Introduction

The ability to regulate emotions effectively is critical for well-being, as dysregulated affect is a feature of numerous psychiatric disorders (Gross & Jazaieri, 2014). Preliminary descriptions of neurocognitive mechanisms have been proposed for cognitive reappraisal (Ochsner, Silvers, & Buhle, 2012; Powers & LaBar, 2019), one clinically relevant strategy for managing emotions (Beck, Rush, Shaw, & Emery, 1979; Wilson 2008), but these mechanisms have not been thoroughly tested. To better support clinical models and treatments, it will be important to understand the cognitive and neural mechanisms of reappraisal as well as how they can be modulated.

One tactic of reappraisal is distancing (Ochsner et al., 2012), in which one simulates a new perspective to alter the psychological distance of an emotion-inducing stimulus (e.g., imagining oneself as an outside observer rather than the person directly involved) and, consequently, one’s emotional response. Distancing is especially well-suited for application in interventions given its effectiveness across a broad spectrum of clinical and nonclinical populations (Winecoff, LaBar, Madden, Cabeza, & Huettel, 2011; Lang et al., 2012; Gaebler, Daniels, Lamke, Fydrich, & Walter, 2014; Wang et al., 2014). Despite its effectiveness, little research has investigated the neurocognitive mechanisms that support distancing. Therefore, we recently reviewed the emerging literature on distancing to develop a model of these mechanisms (see Powers and LaBar, 2019 for a full description). This model proposes that distancing involves three key cognitive components: 1) self-projection, supported by the default mode network; 2) cognitive control, supported by the frontoparietal control network; 3) and affective self-reflection, particularly supported by the dorsomedial prefrontal cortex. A primary limitation of this model is that these associations are based largely on correlational neuroimaging data; however, targeted neurostimulation techniques can offer a stronger test of these components by modulating specific cortical targets in functional contexts.

Noninvasive neurostimulation can induce localized effects on cortical activity to temporarily alter function in neural circuits. Thus far, very little research has employed neurostimulation in the context of distancing. In one study, transcranial direct current stimulation was used to target dorsolateral prefrontal cortical function in the context of a general reappraisal technique that included distancing (Feeser, Prehn, Kazzer, Mungee, & Bajbouj, 2014). This study demonstrated both impairments and enhancements in reappraisal using different types of stimulation, putatively related to the target region’s role in cognitive control. That study’s focus on dorsolateral prefrontal cortex echoes the current intervention protocols for major depressive disorder using repetitive transcranial magnetic stimulation (rTMS; Perera et al., 2016). However, our model identifies another promising cortical target for investigating distancing using neurostimulation, the inferior parietal lobe, which has not yet been explored in this context.

The inferior parietal lobe is an important area for further study given that it has been implicated in both self-projection and cognitive control in distancing, and the current model does not discern more specific regional associations with these functions (Powers & LaBar, 2019). Nevertheless, some evidence suggests that these functions may be dissociable within the inferior parietal lobe. This evidence comes from fMRI comparisons of reappraisal and distraction. In the context of emotion regulation, distraction involves directing attention away from a stimulus to alter its emotional impact, often by redirecting attention towards more neutral or positive content. Distraction serves as a useful comparison tactic for distancing because it shares many of the same cognitive control demands, but it does not involve self-projection, or the ability to mentally simulate a perspective from another time, place, or person (Buckner & Carroll, 2007). Although distancing is supported by the inferior parietal lobe bilaterally (Powers, Graner, & LaBar, 2020; Powers & LaBar, 2019), previous comparisons of distraction and reappraisal suggest that, in the left hemisphere specifically, the temporoparietal junction (TPJ) may relate more to self-projection while more superior parietal regions may relate more to cognitive control (Dörfel et al., 2014; Kanske, Heissler, Schönfelder, Bongers, & Wessa, 2011, McRae et al., 2010).

As a first step in causally testing this role of the TPJ in distancing, we used rTMS to inhibit the function of this region during distancing. In addition, we tested whether this manipulation had an effect on distraction to assess whether any effects observed in distancing were more likely due to alterations in self-projection or cognitive control. Specifically, we hypothesized that rTMS to the TPJ would decrease distancing efficiency, defined as reduced valence changes in response to regulation, increased effort to implement regulation, or both. We further hypothesized that distraction efficiency would be unaffected by rTMS to the TPJ. This pattern of findings would support that the TPJ likely supports self-projection in distancing rather than more general cognitive control processes shared with distraction. This work helps to elucidate the function of the TPJ in distancing, while also exploring a novel cortical target for modifying emotion regulation.

Materials and Methods

Participants

Young adults (ages 18-39 years) were recruited through a community participant pool managed by the Brain Imaging and Analysis Center at the Duke University Medical Center as well as a classified advertisement website for Duke University affiliates. Exclusion criteria included contraindications for MRI and transcranial magnetic stimulation (TMS), history of psychiatric or neurological conditions, and current use of recreational drugs or psychoactive medications (Najib & Horvath, 2014), all assessed though self-report. Additionally, participants completed urine screenings for drugs that may affect seizure threshold and for pregnancy (females only).

Sample size was determined from a preliminary power analysis in G*Power (version 3.1; Faul, Erdfelder, Lang, and Buchner, 2007) based on a previous study targeting the left TPJ with 1-Hz rTMS, which disrupted performance on memory and episodic simulation tasks that share similar key characteristics with the present distancing study, including the use of self-projection (Thakral, Madore, & Schacter, 2017). Given the effect size in this study (d = 0.59), the power analysis prescribed a minimum sample size of 25 participants to achieve 80% power with α < 0.05 based on a two-tailed paired t test comparing the stimulated condition to a control condition. To account for some statistical noise in these procedures, a target of 30 participants was chosen and preregistered, and the final, analyzed sample consisted of 30 participants after exclusions.

For details concerning excluded participants and sample characteristics, see the Appendix. This experiment was undertaken with the understanding and written informed consent of each participant, and participants received $20 per hour. The study was approved by the Duke University Health System Institutional Review Board and preregistered on ClinicalTrials.gov (Identifier: NCT03698591).

Experimental Protocol

Figure 1 outlines the study protocol. Potential participants completed an online questionnaire, which included safety prescreenings for MRI and TMS. For qualified persons, two sessions were scheduled within a 2-week period. At the first session, participants completed an MRI safety screening, task training, and an anatomical MRI scan. At the second session, participants completed a TMS safety screening (Rossi, Hallett, Rossini, & Pascual-Leone, 2011), a task training refresher, and an active motor threshold (AMT) procedure to determine stimulation intensity (AMT was used for consistency with previous theta burst stimulation protocols; Huang, Chen, Rothwell, & Wen, 2007; Huang, Edwards, Rounis, Bhatia, & Rothwell, 2005). The specific rTMS protocol used was continuous theta burst stimulation (cTBS), which has been shown to reliably decrease oxygenation in the affected region and disrupt related cognition (Tupak et al., 2013). Participants completed two testing periods: one following active cTBS and one following sham cTBS (in a counterbalanced crossover design). In this paper, period refers to instances of task performance with respect to chronological order (i.e., period 1 refers to task performance after the first round of stimulation, irrespective of stimulation condition). Participants also completed three shortened and simplified versions of the experimental task (“baselines”), and a debrief interview.

Fig. 1
figure 1

Diagram of experimental protocol

Full size image

Task performance was carefully coordinated to occur during the window of maximum expected effect of cTBS (Huang et al., 2007, 2005). Task performance began at exactly 4 minutes post-cTBS (active and sham) and was completed at 16 minutes post-cTBS. Baselines were performed outside the window of expected effect for cTBS (at 30 minutes post-cTBS, Huang et al., 2005) and were included to assess the sufficiency of the washout period offline (see Appendix).

Active and sham cTBS were performed using a double-blind procedure. The experimenter who delivered stimulation and directed the participant through the experimental tasks exited the testing room during TMS setup and did not learn the true conditions until completion of the study. Participants were debriefed at the completion of the session to assess blinding efficiency (see Appendix).

Experimental Task and Training

Participants viewed 60 negative and 20 positive pictures from the International Affective Picture System (IAPS; Lang, Bradley, & Cuthbert, 2008). For negative pictures, participants were cued to use either a natural response technique (i.e., nonregulation; cue word “VIEW”) or one of two emotion regulation tactics: objective distancing (“OBJECTIVE”) or distraction (“DISTRACT”).Footnote 1 For VIEW, participants were instructed to let themselves experience any emotions they had in response to the pictures and to avoid regulating their emotions in any way. For objective distancing, they were instructed to view the image as if they were a neutral, objective observer at the scene. For distraction, they were instructed to direct their attention toward mentally rehearsing the digits of a previously memorized nine-digit number while looking at the picture (Dörfel et al., 2014). Adding to this previously used tactic, participants were advised that if the number ever became so well-rehearsed that it no longer seemed distracting, they should manipulate the number (e.g., by reordering or replacing digits) so that it again distracted their focus from the pictures. For distancing and distraction, participants were instructed that the goal of the tactic was to decrease emotional responses and that they should try to use the tactic to best achieve that effect. For positive pictures, participants were always cued to VIEW. Positive pictures were included to decrease the predictability of stimulus valence and to decrease the risk of participants developing a negative mood state. Thus, positive trials were not included in any main analyses, but descriptive statistics are still reported for this condition for comparative reasons (see Table 1 under Results).

Table 1. Descriptive summary of task efficiency. Means with standard deviations
Full size table

Refer to Figure 2 for a depiction of the trial structure. Immediately following stimulus presentation, participants rated their emotional valence on a scale from “very negative” (1) to “very positive” (7). For trials in which an emotion regulation tactic had been instructed, participants additionally rated how much effort it required from “very little effort” (1) to “very high effort” (7). Correlations between valence and effort ratings for the tactics tended to be very weak (Table 2), confirming that these ratings likely indexed separate processes. Participants completed a continuous set of 40 trials (10 VIEW-negative, 10 OBJECTIVE-negative, 10 DISTRACT-negative, and 10 VIEW-positive) during each testing period. Trials were presented in one of four pseudorandomized schemes (counterbalanced across participants), such that in every group of four consecutive trials, one trial of each type was presented. Details on task training are provided in the Appendix.

Fig. 2
figure 2

Schematic of the emotion regulation task

Full size image

MRI Acquisition

Scanning was performed on a 3T General Electric MR 750 system with an eight-channel head coil (General Electric Healthcare, Waukesha, WI). High-resolution images were acquired for neuronavigation of TMS using a 3D fast SPGR BRAVO pulse sequence with the following parameters: TR = 7.64 ms, TE = 2.936 ms, matrix = 256 x 256, flip angle = 12°, voxel size = 1 x 1 x 1 mm, 206 contiguous slices. These structural images were aligned to the near-axial plane defined by the anterior and posterior commissures.

TMS Procedures

RTMS was performed with a figure-8 coil (A/P Cool-B65) and a MagPro X100 stimulator with MagOption (MagVenture, Denmark) configured for biphasic pulses. Coil position was continuously monitored using a stereotactic neuronavigation system (Brainsight, Rogue Research, Canada). In the Brainsight software, an MNI-space template brain was registered to each participant’s structural MRI scan. This transformation was used to identify the TPJ target on each participant’s brain based on MNI coordinates (−53, −53, 23), and the corresponding projection onto the scalp. This target site was determined based on the peak group activation in TPJ related to objective distancing in an fMRI study using the same objective distancing task (Powers et al., 2020; Figure 3). The coil was positioned tangentially to the scalp and oriented with the handle directed perpendicular to the lateral sulcus and toward the top of the head.

Fig. 3
figure 3

A. TMS target in the left TPJ overlaid on a heat map of individual OBJECTIVE-negative > VIEW-negative maps from a previous fMRI study of distancing (Powers et al., 2020). The color scheme illustrates the number of individuals with suprathreshold results (cluster-based thresholding with a voxelwise, cluster-forming threshold of p < 0.01 and cluster-size threshold corresponding to p < 0.05) within each voxel. B. Graph illustrates that most individuals from the previous study had an objective distancing activation subpeak within 15 mm of the group peak (i.e., the current TMS target). Subpeaks were determined by applying the same thresholds for significance as the heat map, extracting subpeaks for all clusters using the cluster command in FSL (v5.0.9; https://fsl.fmrib.ox.ac.uk/fsl), and filtering the results for the closest subpeak to the target coordinates (up to a maximum of 40 mm in Euclidean distance). In addition, out of the participants with subpeaks around the group peak (n = 25), participants with stronger distancing effects on valence tended to have subpeaks closer to the group peak, r(48) = −0.47, p = 0.017

Full size image

To determine the AMT, electrodes were placed in a belly-tendon montage on the participant’s right hand to record activity in the first dorsal interosseous muscle. Motor evoked potentials were recorded in Brainsight. The motor hot spot was defined as the position over the left motor cortex that elicited the largest motor evoked potential. The AMT was then defined as the TMS pulse intensity that induced a motor evoked potential of at least 200 μV with 50% likelihood during weak voluntary contraction (20% of maximum voluntary contraction as measured with a pinch gauge; Oliviero et al., 2006) and was determined by using a maximum likelihood estimator (TMS Motor Threshold Assessment Tool, MTAT 2.0, http://www.clinicalresearcher.org/software.htm).

CTBS was performed with standard parameters: bursts of three pulses at 50 Hz delivered at a rate of 5 Hz, 300 pulses in total over 20 seconds, at 80% AMT (Huang et al., 2005). For the sham condition, the opposite face of the coil was positioned over the participant’s scalp to avoid directing a significant electric field into the brain. Using a very similar coil, this orientation has been found to reduce the induced electric field by 92% (Chou et al., 2015). Somatosensory features of the stimulation were mimicked using electrical stimulation via two electrodes placed approximately 1 cm apart on either side of the scalp target. These electrodes remained in place during both stimulation conditions.

Deviations from the planned stimulation protocol occurred for two participants who expressed discomfort at the onset of stimulation. In these cases, the intensity of stimulation was lowered to approximately 91% of original intensities. Both participants indicated that the stimulation was tolerable with these modifications, and the experiment proceeded normally after that point. Only two minor changes in the results of the analyses presented below were noted if these two participants were excluded (see Appendix for details).

Analyses

Summary of ratings by task condition.

First, general analyses of the valence and perceived effort ratings were performed to characterize the emotion regulation tactics on these outcome measures, confirm standard effects of emotion regulation irrespective of stimulation, and test for any general differences between the distancing and distraction tactics. Valence ratings were therefore averaged across both testing periods for OBJECTIVE-negative, DISTRACT-negative, and VIEW-negative trials for each participant, and effort ratings were averaged across the testing periods for OBJECTIVE-negative and DISTRACT-negative (effort was not rated for VIEW trials). A one-way repeated-measures ANOVA was performed to test for an effect of task condition on valence ratings, and paired t tests were performed to compare techniques on valence and effort ratings.

Effects of treatment and period

In the current study design, two tactics were applied during each of two testing periods, and two outcome measures were analyzed—change in valence relative to the VIEW-negative condition and effort. Valence differences were computed for each tactic and testing period as the individual’s mean rating for the regulation tactic minus their mean rating for VIEW-negative. More positive valence differences indicate more positive shifts in valence ratings for regulation versus VIEW-negative and stronger emotion regulation effects. Participants were also randomly assigned to one of two counterbalanced treatment order groups. To evaluate the comparability of these groups, valence ratings for the baseline VIEW-negative condition were compared between order groups for both the active and sham cTBS conditions (see Appendix and Table 1).

The primary data analyses assessing the effects of cTBS on distancing and distraction were performed using two statistical approaches. The first approach, which was planned in study preregistration, was based on repeated-measures ANOVA, with tactic (distancing or distraction) and treatment (active or sham cTBS) as within-subjects factors and temporal order (active-sham or sham-active) as a between-subjects factor. The second approach consisted of analytic procedures developed specifically for crossover study designs (Jones & Kenward, 2003; Wellek & Blettner, 2012). This second approach, which produced consistent results with the first, is documented in the Appendix. Finally, exploratory analyses were conducted to further test for treatment effects using the same repeated-measures ANOVA approach described above, but with the addition of absolute stimulator output intensity and scalp-to-cortex distance as covariates. These features were included to help model individual differences in the electric field delivered to the target site, because these differences could obscure treatment effects.

Results

Summary of Ratings by Task Condition

ANOVA results indicated a significant effect of task condition on valence ratings across the experiment, F(1.41, 41.00) = 43.52, p < 0.001 with Greenhouse-Geisser correction, η2 = 0.25 (Figure 4A; Table 7). Paired t tests indicated no effect of tactic on valence, t(29) = 0.04, p = 0.971, or effort ratings, t(29) = −0.23, p = 0.824 (Figure 4B), but valence ratings were higher (less negative) for OBJECTIVE-negative, t(29) = 7.71, p < 0.001, dav = 1.16, and DISTRACT-negative, t(29) = 6.76, p < 0.001, dav = 1.14, relative to VIEW-negative (all reported t tests are uncorrected). These results demonstrate that both tactics were effective at shifting valence in the positive direction, but they did not differ in terms of valence or effort.

Fig. 4
figure 4

Dot plots of valence and effort by task condition. Each participant’s mean rating per condition and measure is represented by a dot. Black dots represent means across all participants. View here refers to the VIEW-negative condition. Valence ratings were less negative for both distancing and distraction relative to view. ***p < 0.001.

Full size image

Effects of Treatment and Period

Planned analyses

Regarding changes in valence, ANOVA results indicated a significant three-way interaction of tactic*treatment*order, F(1, 28) = 6.16, p = 0.019. Visual inspection of tactic*treatment for each order revealed inverse patterns. Initially, this result would seem to indicate that the treatments had opposite effects in the two order groups. In this crossover design, however, treatment*order is aliased with period (i.e., the variance due to treatment*order and the variance due to period are not dissociable, although these terms indicate different conceptual interpretations). Therefore, the observed three-way interaction more likely captured an interaction of tactic*period rather than a reversal of treatment effects across groups. Indeed, a follow-up repeated-measures ANOVA of valence differences by tactic and period yielded a significant interaction effect, F(1, 29) = 5.91, p = 0.021, η2 = 0.02. This interaction was characterized by increased differences in valence for distancing relative to decreased differences for distraction moving from period 1 to period 2 (Figure 5A), although paired t tests separately evaluating valence differences across periods for distancing and distraction were not significant (both ps > 0.1). No main effect of treatment or interaction of tactic*treatment were found.

Fig. 5
figure 5

Box plots of changes in valence (A; VIEW refers to VIEW-negative) and effort (B) associated with regulation by task period. Panel A displays an interaction between period and tactic, such that the effect of distancing on valence increased across periods relative to a decreased effect of distraction. Panel B displays a simple main effect of period, where, for distancing, perceived effort decreased across periods. *p < 0.05

Full size image

In the repeated-measures ANOVA for effort, only a trending interaction of treatment*order was found, F(1, 28) = 3.91, p = 0.058. Again, visual inspection of the interaction pattern would have initially seemed to indicate a reversal of the treatment effects across the two order groups, but treatment*order is aliased with period in the crossover study design. Therefore, the treatment*order effect more likely reflected a period effect than the treatments having opposite actions in the two groups. Paired t tests of the effect of period for each tactic indicated a significant decrease in effort ratings across periods for distancing, t(29) = 2.69, p = 0.012, dav = 0.51 (Figure 5B), but not distraction (p > 0.1). No effects related to treatment were found (all ps > 0.1), as the trending interaction of treatment*order was explained by a period effect.

Exploratory analyses

Analyses including stimulator intensity and scalp-to-cortex distance as covariates also found no effects on changes in valence related to treatment (all ps > 0.1). Exploratory analyses of effort using these covariates, however, indicated a trend-level main effect of treatment on effort, F(1, 26) = 3.20, p = 0.085, reflecting increased effort following active cTBS relative to sham (stimulator intensity and scalp-to-cortex distance accounted for 2.19% and 0.08% of variance in effort, respectively). This effect of treatment was significant for distancing, F(1, 26) = 7.16, p = 0.013, η2 = 0.06 (Figure 6), but not distraction (p > 0.1), although the interaction of tactic*treatment was not significant (p > 0.1). Scalp-to-cortex distance was negatively correlated with the difference in effort from sham to active cTBS during distancing, r(28) = −0.36, p = 0.049 (i.e., cTBS increased effort more when the scalp-to-cortex distance was smaller and the effect of stimulation presumably greater), but stimulator intensity was not correlated with the difference in effort (p > 0.1).

Fig. 6
figure 6

Mean effort ratings by stimulation condition and task condition in exploratory analyses. Means and error bars (standard error of the mean) are corrected for covariates of stimulator intensity and scalp-to-cortex distance. Perceived effort of distancing was greater following active cTBS relative to sham cTBS. *p < 0.05

Full size image

Discussion

The primary goal of this investigation was to test the function of left TPJ in distancing, by selectively inhibiting this region with cTBS before an emotional regulation task. While planned analyses did not yield any effects of cTBS on distancing, exploratory analyses controlling for stimulation-related parameters revealed an increase in perceived effort specifically for distancing after neuromodulation. These findings partially support our hypotheses that cTBS to TPJ would reduce the efficiency of distancing but not distraction, and therefore suggest that the TPJ more likely supports self-projection in distancing, rather than distraction. Given that effort and valence ratings were generally uncorrelated, this cTBS manipulation seemed to specifically impact the effort of self-projection rather than its efficacy. Results further revealed improvements in distancing efficiency over repeated use, in contrast with distraction, offering novel insights into the temporal dynamics of these tactics.

Given the selective effect of cTBS on distancing, it is important to consider how the cognitive processes of the specific emotion regulation techniques in this study compare. Both techniques placed demands on cognitive control processes (McRae et al., 2010). In the case of distancing, cognitive control was required to maintain the regulatory goal and technique in working memory, monitor the progress of the regulatory goal, and potentially adapt implementation of the technique. The distraction technique shared these demands while additionally requiring the rehearsal and potential manipulation of a number in working memory. However, unlike distraction, distancing also required self-projection to simulate a neutral observer’s perspective. This difference in combination with the present results suggests a dissociation between self-projection and cognitive control in the left TPJ, whereby the left TPJ is specifically associated with self-projection in distancing. This interpretation is consistent with previous fMRI comparisons of distraction and reappraisal (Dörfel et al., 2014; Kanske et al., 2011, McRae et al., 2010), although it is possible that these techniques differed in aspects of cognitive control not considered here, which led to the observed results.

Although these treatment effects were consistent with our hypotheses, they only emerged in exploratory analyses. Furthermore, the distinction between distancing and distraction emerged in separate follow-up analyses of each tactic, but the interactive effect of treatment and tactic was not significant. These findings indicate both that the covariates included in the exploratory analyses may have contributed substantial variance (i.e., noise) to the initial analyses and that the observed treatment effect was relatively weak. One possible explanation for the weakness of this effect is that lateral parietal cortex outside of the stimulated area may have compensated for the disrupted tissue. Notably, distancing was characterized by both left and right TPJ activation in a previous meta-analysis (Powers & LaBar, 2019). Some work has suggested unique functional contributions for these two regions (Perner, Aichhorn, Kronbichler, Staffen, & Ladurner, 2007; Saxe & Wexler, 2005), but it is not yet clear whether their functions are separable in the context of distancing. Similarly, distancing-related activation around left lateral parietal cortex was widespread in the previous fMRI study that informed target selection. Therefore, cTBS effects may have been too focal to affect more robustly the relevant cortical function. Neurostimulation techniques with more diffuse effects (e.g., transcranial direct current stimulation) may be helpful in subsequent work. Furthermore, our targeting procedures may have lacked the precision to achieve more robust effects. We targeted an independent group-based fMRI target using individual anatomical MRI data, but individualized fMRI targets might have strengthened the observed effects.

Consistent with previous research, both tactics were successful in reducing negative valence (Kanske et al., 2011; Sheppes & Meiran, 2007; Webb, Miles, & Sheeran, 2012), and both tactics required comparable effort. However, closer examination revealed unique trajectories for distancing and distraction over the experiment. Specifically, repeated use of distraction led to a loss of efficiency versus a gain for distancing. Distancing required participants to engage in unique stimulus processing for each trial, whereas the same distracting content was processed repeatedly. Thus, the observed decline in the effect of distraction on valence may have resulted from diminishing cognitive control resources, habituation to the distracting content, or a combination of these factors. Indeed, previous literature on self-control suggests a degradation of control processes over short-term repeated use (Baumeister, Bratslavsky, Muraven, & Tice, 1998; Vohs & Faber, 2007). Because improvements were observed in distancing efficiency, any potential declines in cognitive control during this tactic may have been outweighed by positive practice effects on self-projection. These findings then suggest that self-projection may be prone to enhancement rather than depletion over short-term repeated use, making distancing and other cognitively related tasks amenable to practice and training, particularly in therapeutic contexts. These findings complement previous work that has suggested that reappraisal may have more adaptive long-term consequences relative to distraction (Kross & Ayduk, 2008; Thiruchselvam, Blechert, Sheppes, Rydstrom, & Gross, 2011). Nevertheless, distraction may still be preferable in certain contexts, such as brief situations of high emotional intensity (Sheppes, Scheibe, & Suri, 2011).

In addition to the potential limitations discussed above, other limitations include that this study was not designed to evaluate the temporal changes associated with distancing and distraction efficiency. More than two periods would have been ideal to better characterize the trajectories of emotion regulation effects. Also, the fact that the distraction technique did not incorporate novel distracting content for each trial, and therefore conflated the effects of habituation and changes in cognitive control over time, might be considered a limitation of this experimental design. Future work could attempt to dissociate the effect of habituation on distraction by changing the distracting content with each instance of regulation (e.g., with an arithmetic distraction task, Kanske et al., 2011) or explore the parameters needed to recover distraction efficiency, such as the duration of time between use.

Conclusions

This study found some causal evidence linking the left TPJ to self-projection processes in distancing. These findings help to clarify the functions of the inferior parietal lobe in the current model of distancing (Powers & LaBar, 2019) by more specifically characterizing the function of the left TPJ. Furthermore, this study revealed that the self-projection processes in distancing may be facilitated by repeated use. As a result, distancing may be particularly effective with frequent application, whereas distraction may be most effective with short-term use. Together, these findings help to refine our understanding of the neurocognitive processes of distancing and suggest ways to optimize the use of emotion regulation tactics in therapeutic contexts.