

Other factors that may influence the efficacy of different strategies include the personal characteristics and psychological state of the regulator [19, 33]. One such example is depression, which has a significant effect on perceiving and processing negative emotional events and may interrupt the adaptive emotional response to aversive situations [34–36]. The negative potentiation hypothesis posits that depressed individuals have a greater tendency to be aware of and fixate on adverse notions or elements in their lives [37–40]. Relatedly, depressive individuals are prone to excessive negative rumination–non-adaptive, constant focus on mundane negative thoughts [38, 41–43]. As such, they may benefit more from using strategies that do not require them to control their attention and thoughts and that allow them to avoid processing the aversive stimuli, since the processing itself can lead to rumination [44]. The aim of Experiment 2 was to investigate whether perceived intensity levels of emotional stimuli moderate the relationship between depressive symptoms and the efficacy of distraction and reappraisal in reducing distress.

Experiment 1A

Experiment 1A sought to replicate previous findings and test individuals’ preferences for distraction or reappraisal in aversive conditions of low and high intensity. We therefore hypothesized that individuals would favor distraction in high-intensity conditions and reappraisal in low-intensity conditions. We further hypothesized that depressive and state/trait anxiety symptoms would not affect these preferences [32].

Materials and methods

Participants

We used G*Power software to determine a sufficient sample size given an alpha of 0.05, a power of 0.95, and a medium effect size (f = 0.25) [45]. We therefore recruited 58 college students (for a detailed description of the sample in all three experiments see Table 1), who participated in this experiment in exchange for course credit.

Download:

Table 1. Demographic characteristics of the participants in the three experiments (means and standard deviations/frequency).

https://doi.org/10.1371/journal.pone.0254213.t001

All participants were White Israelis with at least 12 years of education and several depressive symptoms. They were recruited through flyers posted in a university setting. Applicants were invited to participate in a screening interview, if they: (i) were 18–45 years of age; (ii) had accurate vision. To reduce confounds related to concurrent disorders, we used the Structured Clinical Interview for DSM-5 (SCID-5-CT) [46] to exclude participants with any current DSM-5 psychopathology according to the following exclusion criteria: (i) present or previous diagnosis of psychiatric disorders; (ii) danger of fatal harm to self or others; (iii) past experience of concussion or other medically relevant condition such as head injury, loss of consciousness for over ten minutes, or neurological conditions including epilepsy, multiple sclerosis, stroke or encephalitis. None of the participants met these criteria. Authorization was obtained from the Ethics Committee School of Education Bar-Ilan University. Participants read and signed a written informed consent form and were told that they could quit at any point during the experiment.

Measures and procedure

In both of the two experiments, we used well-validated performance-based paradigms in which participants had to either choose between two strategies or to apply a given strategy. Based on previous studies that set criteria for low-and high-intensity and showed differential regulatory patterns in these two categories, we selected negative images of high and low intensity [6, 47]. In order to verify the correct use of the strategies, a random sample in each experiment was asked to type one sentence that described the way they employed the strategies in selected trials from each condition (distraction/reappraisal and low/high intensity). Participants wrote one sentence describing their strategy use following 25% of the trials. These sentences were then reviewed blind (with the trial’s focus on strategy and intensity remaining undisclosed) and coded for reappraisal and distraction by a clinical psychologist at a post-doctoral level.

The performance paradigm includes 30 images–all negative–from the International Affective Picture System (IAPS) [48]. The IAPS provides an official normative rating of arousal (1 –low; 9-high) and valance (1- very unpleasant; 9 –highly pleasant). The images are divided into two sets: 15 images with low (M arousal = 5.01; M valence = 3.41) and 15 images with high (M arousal = 6.12; M valence = 1.99) intensity. Low and high arousal images are statistically distinguishable across valence ratings (all ps < .05). The contents of the pictures are roughly matched across the different intensities [see 6]. The paradigm includes two phases: In the practice phase (four practice trials), participants learn to employ distraction (i.e. thinking about something that is emotionally neutral) and reappraisal (i.e. thinking about each picture in a way that reduces its negative meaning). In the following experimental phase, they freely choose between distraction and reappraisal for a total of 30 trials.

Each of these 30 trials begins by showing study participants a fixation cross. This is followed by a quick (500 ms) sample of a picture intended to provoke an emotional response. After the offset of the picture, a computer program issues instructions on the screen asking participants to choose one of two strategies—distraction or reappraisal. The picture then re-appears for an extended duration (5,000 ms) so that participants can implement their choice. To conclude the trial, participants use a Likert scale to assign a number to the degree of distress they felt, ranging from 1 = no distress at all to 9 = great distress. The order of the trials is randomized across participants yet ensures that no picture is presented in sequence.

The coder ratings of the sentences describing strategies use were in nearly perfect accord with the strategies utilized by the participants (99.3%). There were no significant differences between strategy effectiveness levels of the emotion regulation process for participants who wrote down these descriptions and participants who did not (all Fs < 1), suggesting that typing out these descriptions did not influence strategy effectiveness levels.

Two questionnaires were used in order to test and control for possible effects of depression and anxiety: (1) The 21-item Beck Depression Inventory (BDI–II) [49] was used to assesses symptoms of depression over the two weeks–instead of the past week–prior to the experiment (internal consistency in the current study α = .89). Each item is measured on a scale from 0 to 3, with total scores ranging from 0 to 63; higher scores indicate greater levels of depression. (2) The State-Trait Anxiety Inventory (STAI) [50], a 40-item questionnaire (internal consistency in the current study α = .92 and .90, for state and trait anxiety, respectively) was used to evaluate individuals’ state and trait anxiety levels. Items for both state and trait anxiety are rated on a 4-point scale, with scores range from 20 to 80; higher scores reflect greater anxiety.

Statistical analyses

In both Experiment 1 (A and B) and Experiment 2, statistical analyses were performed with SPSS version 25 (Chicago, IL). Descriptive analyses were used to calculate the demographic information of the sample. Pearson correlations were used to test associations between different experimental conditions. Repeated-measures ANOVA was used to test strategy choice preferences under low and high intensity conditions.

Results and discussion

Zero-order correlations are presented in Table 2. We found no significant correlations between clinical measures (depression, anxiety) and experimental conditions (all ps > .05), suggesting that strategy preferences are not related to depressive and anxiety symptoms.

Download:

Table 2. Zero-order correlations between depression, anxiety, and the different conditions in Experiment 1A.

https://doi.org/10.1371/journal.pone.0254213.t002

In order to test possible differences in individuals’ regulatory preferences in reducing distress, we conducted a strategy (distraction vs. reappraisal) by intensity (low vs. high) repeated-measures ANOVA. As predicted, and in accordance with previous research, we found that in low-intensity conditions, participants preferred to employ reappraisal on 77.7% of the low-intensity trials (95% CI: [72.83, 82.57]), whereas in high-intensity conditions, they preferred to employ distraction on 63.48% of the trials (95% CI: [50.78, 63.48), F(1, 56) = 29.08, p < .001, = .34. These preferences were not affected by depressive symptom levels, F(1,56) = .86, p = .36.

Experiment 1B

In this experiment, we compared the actual effectiveness of reappraisal and distraction in reducing distress in aversive conditions with low and high intensity. According to the process-specific timing hypothesis [24], we predicted that distraction would reduce distress more successfully under high-intensity conditions, and reappraisal would reduce distress more successfully under low-intensity conditions.