- Email: [email protected]
Brief mindfulness induction could reduce aggression after depletion
Consciousness and Cognition 33 (2015) 125–134
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Brief mindfulness induction could reduce aggression after depletion Cleoputri Yusainy a,⇑, Claire Lawrence b a b
Psychology Programme, Faculty of Social and Political Sciences, Brawijaya University, Malang, Indonesia Personality, Social Psychology, and Health (PSPH) Group, School of Psychology, University of Nottingham, Nottingham NG7 2RD, UK
a r t i c l e
i n f o
Article history: Received 12 April 2014
Keywords: Mindfulness induction Self-control Ego-depletion Aggressive behaviour
a b s t r a c t Many experiments have shown that one’s ability to refrain from acting on aggressive impulses is likely to decrease following a prior act of self-control. This temporary state of self-control failure is known as ego-depletion. Although mindfulness is increasingly used to treat and manage aggressive behaviour, the extent to which mindfulness may counteract the depletion effect on aggression is yet to be determined. This study (N = 110) investigated the effect of a laboratory induced one-time mindfulness meditation session on aggression following depletion. Aggression was assessed by the intensity of aversive noise blast participants delivered to an opponent on a computerised task. Depleted participants who received mindfulness induction behaved less aggressively than depleted participants with no mindfulness induction. Mindfulness also improved performance on a second measure of self-control (i.e., handgrip perseverance); however, this effect was independent of depletion condition. Motivational factors may help explain the dynamics of mindfulness, self-control, and aggression. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction 1.1. Mindfulness, self-control ego-depletion, and aggression ‘‘Whose mind stands like a mountain. Steady, it is not perturbed. Unattached to things that arouse attachment. Unangered by things that provoke anger. When his mind is cultivated thus. How can suffering come to him?’’ (Udana 4.4, in Anonymous, 2012). At any given moment, the contents of one’s thoughts and emotions may be painful, pleasant, or neutral. Mindfulness — translated from the Pali (the language in which early Buddhist teaching was recorded) word, sati (‘memory’) — is a practice that acknowledges the contents of consciousness as states that arise and pass, but do not necessarily need action (Bodhi, 2000; Sujiva, 2000). The result of this practice is a mental gap between consciousness and its contents, or between attention and its objects (Baer, 2003; Brown, Ryan, & Creswell, 2007a,b; Shapiro, Carlson, Astin, & Freedman, 2006). Central for the present study is the idea that mindfulness may reduce an individual’s reactivity that is typically associated with aggression-triggering events. The current literature has documented the application of mindfulness-based interventions for treatment of aggression in mental health and forensic settings (see Fix & Fix, 2013; Shonin, Gordon, Slade, & Griffiths, 2013). However, much less aggression research has been conducted on the effect of brief mindfulness induction as opposed to the context of extensive ⇑ Corresponding author. E-mail address: [email protected] (C. Yusainy). http://dx.doi.org/10.1016/j.concog.2014.12.008 1053-8100/Ó 2014 Elsevier Inc. All rights reserved.
126
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
training in mindfulness. Preliminary support for the benefit of laboratory inductions of mindfulness on provoked aggression has been provided by Heppner et al. (2008). In their study, participants were randomly assigned to one of three conditions in a social rejection task: acceptance from partner, rejection by partner, or rejection plus mindfulness (i.e., a 5-min task of eating raisins while being focused in a mindful way on the experience of eating the raisin). In a subsequent aggression task using the administration of aversive noise blasts as the aggressive behaviour, those in the rejection plus mindfulness condition behaved less aggressively than those in the rejection only condition, and their behaviour did not differ from those in the non-rejection condition. We expand on this method by looking at prior self-control behaviour rather than social rejection as a possible precursor to aggression. The importance of trait self-control on the link between mindfulness and aggression has been reported in a recent correlational study (Yusainy & Lawrence, 2014). Even though trait self-control appears to be a highly desirable quality, a momentary act of self-control may be costly. As suggested in ‘‘test-operate-test-exit’’ theory (TOTE, see Carver & Scheier, 1982), effective self-control requires a continuous monitoring of the self’s current state against some desirable goals or standards. Once a person has detected a discrepancy between current and desired states, some operations to change the self can be initiated. Much like a muscle, however, these operations are proposed to rely on the availability of a common and limited pool of ‘resources’, or self-control strength (Baumeister, Vohs, & Tice, 2007). After performing initial acts requiring selfcontrol, the capacity for further control, even on unrelated self-control tasks, is temporarily decreased (i.e., ‘ego-depletion’: a meta-analysis of this model is provided in Hagger, Wood, Chris, & Chatzisarantis, 2010). Experimental research has shown that compared to non ego-depleted participants, depleted participants (who had previously resisted eating tempting food or stifled physical and emotional reactions) delivered more negative ratings to an experimenter that had previously insulted them (Stucke & Baumeister, 2006; see also DeWall, Baumeister, Stillman, & Gailliot, 2007). Participants also behaved more aggressively toward intimate partners had who ostensibly provoked them with negative feedback when they were depleted than when they were not (Finkel, DeWall, Slotter, Oaten, & Foshee, 2009). These studies conclude that under depletion conditions, subsequent attempts at controlling aggressive impulses become more prone to failure. Aggression is, arguably, a prime example of self-control failure. Unlike self-controlled attention, which encompasses both monitoring and controlling the self in order to achieve some future goals, mindfulness serves solely to monitor the contents of consciousness (Brown et al., 2007a,b). It has been demonstrated elsewhere that the neural system that monitors the mismatches between current and desired states (ERN: errorrelated negativity) is weakened after performing an initial self-control act (Inzlicht & Gutsell, 2007). As individuals who are mindful make no direct attempts to modify their experiences associated with being depleted, they become better attuned to when self-control is at stake (Teper, Segal, & Inzlicht, 2013). Indeed, greater self-control in terms of increased ERN and heightened emotional acceptance has been reported in experienced mindfulness meditators (Teper & Inzlicht, 2013). Of particular relevance is Friese, Messner, and Schaffner’s (2012) experiment with participants of a 3-day introductory seminar on mindfulness meditation. They showed that participants who had engaged in a 5-min period of mindfulness meditation after an emotion suppression task performed better on a subsequent self-control task (i.e., the crossing-out-letters task to discriminate between adjacent and similarly looking, but slightly different letters) than those in the suppression-only condition, and performed equally well as those not suppressing emotion (where participants were asked simply to watch the same movie clip). In the same way, mindfulness may act as a counteracting variable for the effect of depletion on aggression. It should be noted, however, that many measures of self-control exist, and many do not correlate highly (see Duckworth & Kern, 2011). While resisting aggressive impulses requires self-control, the direct effect of depletion on aggression, and the effect of mindfulness on post-depletion aggression may not be observed in other self-control tasks. To test this possibility, a frequently used behavioural measure of self-control (i.e., handgrip perseverance: Hagger et al., 2010) is included as a second dependent variable. As a whole, therefore, the current study examines the moderating effect of mindfulness induction on the link between ego-depletion and two consecutive self-control measures: one in the domain of aggression and one in the domain of physical stamina. In this way, we will be able to investigate whether the effects of brief mindfulness induction are consistent across different tasks of self-control. 1.2. The current study In the current study, aggression is provoked using Lawrence and Hutchinson’s (2013, 2014) adapted version of the Taylor Competitive Reaction Time (TCRT: Taylor, 1967) task. In typical work using the TCRT task, participants play a computerbased competitive reaction-time task against a bogus opponent, where the winner of each trial is given opportunity to deliver a noise blast/electric shock to the loser (see Giancola & Parrott, 2008). Following Lawrence and Hutchinson (2013, 2014), we (i) use participants’ intensity of noise blasts delivered to the bogus opponent as a measure of direct aggression, (ii) provide a non-aggressive response option, and (iii) vary the levels of provocation from the opponent (i.e., noise blast intensity delivered to participant by the bogus opponent) such that participants are exposed to no provocation, low/moderate provocation, and high provocation trials. We expect that the main effect of depletion on aggression, and the moderation of mindfulness on the depletion-aggression link would be strongest under conditions of a moderate degree of provocation (i.e., low provocation trial in the current TCRT task). The logic behind this prediction follows the logic of person x situation models. In the Traits as Situational Sensitivities model (TASS: Marshall & Brown, 2006), the person-based predictors of aggression are proposed to function in a threshold-like manner to situation-based predictors of aggression. For example, a person who is high in trait aggression
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
127
would require less provocation to activate aggressive behaviour compared to a person who is low in trait aggression. In the same way, less situational strength would be needed to evoke aggression in those who are already depleted than in those who are not depleted. Past studies have shown that the effect of depletion on aggression is less clear in the absence of provocation (DeWall et al., 2007; Finkel et al., 2009) and that interventions aimed at reducing aggression in the TCRT task tend to lose their effectiveness as the levels of provocation increase (Lawrence & Hutchinson, 2013, 2014). Therefore, compared to non-depleted participants, depleted participants should be more responsive to mindfulness induction following low/moderate provocation, than following no or high provocation. We test the following primary hypotheses: 1. Compared to non ego-depleted participants, ego-depleted participants will deliver higher levels of blast intensity to the opponent, particularly on low/moderate provocation trials. No strong prediction is made for the effect following no provocation or high provocation trials. 2. Mindfulness induction will moderate the effect of ego-depletion on aggression. Specifically, compared to depleted participants without mindfulness induction, depleted participants who receive mindfulness induction will deliver lower levels of blast intensity to the opponent particularly on low/moderate provocation trials. No strong prediction is made for the effect following no provocation or high provocation trials. If self-control ability in the domain of aggression is related to performance in other self-control tasks, physical stamina in a handgrip task should similarly be susceptible to any effect of depletion and the possible counteracting effects of mindfulness induction. The secondary hypotheses predict that: 1. Compared to non ego-depleted participants, ego-depleted participants will be less able to maintain their baseline handgrip performance both a) before and b) after the TCRT task. 2. Mindfulness induction will moderate the effect of ego-depletion on handgrip performance. Specifically, compared to depleted participants without mindfulness induction, depleted participants who receive mindfulness induction will be more able to maintain their baseline handgrip performance both a) before and b) after the TCRT task. We make no prediction with regards to the main effect of mindfulness induction on aggression or on physical stamina, since many ego-depletion studies have shown that it is difficult for brief manipulations to improve self-control performance beyond baseline levels (cf. Friese et al., 2012). The efficacy of the mindfulness induction will be measured with the Toronto Mindfulness Scale (TMS: Lau et al., 2006), which is designed to be administered after meditation practices. To examine any effect of negative affect on aggression (Anderson & Bushman, 2002), participants’ mood will be assessed with the Positive Affect, Negative Affect Schedule (PANAS: Watson, Clark, & Tellegen, 1988). 2. Materials and methods 2.1. Participants and design We used a 2 (ego-depletion task vs. no-depletion) 2 (mindfulness induction vs. no-mindfulness) between-subjects design, crossed with participants’ sex (Fig. 1). Equal numbers of participants were allocated to each condition using a random order generator. Only native British participants were invited to control for any potential cultural impact on aggression (Bergeron & Schneider, 2005). The final sample consisted of 110 subjects (58 females), with ages ranging from 18 to 31 (M = 19.52, SD = 2.03). Of these, 107 subjects reported never having encountered mindfulness meditation, 2 subjects practised mindfulness once a year, 1 subject on monthly basis, and none practised mindfulness on weekly or daily basis. Given the small number of subjects who had experiences with mindfulness practices, no further comparisons were carried out based on prior mindfulness experience. 2.2. Procedure Participants were invited via posters/leaflets on campus and departmental mailing lists of a large UK university to participate in a study investigating the effects of personality and experimental treatments on the way people perform in a competitive reaction-time task. It was specified that participants would be included only if British English was their first language. They were given a £5 inconvenience allowance and a chance to win £50 for the fastest reaction-time across the entire study. Alternatively, course credit points were given for students signing up via a research scheme. Participants came into the lab individually. After giving consent, they were asked to provide demographic information and furnish a baseline measure of handgrip performance. Then they were left alone to watch a 6-min attention control videotape, and were asked to open the cubicle door once the videotape ended. The experimenter then returned to the cubicle and asked participants to complete two questions assessing the effectiveness of the depletion procedure, three dummy questions about the interviewee on the videotape, and the first PANAS.
128
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Fig. 1. Flow of participants. Note: M = Males; F = Females.
Half of participants in the depletion group and half of those in the no-depletion condition were then given a mindfulness induction task, whereas the rest of participants were given neutral education-themed information. The experimenter left the room again while participants listened to and practised one of the appropriate audio guided instructions for 15 min. When participants opened the cubicle door, the experimenter entered the cubicle and asked them to complete the TMS and the second measure of handgrip performance. Next, participants were told that the experimenter needed to prepare their TCRT task opponent. After leaving the room for a while (ostensibly to prepare the opponent), the experimenter returned and gave participants instruction to begin the TCRT task. Participants were then left alone to perform this task. When participants opened the cubicle door to indicate the end of the TCRT task, the experimenter asked them to complete the second PANAS and perform the final measure of handgrip task. Finally participants were probed for suspicions, debriefed, and given help and support information. 2.3. Materials and apparatus 2.3.1. Self-report measures Participants’ mood was measured using the Positive Affect, Negative Affect Schedule (PANAS: Watson et al., 1988), which comprises 10 items assessing current positive affect (e.g., ‘‘interested’’) and 10 items measuring current negative affect (e.g., ‘‘irritable’’), on a 5-point scale (1 = very slightly or not at all and 5 extremely). Since negative affect due to unpleasant experiences often triggers aggression (Anderson & Bushman, 2002), the mood measure was employed after the depletion task and then again after the TCRT task. In the current sample, both mood subscales after the depletion task showed good internal consistency (a = .87 and .72 for PA and NA, respectively) but were not significantly correlated (r = .14, p = .15, 95% CI = .02 to .35). Both mood subscales also showed good internal consistency (a = .91 and .77 for PA and NA, respectively) and were positively correlated (r = .63, p < .001, 95% CI = .36 to .81) when given after the TCRT task. The efficacy of the mindfulness induction was assessed with Toronto Mindfulness Scale (TMS: Lau et al., 2006), which measures the action of experiencing the present with (i) curiosity (6 items, e.g., ‘‘I was curious about my reactions to things’’) and (ii) decentering (7 items, e.g., ‘‘I was aware of my thoughts and feelings without overidentifying with them’’), on a 5point Likert-scale (0 = not at all and 4 = very much). In the current sample, both mindfulness subscales showed good internal consistency (a = .84 and .71 for curiosity and decentering, respectively) and were positively correlated (r = .58, p < .001, 95% CI = .43 to .71). The efficacy of the depletion manipulation was assessed with two following questions (i) ‘‘How difficult was it for you to follow the instructions during the video clip’’, and (ii) ‘‘How much did you have to control your attention to follow the instructions during the video clip’’, on a 7 point Likert-scale (1 = not at all and 7 = very much).
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
129
2.3.2. Ego-depletion task Depletion was induced using the dual-task self-control procedure, in which participants in the depletion group were asked to perform two unrelated tasks requiring self-control, whereas participants in the no-depletion condition were also asked to perform two unrelated tasks, but only the second task required self-control (Baumeister et al., 2007). For the first self-control task, we used an attention control task (DeWall et al., 2007; Finkel et al., 2009). All participants were asked to watch a 6-min videotape without audio of a woman being interviewed by an off-camera interviewer, and were told that they would later be making judgments of the interviewee. During the video clip, a series of common one-syllable words (e.g., ‘‘glue’’) appeared at the bottom of the screen for 10 s each. Participants in the depletion condition were instructed ‘‘not to read or look at any words that may appear on the screen’’, and to redirect their gaze to the interviewee’s face immediately if they caught themselves looking at the words. Participants in the no-depletion condition were not given any such instructions and were not told in advance that there would be words at the bottom of the screen. For the second self-control task, we used performance in a handgrip task (see Section 2.3.5). 2.3.3. Mindfulness induction Participants in the mindfulness condition were guided with a 15-min audio instruction for ‘‘Mindfulness of body and breath’’ (Williams & Penman, 2011) to direct their attention towards witnessing the full sensations of breathing without the intention of altering these experiences, and to notice in an accepting manner when their minds wander and gently return their focus to their breathing. This task has been used in Mindfulness-Based Cognitive Therapy (MBCT: Oxford Mindfulness Centre, 2002) and in past laboratory research (Kramer, Weger, & Sharma, 2013). Mindfulness with regards to the body and breath is used because the behaviour can be observed for a fairly extended period by participants without creating too much stress for novice practitioners (Sujiva, 2000). For a total 15 min, participants in the control condition listened to two neutral educational excerpts and composed all the possible words that came to mind using 100 tile of letters from a scrabble set. A similar procedure has been used elsewhere (Erisman & Roemer, 2010; Kramer et al., 2013) as a control condition for mindfulness induction. Although this type of task still requires participants to concentrate, it is unlikely to induce additional depleting effects because it does not require effortful suppression of an impulse (Baumeister et al., 2007) and is neither high in difficulty nor complexity (Hagger et al., 2010). 2.3.4. Adapted Taylor Competitive Reaction Time (TCRT) task The adapted TCRT task (Lawrence & Hutchinson, 2013; Lawrence & Hutchinson, 2014) was presented to participants on computer using E-Prime software version 2.0. A similar paradigm has been used in studies measuring the effect of depletion (DeWall et al., 2007; Finkel et al., 2009) and mindfulness induction (Heppner et al., 2008) on aggression. Participants were informed that they would be playing a computer-based competition task (i.e., hitting the spacebar on the computer keyboard as soon as a white circle stimulus appears on the screen) against an opponent located in a nearby room. In reality there was no opponent and the outcomes of the task was pre-programmed. Each time participants ‘won’ a trial, they could select a level of noise blast to be delivered to the opponent, ranging from 0 (no blast, non-aggressive option) through to 8 (maximum blast). When they ‘lost’ a trial, they received a noise blast through surround earphones for a fixed duration of 500 ms, at a pre-programmed level ostensibly chosen by the opponent (up to a maximum of 90 db). The TCRT task comprised a total of 81 trials. Participants always won the first trial to assess the baseline blast intensity that they delivered to the opponent prior to any provocation from the opponent (no provocation trial). On subsequent losing trials, they received blasts in two blocks of (i) low/moderate intensity (40 trials, participants lost 50% of time at random and received blast levels 1–4: low/moderate provocation) and (ii) higher intensity (40 trials, participants lost 50% of time at random and received blast levels 5–8: high provocation). The intensity of blasts delivered to the opponent under each block of trials (no provocation, mean low/moderate provocation, and mean high provocation) was used as a measure of participants’ aggression. 2.3.5. Handgrip task A manual handgrip (consisting of two handles and a connecting metal spring) was used as the second self-control measure. This measure compromises resisting discomfort in the forearm muscles and the resulting urge to quit (Baumeister et al., 2007). A small wad of paper was inserted between the two handles when participants gripped them together. Participants were required to continuously squeeze the handgrip using the dominant hand until the wad of paper fell out. Participants performed this task at three time points: at the beginning of the study (baseline handgrip performance), prior to the TCRT task (as a measure of performance in the second self-control task), and after the TCRT task. Handgrip performance was calculated as the duration time(s) elapsing before the wad of paper fell. Crucially, participants were not informed about the number of handgrip task they would be required to perform, since the expectation of further self-control acts may increase the depletion effects (see Hagger et al., 2010).
130
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
3. Results 3.1. Manipulation checks Unless specified, all analyses were performed using 5,000 bootstrap resamples (N = 110) with 95% bias-corrected confidence intervals. A bootstrap sampling distribution is estimated with replacement from the original sample; therefore, it reduces the impact of anomalies and helps ensure the stability of the models (Hayes, 2012a). The depletion manipulation was partly effective. Participants in the depletion condition (n = 55) reported having controlled their attention to a greater extent than no-depletion participants (n = 55; t(108) = 2.26, p = .03, 95% CI = 1.31 to .09; M depletion = 4.91, SD = 1.78 vs. M no-depletion = 4.20, SD = 1.50), but did not rate the task as more difficult than those in the no-depletion condition (t(108) = .42, p = .68, 95% CI = .52 to .84; M depletion = 3.31, SD = 1.95 vs. M no-depletion = 3.45, SD = 1.69). Participants receiving mindfulness induction (n = 55) reported higher decentering than participants not receiving mindfulness induction (n = 55; t(105.77) = 2.65, p = .01, 95% CI = .58 to .09; M mindfulness = 3.03, SD = .60 vs. M no-mindfulness = 2.70, SD = .70), but did not show higher levels of curiosity (t(117) = 1.15, p = .25, 95% CI = .50 to .13; M mindfulness = 3.50, SD = .81 vs. M no-mindfulness = 3.31, SD = .88). Participants in the depletion condition reported no differences than non-depleted participants in negative affect (t(108) = .57, p = .57, 95% CI = .16 to .09; M depletion = 1.26, SD = .38 vs. M no-depletion = 1.23, SD = .24), but scored higher in positive affect (t(108) = 2.24, p = .03, 95% CI = .55 to .03; M depletion = 2.66, SD = .68 vs. M no-depletion = 2.37, SD = .70). This positive affect, however, was not related to aggression under any of the three provocation levels (rs < .10, ps > .29) or to handgrip performance (r = .07 and .08 at pre-TCRT task and post-TCRT task, all controlling for baseline handgrip duration time, ps > .42). No differences were reported between the groups in positive (ps > .23) or negative affect (ps > .40) when the mood scale was given again after the TCRT task. Thus, mood fluctuations were not related to the experimental outcomes. 3.2. Association between aggression and handgrip performance Table 1 depicts correlations between the dependent measures, with handgrip performance reported controlling for baseline handgrip duration time (M baseline = 51.65, SD = 39.63). Participants’ blast intensities to the opponent were positively related to each other under all provocation levels (ps < .001). Participants’ measures of handgrip performance were also positively related to each other (ps < .001). Handgrip performance pre-TCRT task was associated, negatively, with blast intensity under no provocation trial. This indicates that those who were more able to maintain their baseline handgrip performance were also less aggressive prior to the provocation procedure. The correlations between blast intensity and handgrip performance did not occur after the TCRT task (ps > .33). 3.3. Moderation of mindfulness induction on the link between depletion and aggression As typically found in TCRT studies varying levels of provocation from the opponent (see Lawrence & Hutchinson, 2013; Lawrence & Hutchinson, 2014), participants delivered more intense blasts following high provocation compared to low provocations (p < .001), and following low provocation compared to no provocation (p < .001; see Table 1 for mean differences in blast intensity). The moderating effect of mindfulness induction on the link between depletion and aggression was tested at each provocation level using a bootstrap method (Hayes, 2012b). Theoretically, X (depletion) is depicted to exert an influence on Y (aggression), and this effect is proposed to be influenced by M (mindfulness induction). These effects are estimated mathematically in the form of a linear equation: Y = i + (c1 + c3M)X + c2M + ey. The (c1 + c3M) function represents the conditional Table 1 Correlations between dependent measures. Measures
Blast intensity under no provocation (1) Blast intensity under low/moderate provocation (2) Blast intensity under high provocation (3) Handgrip performance pre-TCRT task (4) Handgrip performance post-TCRT task (5) M (SD) absolute scores
Bias-corrected bootstrap 95% CI 1
2
3
4
5
[1.00, 1.00] [0.65, 0.83]*** [0.40, 0.66]*** [ 0.41, 0.04]** [ 0.27, 0.08] 2.61 (2.50)
[1.00, 1.00] [0.77, 0.91]*** [ 0.37, 0.04] [ 0.31, 0.15] 3.39 (2.00)
[1.00, 1.00] [ 0.35, 0.09] [ 0.32, 0.13] 4.14 (2.17)
[1.00, 1.00] [0.26, 0.73]*** 42.38 (30.89)
[1.00, 1.00] 43.18 (31.05)
Note: Handgrip performance reported controlling for baseline handgrip duration time(s). Higher positive scores indicate greater aggression and better physical (handgrip) stamina. ** p < .01. *** p < .001.
131
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
effect of X on Y, in which c1 estimates the effect of X on Y when M = 0, and c3 estimates how much the effect of X on Y changes as M changes by one unit. The main focus in a moderation model is whether c3, the interaction coefficient between X and M, is statistically different from zero (for further details, consult Hayes, 2012a). Findings shown in Table 2 support Hypothesis 1, in that participants’ aggression was affected by a main effect of depletion. Specifically, compared to non-depleted participants, depleted participants delivered higher levels of blast intensity as compared to non-depleted participants under no provocation and low/moderate provocation trials (p < .01). The main effect of mindfulness was non-significant across all three levels of provocation (ps > .78). Importantly, Hypothesis 2 was supported, such that the link between depletion and blast intensity was moderated by mindfulness induction under no provocation and low/moderate provocation trials. Under no provocation and low/moderate provocation, depleted participants who received mindfulness induction delivered lower levels of blast intensity than depleted participant without mindfulness induction (all ps = .04), whereas non-depleted participants with mindfulness induction delivered comparable blast intensities to non-depleted participants without mindfulness induction (see Fig. 2). The moderating effect of mindfulness diminished under conditions of high provocation (p = .13). 3.4. Moderation of mindfulness induction on the link between depletion and handgrip performance In the ego-depletion literature, handgrip performance is commonly calculated by subtracting each participant’s baseline duration time from their subsequent duration time(s) (see Hagger et al., 2010). A negative change score indicates that the participant is less able to maintain baseline handgrip duration, as would be expected if the person is under depleted condition. In this current study, residualising handgrip scores pre-TCRT task and post-TCRT task by baseline duration time resulted in significant correlations with the baseline score (ps < .001). When such regression towards the mean influences the measurement then covariance analysis is more recommended than analysis of change scores (Bonate, 2000). Thus, to analyse the effect of mindfulness on post-depletion handgrip performance, we used the same bootstrap procedure while controlling for baseline handgrip duration time as a covariate. The findings shown in Table 2 reveal that the standard main effect of depletion on handgrip performance was found prior to the TCRT task (Hypothesis 3a; p = .01), but not after the TCRT task (Hypothesis 3b; p > .75). No support was found for the moderation of mindfulness on the depletion and handgrip performance link at both time points (Hypotheses 4a and 4b; ps > .38). Unexpectedly, after the TCRT task, mindfulness induced participants became more able to maintain their baseline handgrip performance independent of depletion condition (p = .03). 4. Discussions This study is the first to examine the effect of a brief laboratory mindfulness induction on aggressive behaviour using a self-control resource depletion model. It extends past research on the role of mindfulness induction on provoked aggression (Heppner et al., 2008) by showing that mindfulness could be useful also to reduce post-depletion aggression. This finding is important as self-control failure has been acknowledged to increase the association between aggressive triggers and various types of aggressive behaviour (DeWall et al., 2007; Finkel et al., 2009; Stucke & Baumeister, 2006). In this study, we specifically predicted that the main effect of depletion on aggression, and the moderation of mindfulness on the depletion-aggression link would vary as a function of the level of provocation. In line with the TAAS model (Marshall & Brown, 2006), we found that in response to low/moderate provocation, the depletion effect was moderated by mindfulness induction. The descriptive pattern was similar but less strong following high provocation trials, as in previous research using the TCRT task (e.g., Lawrence & Hutchinson, 2014). Table 2 Moderation with bootstrap method, using depletion as predictor, mindfulness as moderator, blast intensity and handgrip performance as outcomes. Model tested
Blast intensity No provocation
Depletion (c1) Mindfulness induction (c2) Depletion x mindfulness (c3) R2 on interaction (%) Conditional effect of depletion without mindfulness Conditional effect of depletion with mindfulness
Handgrip performance
B (SE)
Low/moderate provocation B (SE)
High provocation B (SE)
Pre-TCRT task B (SE)
Post-TCRT task B (SE)
2.19*** (0.64) 0.00 (0.64) 1.90* (0.90) 3.63%* 2.19*** (0.64) 0.29 (0.64)
1.40 (0.52)** 0.15 (0.52) 1.55* (0.74) 3.79%* 1.40** (0.52) 0.15 (0.52)
1.04 (0.58) 0.01 (0.58) 1.25 (0.82) 2.07% 1.04 (0.58) 0.20 (0.58)
9.49 (3.75)** 5.00 (3.75) 4.70 (5.30) 0.002% 9.49 (3.75) 4.79 (3.74)
1.12 (3.58) 7.70 (3.58)* 2.65 (5.06) 0.001% 1.12 (3.58) 3.77 (3.57)
Note: Handgrip performance reported controlling for baseline handgrip duration time(s). Higher positive scores indicate greater aggression and better physical (handgrip) stamina. Coding: 1 = depletion or mindfulness induction condition; 0 = non-depletion or no mindfulness induction condition. B = unstandardised regression coefficient; SE = standard error; R2 = variance increase * p < .05. ** p < .01. *** p < .001.
132
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Fig. 2. Mindfulness induction depletion effect on blast intensity across provocation levels. Error bars indicate standard error of mean, asterisk indicates significant (p < .05) decrease in blast intensity. Scale goes up to 8 as a maximum.
To scrutinise the role of mindfulness induction across different tasks of self-control, this study incorporated a measure of physical stamina in a handgrip task. The initial analysis indicated that participants who performed better on the handgrip task also tend to be less aggressive. However, the role that mindfulness had on these two self-control tasks (aggression and handgrip) appeared to differ. While the effect of mindfulness on aggression was more observable amongst those egodepleted, its benefit on handgrip performance was not dependent upon individuals having been subjected to the depletion. The main effect of mindfulness induction occurred on handgrip performance in particular when people have experienced provocation (after the TCRT task) instead. As suggested by Friese et al. (2012), brief experimental manipulations will be more effective in increasing self-control performance in individuals who are under the influence of various factors that trigger impulsive responding. Thus in this study, mindfulness induction may have improved handgrip performance after the TCRT task because the provocation experienced during this task acted as one such trigger of impulsive responding. Based on the resource depletion model, one would hypothesize that the depletion effect should become increasingly large over the course of multiple self-control tasks. In contrast, at the end of this study the negative effect of depletion in relation to handgrip performance appeared to diminish, resulting in a comparable level of final performance between depletion/control groups. One possible explanation is that instead of the depletion effect, an adaptation effect might have occurred with repeated self-control acts. Indeed, past study has shown that for participants who were required to perform an additional third self-control task after the second self-control task, their performance might remain stable or even improve (Converse & DeShon, 2009). Another possibility is that the depletion effect could have been attenuated by participants’ motivational factors. Because all participants were told that the final measure of handgrip was their last task to be accomplished, they became sufficiently motivated to accomplish this task and performed equally effective regardless of their depletion condition. Relatedly, this could be due to conceptual differences between the two self-control tasks used in this study. The handgrip task explicitly required physical muscle control as well as motivational control (they were explicitly told to keep the paper wad from falling for as long as possible), whereas the TCRT task required no such explicit instruction to inhibit aggressive behaviour. From a recent motivational viewpoint of ego-depletion, self-control failure is seen as a motivational deficit to perform a demanding task rather than a temporary loss of some limited resources or energy (Inzlicht, Schmeichel, & Macrae, 2014). More research is needed to explore the dynamics of self-control over time as well as non resource-based accounts of self-control. Our findings are important also as it could have been predicted that a mindfulness meditation task such as the one used in this study could be expected to add to the existing self-control demands because the manipulation is, essentially, an attention control task (e.g., Masicampo & Baumeister, 2007). It has been speculated that to be able to sustain attention on the present moment, individuals naïve to mindfulness practices might need to exert cognitive control over their emotional reactions (Holzel et al., 2011). Instead, in the present study and in the work by Friese et al. (2012), such mindfulness manipulations counteract depletion. Thus, it is timely to identify the underlying emotional, cognitive, and biological processes that might come into play in the association between mindfulness, self-control, and aggression. Possible mediating mechanisms such as the mindfulness acceptance of internal experiences, nonattachment, and clarification of values (Baer, 2003; Shapiro et al., 2006) should be investigated. Importantly, previous research has shown that depleted individuals who were asked to monitor their performance (i.e., using an accurate clock) or who were already high in trait self-monitoring persisted longer on subsequent self-control task (Wan & Sternhal, 2008). Although mindfulness appears to share considerable variance with self-monitoring, the focus of the latter construct is on attention towards aspects of the self, whereas mindfulness simply
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
133
offers attention on the ‘in-the-moment’ experience (Brown et al., 2007a; Brown et al., 2007b) which can extend beyond the self. Teasing the monitoring function of mindfulness from a comparative self-monitoring against some desirable standards is important for future studies. It should be emphasised, however, that for an exercise to count as a true mindfulness meditation, some personal experience with mindfulness exercises may be necessary. Although most participants seemed to be serious in their attempts to engage with the task, and the manipulation check did reveal that participants in the mindfulness condition reported higher levels of decentering, the brief exercise might have not been sufficiently strong to evoke the whole range of state mindfulness (see also Erisman & Roemer, 2010). Decentering, or observing one’s experience with some distance rather than being carried away by thoughts and feelings, was previously shown to predict lower perceived stress and distress (Lau et al., 2006). So it could be that the cultivation of decentering might have enabled individuals to become more tolerant to unpleasant experiences related to depletion and provocation. The current study used neutral educational information as control condition for the mindfulness manipulation. Although a similar control has been employed in past research on mindfulness (Erisman & Roemer, 2010; Kramer et al., 2013) and egodepletion (Friese et al. 2012), ideally a control group that has the same expectation of improvement as the experimental group would be more effective as an active control (Boot, Simons, Stothart, & Stutts, 2013). As a result, we cannot thoroughly rule out the possibility of placebo or demand effects contributing to the results in this study. Future research could employ other types of meditation or use alternative emotion regulation strategies (e.g., cognitive reappraisal, distraction, unfocused attention). These latter conditions were not employed in this study because regulating emotions risked depleting participants’ self-control ability (Baumeister et al., 2007). Although we included a manipulation check to assess the efficacy of the mindfulness induction, no measure of expectation effects of the mindfulness induction or control condition was included. This was in order to avoid sensitising participants into linking the mindfulness procedure to the aggression and handgrip tasks and leading to possible demand characteristics. Nonetheless, it is possible that participants may have implicitly expected a ‘self-improvement’ effect of the mindfulness induction that could explain the results. Finally, this study consisted of entirely British university students naïve to mindfulness techniques. Replication with different populations is necessary to examine whether the differential association between mindfulness, self-control, and aggression is robust. Despite the limitations, findings from the present study contribute to the literature by showing that the benefits of mindfulness on post-depletion aggression can be demonstrated outside extensive prior training in mindfulness. Additionally, the benefit of short mindfulness induction on a behavioural self-control task appears to hold up after people have also experienced provocation from others. The inclusion of mindfulness interventions for those who have a particular difficulty in effortful control of aggressive impulses should now be examined further. Conflicts of interest The authors have no conflicts of interest to declare. Acknowledgment The authors would like to thank the anonymous reviewers for their thorough feedbacks. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.concog.2014.12.008. References Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology, 53(1), 27–51. http://dx.doi.org/10.1146/ annurev.psych.53.100901.135231. Anonymous (2012). Udana: Exclamations (Thanissaro Bhikkhu, Trans.). Valley Center, CA: Metta Forest Monastery. Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10(2), 125–143. http://dx.doi.org/10.1093/clipsy.bpg015. Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control. Current Directions in Psychological Science, 16(6), 351–355. http:// dx.doi.org/10.1111/j.1467-8721.2007.00534.x. Bergeron, N., & Schneider, B. H. (2005). Explaining cross-national differences in peer-directed aggression: A quantitative synthesis. Aggressive Behavior, 31(2), 116–137. http://dx.doi.org/10.1002/ab.20049. Bodhi, B. (2000). A comprehensive manual of Adhidhamma. Seattle: BPS Pariyatti. Bonate, P. L. (2000). Analysis of pretest-posttest designs. Boca Raton, Florida: Chapman & Hall/CRC. Boot, W. R., Simons, D. J., Stothart, C., & Stutts, C. (2013). The pervasive problem with placebos in psychology: Why active control groups are not sufficient to rule out placebo effects. Perspectives on Psychological Science, 8(4), 445–454. http://dx.doi.org/10.1177/1745691613491271. Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007a). Addressing fundamental questions about mindfulness. Psychological Inquiry, 18(4), 272–281. http:// dx.doi.org/10.1080/10478400701703344. Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007b). Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry, 18(4), 211–237. http://dx.doi.org/10.1080/10478400701598298.
134
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality-social, clinical and health psychology. Psychological Bulletin, 92(1), 111–135. http://dx.doi.org/10.1037/0033-2909.92.1.111. Converse, P. D., & DeShon, R. P. (2009). A tale of two tasks: Reversing the self-regulatory resource depletion effect. Journal of Applied Psychology, 94(5), 1318–1324. http://dx.doi.org/10.1037/a0014604. DeWall, C. N., Baumeister, R. F., Stillman, T., & Gailliot, M. T. (2007). Violence restrained: Effects of self-regulation and its depletion on aggression. Journal of Experimental Social Psychology, 43(1), 62–76. http://dx.doi.org/10.1016/j.jesp.2005.12.005. Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the convergent validity of self-control measures. Journal of Research in Personality, 45(3), 259–268. http://dx.doi.org/10.1016/j.jrp.2011.02.004. Erisman, S. M., & Roemer, L. (2010). A preliminary investigation of the effects of experimentally induced mindfulness on emotional responding to film clips. Emotion, 10(1), 72–82. http://dx.doi.org/10.1037/a0017162. Finkel, E. J., DeWall, C. N., Slotter, E. B., Oaten, M., & Foshee, V. A. (2009). Self-regulatory failure and intimate partner violence perpetration. Journal of Personality and Social Psychology, 97(3), 483–499. http://dx.doi.org/10.1037/a0015433. Fix, R. L., & Fix, S. T. (2013). The effects of mindfulness-based treatments for aggression: A critical review. Aggression and Violent Behavior, 18(2), 219–227. http://dx.doi.org/10.1016/j.avb.2012.11.009. Friese, M., Messner, C., & Schaffner, Y. (2012). Mindfulness meditation counteracts self-control depletion. Consciousness and Cognition, 21(2), 1016–1022. http://dx.doi.org/10.1016/j.concog.2012.01.00. Giancola, P. R., & Parrott, D. J. (2008). Further evidence for the validity of the Taylor aggression paradigm. Aggressive Behavior, 34(2), 214–229. http:// dx.doi.org/10.1002/ab.20235. Hagger, M. S., Wood, C., Chris, S., & Chatzisarantis, N. L. D. (2010). Ego-depletion and the strength model of self-control: A meta-analysis. Psychological Bulletin, 136(4), 495–525. http://dx.doi.org/10.1037/a0019486. Hayes, A. F. (2012a). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from. Hayes, A. F. (2012b). PROCESS for SPSS and SAS [Macro]. Retrieved from . Heppner, W. L., Kernis, M. H., Lakey, C. E., Campbell, W. K., Goldman, B. M., Davis, P. J., et al (2008). Mindfulness as a means of reducing aggressive behavior: Dispositional and situational evidence. Aggressive behavior, 34(5), 486–496. http://dx.doi.org/10.1002/ab.20258. Holzel, B. K., Lazar, S. W., Gard, T., Schuman-Olivier, Z., Vago, D. R., & Ott, U. (2011). How does mindfulness meditation work? Proposing mechanisms of action from a conceptual and neural perspective. Perspectives on Psychological Science, 6(6), 537–559. http://dx.doi.org/10.1177/1745691611419671. Inzlicht, M., & Gutsell, J. (2007). Running on empty: Neural signals for self-control failure. Psychological Science, 18(11), 933–937. http://dx.doi.org/10.1111/ j.1467-9280.2007.02004.x. Inzlicht, M., Schmeichel, B. J., & Macrae, C. N. (2014). Why self-control seems (but may not be) limited. Trends in Cognitive Sciences, 18(3), 127–133. http:// dx.doi.org/10.1016/j.tics.2013.12.009. Kramer, R. S. S., Weger, U. W., & Sharma, D. (2013). The effect of mindfulness meditation on time perception. Consciousness and Cognition, 22(3), 846–852. http://dx.doi.org/10.1016/j.concog.2013.05.008. Lau, M. A., Bishop, S. R., Segal, Z. V., Buis, T., Anderson, N. D., Carlson, L., et al (2006). The Toronto mindfulness scale: development and validation. Journal of Clinical Psychology, 62(12), 1445–1467. http://dx.doi.org/10.1002/jclp.20326. Lawrence, C., & Hutchinson, L. (2013). The influence of individual differences in sensitivity to provocations on provoked aggression. Aggressive Behavior, 39(3), 212–221. http://dx.doi.org/10.1002/ab.21473. Lawrence, C., & Hutchinson, L. (2014). The impact of non-aggressive behaviour early in aggressive interactions: Sex differences in direct and indirect aggression in response to provocation. British Journal of Psychology, 105(1), 127–144. http://dx.doi.org/10.1111/bjop.12020. Marshall, M. A., & Brown, J. D. (2006). Trait aggressiveness and situational provocation: A test of the traits as situational sensitivities (TASS) model. Personality and Social Psychology Bulletin, 32(8), 1100–1113. http://dx.doi.org/10.1177/0146167206288488. Masicampo, E. J., & Baumeister, R. F. (2007). Relating mindfulness and self-regulatory processes. Psychological Inquiry, 18(4), 255–258. http://dx.doi.org/ 10.1080/10478400701598363. Oxford Mindfulness Centre (University of Oxford Department of Psychiatry) (2002). Mindfulness-based Cognitive Therapy. Retrieved from
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Brief mindfulness induction could reduce aggression after depletion Cleoputri Yusainy a,⇑, Claire Lawrence b a b
Psychology Programme, Faculty of Social and Political Sciences, Brawijaya University, Malang, Indonesia Personality, Social Psychology, and Health (PSPH) Group, School of Psychology, University of Nottingham, Nottingham NG7 2RD, UK
a r t i c l e
i n f o
Article history: Received 12 April 2014
Keywords: Mindfulness induction Self-control Ego-depletion Aggressive behaviour
a b s t r a c t Many experiments have shown that one’s ability to refrain from acting on aggressive impulses is likely to decrease following a prior act of self-control. This temporary state of self-control failure is known as ego-depletion. Although mindfulness is increasingly used to treat and manage aggressive behaviour, the extent to which mindfulness may counteract the depletion effect on aggression is yet to be determined. This study (N = 110) investigated the effect of a laboratory induced one-time mindfulness meditation session on aggression following depletion. Aggression was assessed by the intensity of aversive noise blast participants delivered to an opponent on a computerised task. Depleted participants who received mindfulness induction behaved less aggressively than depleted participants with no mindfulness induction. Mindfulness also improved performance on a second measure of self-control (i.e., handgrip perseverance); however, this effect was independent of depletion condition. Motivational factors may help explain the dynamics of mindfulness, self-control, and aggression. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction 1.1. Mindfulness, self-control ego-depletion, and aggression ‘‘Whose mind stands like a mountain. Steady, it is not perturbed. Unattached to things that arouse attachment. Unangered by things that provoke anger. When his mind is cultivated thus. How can suffering come to him?’’ (Udana 4.4, in Anonymous, 2012). At any given moment, the contents of one’s thoughts and emotions may be painful, pleasant, or neutral. Mindfulness — translated from the Pali (the language in which early Buddhist teaching was recorded) word, sati (‘memory’) — is a practice that acknowledges the contents of consciousness as states that arise and pass, but do not necessarily need action (Bodhi, 2000; Sujiva, 2000). The result of this practice is a mental gap between consciousness and its contents, or between attention and its objects (Baer, 2003; Brown, Ryan, & Creswell, 2007a,b; Shapiro, Carlson, Astin, & Freedman, 2006). Central for the present study is the idea that mindfulness may reduce an individual’s reactivity that is typically associated with aggression-triggering events. The current literature has documented the application of mindfulness-based interventions for treatment of aggression in mental health and forensic settings (see Fix & Fix, 2013; Shonin, Gordon, Slade, & Griffiths, 2013). However, much less aggression research has been conducted on the effect of brief mindfulness induction as opposed to the context of extensive ⇑ Corresponding author. E-mail address: [email protected] (C. Yusainy). http://dx.doi.org/10.1016/j.concog.2014.12.008 1053-8100/Ó 2014 Elsevier Inc. All rights reserved.
126
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
training in mindfulness. Preliminary support for the benefit of laboratory inductions of mindfulness on provoked aggression has been provided by Heppner et al. (2008). In their study, participants were randomly assigned to one of three conditions in a social rejection task: acceptance from partner, rejection by partner, or rejection plus mindfulness (i.e., a 5-min task of eating raisins while being focused in a mindful way on the experience of eating the raisin). In a subsequent aggression task using the administration of aversive noise blasts as the aggressive behaviour, those in the rejection plus mindfulness condition behaved less aggressively than those in the rejection only condition, and their behaviour did not differ from those in the non-rejection condition. We expand on this method by looking at prior self-control behaviour rather than social rejection as a possible precursor to aggression. The importance of trait self-control on the link between mindfulness and aggression has been reported in a recent correlational study (Yusainy & Lawrence, 2014). Even though trait self-control appears to be a highly desirable quality, a momentary act of self-control may be costly. As suggested in ‘‘test-operate-test-exit’’ theory (TOTE, see Carver & Scheier, 1982), effective self-control requires a continuous monitoring of the self’s current state against some desirable goals or standards. Once a person has detected a discrepancy between current and desired states, some operations to change the self can be initiated. Much like a muscle, however, these operations are proposed to rely on the availability of a common and limited pool of ‘resources’, or self-control strength (Baumeister, Vohs, & Tice, 2007). After performing initial acts requiring selfcontrol, the capacity for further control, even on unrelated self-control tasks, is temporarily decreased (i.e., ‘ego-depletion’: a meta-analysis of this model is provided in Hagger, Wood, Chris, & Chatzisarantis, 2010). Experimental research has shown that compared to non ego-depleted participants, depleted participants (who had previously resisted eating tempting food or stifled physical and emotional reactions) delivered more negative ratings to an experimenter that had previously insulted them (Stucke & Baumeister, 2006; see also DeWall, Baumeister, Stillman, & Gailliot, 2007). Participants also behaved more aggressively toward intimate partners had who ostensibly provoked them with negative feedback when they were depleted than when they were not (Finkel, DeWall, Slotter, Oaten, & Foshee, 2009). These studies conclude that under depletion conditions, subsequent attempts at controlling aggressive impulses become more prone to failure. Aggression is, arguably, a prime example of self-control failure. Unlike self-controlled attention, which encompasses both monitoring and controlling the self in order to achieve some future goals, mindfulness serves solely to monitor the contents of consciousness (Brown et al., 2007a,b). It has been demonstrated elsewhere that the neural system that monitors the mismatches between current and desired states (ERN: errorrelated negativity) is weakened after performing an initial self-control act (Inzlicht & Gutsell, 2007). As individuals who are mindful make no direct attempts to modify their experiences associated with being depleted, they become better attuned to when self-control is at stake (Teper, Segal, & Inzlicht, 2013). Indeed, greater self-control in terms of increased ERN and heightened emotional acceptance has been reported in experienced mindfulness meditators (Teper & Inzlicht, 2013). Of particular relevance is Friese, Messner, and Schaffner’s (2012) experiment with participants of a 3-day introductory seminar on mindfulness meditation. They showed that participants who had engaged in a 5-min period of mindfulness meditation after an emotion suppression task performed better on a subsequent self-control task (i.e., the crossing-out-letters task to discriminate between adjacent and similarly looking, but slightly different letters) than those in the suppression-only condition, and performed equally well as those not suppressing emotion (where participants were asked simply to watch the same movie clip). In the same way, mindfulness may act as a counteracting variable for the effect of depletion on aggression. It should be noted, however, that many measures of self-control exist, and many do not correlate highly (see Duckworth & Kern, 2011). While resisting aggressive impulses requires self-control, the direct effect of depletion on aggression, and the effect of mindfulness on post-depletion aggression may not be observed in other self-control tasks. To test this possibility, a frequently used behavioural measure of self-control (i.e., handgrip perseverance: Hagger et al., 2010) is included as a second dependent variable. As a whole, therefore, the current study examines the moderating effect of mindfulness induction on the link between ego-depletion and two consecutive self-control measures: one in the domain of aggression and one in the domain of physical stamina. In this way, we will be able to investigate whether the effects of brief mindfulness induction are consistent across different tasks of self-control. 1.2. The current study In the current study, aggression is provoked using Lawrence and Hutchinson’s (2013, 2014) adapted version of the Taylor Competitive Reaction Time (TCRT: Taylor, 1967) task. In typical work using the TCRT task, participants play a computerbased competitive reaction-time task against a bogus opponent, where the winner of each trial is given opportunity to deliver a noise blast/electric shock to the loser (see Giancola & Parrott, 2008). Following Lawrence and Hutchinson (2013, 2014), we (i) use participants’ intensity of noise blasts delivered to the bogus opponent as a measure of direct aggression, (ii) provide a non-aggressive response option, and (iii) vary the levels of provocation from the opponent (i.e., noise blast intensity delivered to participant by the bogus opponent) such that participants are exposed to no provocation, low/moderate provocation, and high provocation trials. We expect that the main effect of depletion on aggression, and the moderation of mindfulness on the depletion-aggression link would be strongest under conditions of a moderate degree of provocation (i.e., low provocation trial in the current TCRT task). The logic behind this prediction follows the logic of person x situation models. In the Traits as Situational Sensitivities model (TASS: Marshall & Brown, 2006), the person-based predictors of aggression are proposed to function in a threshold-like manner to situation-based predictors of aggression. For example, a person who is high in trait aggression
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
127
would require less provocation to activate aggressive behaviour compared to a person who is low in trait aggression. In the same way, less situational strength would be needed to evoke aggression in those who are already depleted than in those who are not depleted. Past studies have shown that the effect of depletion on aggression is less clear in the absence of provocation (DeWall et al., 2007; Finkel et al., 2009) and that interventions aimed at reducing aggression in the TCRT task tend to lose their effectiveness as the levels of provocation increase (Lawrence & Hutchinson, 2013, 2014). Therefore, compared to non-depleted participants, depleted participants should be more responsive to mindfulness induction following low/moderate provocation, than following no or high provocation. We test the following primary hypotheses: 1. Compared to non ego-depleted participants, ego-depleted participants will deliver higher levels of blast intensity to the opponent, particularly on low/moderate provocation trials. No strong prediction is made for the effect following no provocation or high provocation trials. 2. Mindfulness induction will moderate the effect of ego-depletion on aggression. Specifically, compared to depleted participants without mindfulness induction, depleted participants who receive mindfulness induction will deliver lower levels of blast intensity to the opponent particularly on low/moderate provocation trials. No strong prediction is made for the effect following no provocation or high provocation trials. If self-control ability in the domain of aggression is related to performance in other self-control tasks, physical stamina in a handgrip task should similarly be susceptible to any effect of depletion and the possible counteracting effects of mindfulness induction. The secondary hypotheses predict that: 1. Compared to non ego-depleted participants, ego-depleted participants will be less able to maintain their baseline handgrip performance both a) before and b) after the TCRT task. 2. Mindfulness induction will moderate the effect of ego-depletion on handgrip performance. Specifically, compared to depleted participants without mindfulness induction, depleted participants who receive mindfulness induction will be more able to maintain their baseline handgrip performance both a) before and b) after the TCRT task. We make no prediction with regards to the main effect of mindfulness induction on aggression or on physical stamina, since many ego-depletion studies have shown that it is difficult for brief manipulations to improve self-control performance beyond baseline levels (cf. Friese et al., 2012). The efficacy of the mindfulness induction will be measured with the Toronto Mindfulness Scale (TMS: Lau et al., 2006), which is designed to be administered after meditation practices. To examine any effect of negative affect on aggression (Anderson & Bushman, 2002), participants’ mood will be assessed with the Positive Affect, Negative Affect Schedule (PANAS: Watson, Clark, & Tellegen, 1988). 2. Materials and methods 2.1. Participants and design We used a 2 (ego-depletion task vs. no-depletion) 2 (mindfulness induction vs. no-mindfulness) between-subjects design, crossed with participants’ sex (Fig. 1). Equal numbers of participants were allocated to each condition using a random order generator. Only native British participants were invited to control for any potential cultural impact on aggression (Bergeron & Schneider, 2005). The final sample consisted of 110 subjects (58 females), with ages ranging from 18 to 31 (M = 19.52, SD = 2.03). Of these, 107 subjects reported never having encountered mindfulness meditation, 2 subjects practised mindfulness once a year, 1 subject on monthly basis, and none practised mindfulness on weekly or daily basis. Given the small number of subjects who had experiences with mindfulness practices, no further comparisons were carried out based on prior mindfulness experience. 2.2. Procedure Participants were invited via posters/leaflets on campus and departmental mailing lists of a large UK university to participate in a study investigating the effects of personality and experimental treatments on the way people perform in a competitive reaction-time task. It was specified that participants would be included only if British English was their first language. They were given a £5 inconvenience allowance and a chance to win £50 for the fastest reaction-time across the entire study. Alternatively, course credit points were given for students signing up via a research scheme. Participants came into the lab individually. After giving consent, they were asked to provide demographic information and furnish a baseline measure of handgrip performance. Then they were left alone to watch a 6-min attention control videotape, and were asked to open the cubicle door once the videotape ended. The experimenter then returned to the cubicle and asked participants to complete two questions assessing the effectiveness of the depletion procedure, three dummy questions about the interviewee on the videotape, and the first PANAS.
128
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Fig. 1. Flow of participants. Note: M = Males; F = Females.
Half of participants in the depletion group and half of those in the no-depletion condition were then given a mindfulness induction task, whereas the rest of participants were given neutral education-themed information. The experimenter left the room again while participants listened to and practised one of the appropriate audio guided instructions for 15 min. When participants opened the cubicle door, the experimenter entered the cubicle and asked them to complete the TMS and the second measure of handgrip performance. Next, participants were told that the experimenter needed to prepare their TCRT task opponent. After leaving the room for a while (ostensibly to prepare the opponent), the experimenter returned and gave participants instruction to begin the TCRT task. Participants were then left alone to perform this task. When participants opened the cubicle door to indicate the end of the TCRT task, the experimenter asked them to complete the second PANAS and perform the final measure of handgrip task. Finally participants were probed for suspicions, debriefed, and given help and support information. 2.3. Materials and apparatus 2.3.1. Self-report measures Participants’ mood was measured using the Positive Affect, Negative Affect Schedule (PANAS: Watson et al., 1988), which comprises 10 items assessing current positive affect (e.g., ‘‘interested’’) and 10 items measuring current negative affect (e.g., ‘‘irritable’’), on a 5-point scale (1 = very slightly or not at all and 5 extremely). Since negative affect due to unpleasant experiences often triggers aggression (Anderson & Bushman, 2002), the mood measure was employed after the depletion task and then again after the TCRT task. In the current sample, both mood subscales after the depletion task showed good internal consistency (a = .87 and .72 for PA and NA, respectively) but were not significantly correlated (r = .14, p = .15, 95% CI = .02 to .35). Both mood subscales also showed good internal consistency (a = .91 and .77 for PA and NA, respectively) and were positively correlated (r = .63, p < .001, 95% CI = .36 to .81) when given after the TCRT task. The efficacy of the mindfulness induction was assessed with Toronto Mindfulness Scale (TMS: Lau et al., 2006), which measures the action of experiencing the present with (i) curiosity (6 items, e.g., ‘‘I was curious about my reactions to things’’) and (ii) decentering (7 items, e.g., ‘‘I was aware of my thoughts and feelings without overidentifying with them’’), on a 5point Likert-scale (0 = not at all and 4 = very much). In the current sample, both mindfulness subscales showed good internal consistency (a = .84 and .71 for curiosity and decentering, respectively) and were positively correlated (r = .58, p < .001, 95% CI = .43 to .71). The efficacy of the depletion manipulation was assessed with two following questions (i) ‘‘How difficult was it for you to follow the instructions during the video clip’’, and (ii) ‘‘How much did you have to control your attention to follow the instructions during the video clip’’, on a 7 point Likert-scale (1 = not at all and 7 = very much).
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
129
2.3.2. Ego-depletion task Depletion was induced using the dual-task self-control procedure, in which participants in the depletion group were asked to perform two unrelated tasks requiring self-control, whereas participants in the no-depletion condition were also asked to perform two unrelated tasks, but only the second task required self-control (Baumeister et al., 2007). For the first self-control task, we used an attention control task (DeWall et al., 2007; Finkel et al., 2009). All participants were asked to watch a 6-min videotape without audio of a woman being interviewed by an off-camera interviewer, and were told that they would later be making judgments of the interviewee. During the video clip, a series of common one-syllable words (e.g., ‘‘glue’’) appeared at the bottom of the screen for 10 s each. Participants in the depletion condition were instructed ‘‘not to read or look at any words that may appear on the screen’’, and to redirect their gaze to the interviewee’s face immediately if they caught themselves looking at the words. Participants in the no-depletion condition were not given any such instructions and were not told in advance that there would be words at the bottom of the screen. For the second self-control task, we used performance in a handgrip task (see Section 2.3.5). 2.3.3. Mindfulness induction Participants in the mindfulness condition were guided with a 15-min audio instruction for ‘‘Mindfulness of body and breath’’ (Williams & Penman, 2011) to direct their attention towards witnessing the full sensations of breathing without the intention of altering these experiences, and to notice in an accepting manner when their minds wander and gently return their focus to their breathing. This task has been used in Mindfulness-Based Cognitive Therapy (MBCT: Oxford Mindfulness Centre, 2002) and in past laboratory research (Kramer, Weger, & Sharma, 2013). Mindfulness with regards to the body and breath is used because the behaviour can be observed for a fairly extended period by participants without creating too much stress for novice practitioners (Sujiva, 2000). For a total 15 min, participants in the control condition listened to two neutral educational excerpts and composed all the possible words that came to mind using 100 tile of letters from a scrabble set. A similar procedure has been used elsewhere (Erisman & Roemer, 2010; Kramer et al., 2013) as a control condition for mindfulness induction. Although this type of task still requires participants to concentrate, it is unlikely to induce additional depleting effects because it does not require effortful suppression of an impulse (Baumeister et al., 2007) and is neither high in difficulty nor complexity (Hagger et al., 2010). 2.3.4. Adapted Taylor Competitive Reaction Time (TCRT) task The adapted TCRT task (Lawrence & Hutchinson, 2013; Lawrence & Hutchinson, 2014) was presented to participants on computer using E-Prime software version 2.0. A similar paradigm has been used in studies measuring the effect of depletion (DeWall et al., 2007; Finkel et al., 2009) and mindfulness induction (Heppner et al., 2008) on aggression. Participants were informed that they would be playing a computer-based competition task (i.e., hitting the spacebar on the computer keyboard as soon as a white circle stimulus appears on the screen) against an opponent located in a nearby room. In reality there was no opponent and the outcomes of the task was pre-programmed. Each time participants ‘won’ a trial, they could select a level of noise blast to be delivered to the opponent, ranging from 0 (no blast, non-aggressive option) through to 8 (maximum blast). When they ‘lost’ a trial, they received a noise blast through surround earphones for a fixed duration of 500 ms, at a pre-programmed level ostensibly chosen by the opponent (up to a maximum of 90 db). The TCRT task comprised a total of 81 trials. Participants always won the first trial to assess the baseline blast intensity that they delivered to the opponent prior to any provocation from the opponent (no provocation trial). On subsequent losing trials, they received blasts in two blocks of (i) low/moderate intensity (40 trials, participants lost 50% of time at random and received blast levels 1–4: low/moderate provocation) and (ii) higher intensity (40 trials, participants lost 50% of time at random and received blast levels 5–8: high provocation). The intensity of blasts delivered to the opponent under each block of trials (no provocation, mean low/moderate provocation, and mean high provocation) was used as a measure of participants’ aggression. 2.3.5. Handgrip task A manual handgrip (consisting of two handles and a connecting metal spring) was used as the second self-control measure. This measure compromises resisting discomfort in the forearm muscles and the resulting urge to quit (Baumeister et al., 2007). A small wad of paper was inserted between the two handles when participants gripped them together. Participants were required to continuously squeeze the handgrip using the dominant hand until the wad of paper fell out. Participants performed this task at three time points: at the beginning of the study (baseline handgrip performance), prior to the TCRT task (as a measure of performance in the second self-control task), and after the TCRT task. Handgrip performance was calculated as the duration time(s) elapsing before the wad of paper fell. Crucially, participants were not informed about the number of handgrip task they would be required to perform, since the expectation of further self-control acts may increase the depletion effects (see Hagger et al., 2010).
130
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
3. Results 3.1. Manipulation checks Unless specified, all analyses were performed using 5,000 bootstrap resamples (N = 110) with 95% bias-corrected confidence intervals. A bootstrap sampling distribution is estimated with replacement from the original sample; therefore, it reduces the impact of anomalies and helps ensure the stability of the models (Hayes, 2012a). The depletion manipulation was partly effective. Participants in the depletion condition (n = 55) reported having controlled their attention to a greater extent than no-depletion participants (n = 55; t(108) = 2.26, p = .03, 95% CI = 1.31 to .09; M depletion = 4.91, SD = 1.78 vs. M no-depletion = 4.20, SD = 1.50), but did not rate the task as more difficult than those in the no-depletion condition (t(108) = .42, p = .68, 95% CI = .52 to .84; M depletion = 3.31, SD = 1.95 vs. M no-depletion = 3.45, SD = 1.69). Participants receiving mindfulness induction (n = 55) reported higher decentering than participants not receiving mindfulness induction (n = 55; t(105.77) = 2.65, p = .01, 95% CI = .58 to .09; M mindfulness = 3.03, SD = .60 vs. M no-mindfulness = 2.70, SD = .70), but did not show higher levels of curiosity (t(117) = 1.15, p = .25, 95% CI = .50 to .13; M mindfulness = 3.50, SD = .81 vs. M no-mindfulness = 3.31, SD = .88). Participants in the depletion condition reported no differences than non-depleted participants in negative affect (t(108) = .57, p = .57, 95% CI = .16 to .09; M depletion = 1.26, SD = .38 vs. M no-depletion = 1.23, SD = .24), but scored higher in positive affect (t(108) = 2.24, p = .03, 95% CI = .55 to .03; M depletion = 2.66, SD = .68 vs. M no-depletion = 2.37, SD = .70). This positive affect, however, was not related to aggression under any of the three provocation levels (rs < .10, ps > .29) or to handgrip performance (r = .07 and .08 at pre-TCRT task and post-TCRT task, all controlling for baseline handgrip duration time, ps > .42). No differences were reported between the groups in positive (ps > .23) or negative affect (ps > .40) when the mood scale was given again after the TCRT task. Thus, mood fluctuations were not related to the experimental outcomes. 3.2. Association between aggression and handgrip performance Table 1 depicts correlations between the dependent measures, with handgrip performance reported controlling for baseline handgrip duration time (M baseline = 51.65, SD = 39.63). Participants’ blast intensities to the opponent were positively related to each other under all provocation levels (ps < .001). Participants’ measures of handgrip performance were also positively related to each other (ps < .001). Handgrip performance pre-TCRT task was associated, negatively, with blast intensity under no provocation trial. This indicates that those who were more able to maintain their baseline handgrip performance were also less aggressive prior to the provocation procedure. The correlations between blast intensity and handgrip performance did not occur after the TCRT task (ps > .33). 3.3. Moderation of mindfulness induction on the link between depletion and aggression As typically found in TCRT studies varying levels of provocation from the opponent (see Lawrence & Hutchinson, 2013; Lawrence & Hutchinson, 2014), participants delivered more intense blasts following high provocation compared to low provocations (p < .001), and following low provocation compared to no provocation (p < .001; see Table 1 for mean differences in blast intensity). The moderating effect of mindfulness induction on the link between depletion and aggression was tested at each provocation level using a bootstrap method (Hayes, 2012b). Theoretically, X (depletion) is depicted to exert an influence on Y (aggression), and this effect is proposed to be influenced by M (mindfulness induction). These effects are estimated mathematically in the form of a linear equation: Y = i + (c1 + c3M)X + c2M + ey. The (c1 + c3M) function represents the conditional Table 1 Correlations between dependent measures. Measures
Blast intensity under no provocation (1) Blast intensity under low/moderate provocation (2) Blast intensity under high provocation (3) Handgrip performance pre-TCRT task (4) Handgrip performance post-TCRT task (5) M (SD) absolute scores
Bias-corrected bootstrap 95% CI 1
2
3
4
5
[1.00, 1.00] [0.65, 0.83]*** [0.40, 0.66]*** [ 0.41, 0.04]** [ 0.27, 0.08] 2.61 (2.50)
[1.00, 1.00] [0.77, 0.91]*** [ 0.37, 0.04] [ 0.31, 0.15] 3.39 (2.00)
[1.00, 1.00] [ 0.35, 0.09] [ 0.32, 0.13] 4.14 (2.17)
[1.00, 1.00] [0.26, 0.73]*** 42.38 (30.89)
[1.00, 1.00] 43.18 (31.05)
Note: Handgrip performance reported controlling for baseline handgrip duration time(s). Higher positive scores indicate greater aggression and better physical (handgrip) stamina. ** p < .01. *** p < .001.
131
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
effect of X on Y, in which c1 estimates the effect of X on Y when M = 0, and c3 estimates how much the effect of X on Y changes as M changes by one unit. The main focus in a moderation model is whether c3, the interaction coefficient between X and M, is statistically different from zero (for further details, consult Hayes, 2012a). Findings shown in Table 2 support Hypothesis 1, in that participants’ aggression was affected by a main effect of depletion. Specifically, compared to non-depleted participants, depleted participants delivered higher levels of blast intensity as compared to non-depleted participants under no provocation and low/moderate provocation trials (p < .01). The main effect of mindfulness was non-significant across all three levels of provocation (ps > .78). Importantly, Hypothesis 2 was supported, such that the link between depletion and blast intensity was moderated by mindfulness induction under no provocation and low/moderate provocation trials. Under no provocation and low/moderate provocation, depleted participants who received mindfulness induction delivered lower levels of blast intensity than depleted participant without mindfulness induction (all ps = .04), whereas non-depleted participants with mindfulness induction delivered comparable blast intensities to non-depleted participants without mindfulness induction (see Fig. 2). The moderating effect of mindfulness diminished under conditions of high provocation (p = .13). 3.4. Moderation of mindfulness induction on the link between depletion and handgrip performance In the ego-depletion literature, handgrip performance is commonly calculated by subtracting each participant’s baseline duration time from their subsequent duration time(s) (see Hagger et al., 2010). A negative change score indicates that the participant is less able to maintain baseline handgrip duration, as would be expected if the person is under depleted condition. In this current study, residualising handgrip scores pre-TCRT task and post-TCRT task by baseline duration time resulted in significant correlations with the baseline score (ps < .001). When such regression towards the mean influences the measurement then covariance analysis is more recommended than analysis of change scores (Bonate, 2000). Thus, to analyse the effect of mindfulness on post-depletion handgrip performance, we used the same bootstrap procedure while controlling for baseline handgrip duration time as a covariate. The findings shown in Table 2 reveal that the standard main effect of depletion on handgrip performance was found prior to the TCRT task (Hypothesis 3a; p = .01), but not after the TCRT task (Hypothesis 3b; p > .75). No support was found for the moderation of mindfulness on the depletion and handgrip performance link at both time points (Hypotheses 4a and 4b; ps > .38). Unexpectedly, after the TCRT task, mindfulness induced participants became more able to maintain their baseline handgrip performance independent of depletion condition (p = .03). 4. Discussions This study is the first to examine the effect of a brief laboratory mindfulness induction on aggressive behaviour using a self-control resource depletion model. It extends past research on the role of mindfulness induction on provoked aggression (Heppner et al., 2008) by showing that mindfulness could be useful also to reduce post-depletion aggression. This finding is important as self-control failure has been acknowledged to increase the association between aggressive triggers and various types of aggressive behaviour (DeWall et al., 2007; Finkel et al., 2009; Stucke & Baumeister, 2006). In this study, we specifically predicted that the main effect of depletion on aggression, and the moderation of mindfulness on the depletion-aggression link would vary as a function of the level of provocation. In line with the TAAS model (Marshall & Brown, 2006), we found that in response to low/moderate provocation, the depletion effect was moderated by mindfulness induction. The descriptive pattern was similar but less strong following high provocation trials, as in previous research using the TCRT task (e.g., Lawrence & Hutchinson, 2014). Table 2 Moderation with bootstrap method, using depletion as predictor, mindfulness as moderator, blast intensity and handgrip performance as outcomes. Model tested
Blast intensity No provocation
Depletion (c1) Mindfulness induction (c2) Depletion x mindfulness (c3) R2 on interaction (%) Conditional effect of depletion without mindfulness Conditional effect of depletion with mindfulness
Handgrip performance
B (SE)
Low/moderate provocation B (SE)
High provocation B (SE)
Pre-TCRT task B (SE)
Post-TCRT task B (SE)
2.19*** (0.64) 0.00 (0.64) 1.90* (0.90) 3.63%* 2.19*** (0.64) 0.29 (0.64)
1.40 (0.52)** 0.15 (0.52) 1.55* (0.74) 3.79%* 1.40** (0.52) 0.15 (0.52)
1.04 (0.58) 0.01 (0.58) 1.25 (0.82) 2.07% 1.04 (0.58) 0.20 (0.58)
9.49 (3.75)** 5.00 (3.75) 4.70 (5.30) 0.002% 9.49 (3.75) 4.79 (3.74)
1.12 (3.58) 7.70 (3.58)* 2.65 (5.06) 0.001% 1.12 (3.58) 3.77 (3.57)
Note: Handgrip performance reported controlling for baseline handgrip duration time(s). Higher positive scores indicate greater aggression and better physical (handgrip) stamina. Coding: 1 = depletion or mindfulness induction condition; 0 = non-depletion or no mindfulness induction condition. B = unstandardised regression coefficient; SE = standard error; R2 = variance increase * p < .05. ** p < .01. *** p < .001.
132
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Fig. 2. Mindfulness induction depletion effect on blast intensity across provocation levels. Error bars indicate standard error of mean, asterisk indicates significant (p < .05) decrease in blast intensity. Scale goes up to 8 as a maximum.
To scrutinise the role of mindfulness induction across different tasks of self-control, this study incorporated a measure of physical stamina in a handgrip task. The initial analysis indicated that participants who performed better on the handgrip task also tend to be less aggressive. However, the role that mindfulness had on these two self-control tasks (aggression and handgrip) appeared to differ. While the effect of mindfulness on aggression was more observable amongst those egodepleted, its benefit on handgrip performance was not dependent upon individuals having been subjected to the depletion. The main effect of mindfulness induction occurred on handgrip performance in particular when people have experienced provocation (after the TCRT task) instead. As suggested by Friese et al. (2012), brief experimental manipulations will be more effective in increasing self-control performance in individuals who are under the influence of various factors that trigger impulsive responding. Thus in this study, mindfulness induction may have improved handgrip performance after the TCRT task because the provocation experienced during this task acted as one such trigger of impulsive responding. Based on the resource depletion model, one would hypothesize that the depletion effect should become increasingly large over the course of multiple self-control tasks. In contrast, at the end of this study the negative effect of depletion in relation to handgrip performance appeared to diminish, resulting in a comparable level of final performance between depletion/control groups. One possible explanation is that instead of the depletion effect, an adaptation effect might have occurred with repeated self-control acts. Indeed, past study has shown that for participants who were required to perform an additional third self-control task after the second self-control task, their performance might remain stable or even improve (Converse & DeShon, 2009). Another possibility is that the depletion effect could have been attenuated by participants’ motivational factors. Because all participants were told that the final measure of handgrip was their last task to be accomplished, they became sufficiently motivated to accomplish this task and performed equally effective regardless of their depletion condition. Relatedly, this could be due to conceptual differences between the two self-control tasks used in this study. The handgrip task explicitly required physical muscle control as well as motivational control (they were explicitly told to keep the paper wad from falling for as long as possible), whereas the TCRT task required no such explicit instruction to inhibit aggressive behaviour. From a recent motivational viewpoint of ego-depletion, self-control failure is seen as a motivational deficit to perform a demanding task rather than a temporary loss of some limited resources or energy (Inzlicht, Schmeichel, & Macrae, 2014). More research is needed to explore the dynamics of self-control over time as well as non resource-based accounts of self-control. Our findings are important also as it could have been predicted that a mindfulness meditation task such as the one used in this study could be expected to add to the existing self-control demands because the manipulation is, essentially, an attention control task (e.g., Masicampo & Baumeister, 2007). It has been speculated that to be able to sustain attention on the present moment, individuals naïve to mindfulness practices might need to exert cognitive control over their emotional reactions (Holzel et al., 2011). Instead, in the present study and in the work by Friese et al. (2012), such mindfulness manipulations counteract depletion. Thus, it is timely to identify the underlying emotional, cognitive, and biological processes that might come into play in the association between mindfulness, self-control, and aggression. Possible mediating mechanisms such as the mindfulness acceptance of internal experiences, nonattachment, and clarification of values (Baer, 2003; Shapiro et al., 2006) should be investigated. Importantly, previous research has shown that depleted individuals who were asked to monitor their performance (i.e., using an accurate clock) or who were already high in trait self-monitoring persisted longer on subsequent self-control task (Wan & Sternhal, 2008). Although mindfulness appears to share considerable variance with self-monitoring, the focus of the latter construct is on attention towards aspects of the self, whereas mindfulness simply
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
133
offers attention on the ‘in-the-moment’ experience (Brown et al., 2007a; Brown et al., 2007b) which can extend beyond the self. Teasing the monitoring function of mindfulness from a comparative self-monitoring against some desirable standards is important for future studies. It should be emphasised, however, that for an exercise to count as a true mindfulness meditation, some personal experience with mindfulness exercises may be necessary. Although most participants seemed to be serious in their attempts to engage with the task, and the manipulation check did reveal that participants in the mindfulness condition reported higher levels of decentering, the brief exercise might have not been sufficiently strong to evoke the whole range of state mindfulness (see also Erisman & Roemer, 2010). Decentering, or observing one’s experience with some distance rather than being carried away by thoughts and feelings, was previously shown to predict lower perceived stress and distress (Lau et al., 2006). So it could be that the cultivation of decentering might have enabled individuals to become more tolerant to unpleasant experiences related to depletion and provocation. The current study used neutral educational information as control condition for the mindfulness manipulation. Although a similar control has been employed in past research on mindfulness (Erisman & Roemer, 2010; Kramer et al., 2013) and egodepletion (Friese et al. 2012), ideally a control group that has the same expectation of improvement as the experimental group would be more effective as an active control (Boot, Simons, Stothart, & Stutts, 2013). As a result, we cannot thoroughly rule out the possibility of placebo or demand effects contributing to the results in this study. Future research could employ other types of meditation or use alternative emotion regulation strategies (e.g., cognitive reappraisal, distraction, unfocused attention). These latter conditions were not employed in this study because regulating emotions risked depleting participants’ self-control ability (Baumeister et al., 2007). Although we included a manipulation check to assess the efficacy of the mindfulness induction, no measure of expectation effects of the mindfulness induction or control condition was included. This was in order to avoid sensitising participants into linking the mindfulness procedure to the aggression and handgrip tasks and leading to possible demand characteristics. Nonetheless, it is possible that participants may have implicitly expected a ‘self-improvement’ effect of the mindfulness induction that could explain the results. Finally, this study consisted of entirely British university students naïve to mindfulness techniques. Replication with different populations is necessary to examine whether the differential association between mindfulness, self-control, and aggression is robust. Despite the limitations, findings from the present study contribute to the literature by showing that the benefits of mindfulness on post-depletion aggression can be demonstrated outside extensive prior training in mindfulness. Additionally, the benefit of short mindfulness induction on a behavioural self-control task appears to hold up after people have also experienced provocation from others. The inclusion of mindfulness interventions for those who have a particular difficulty in effortful control of aggressive impulses should now be examined further. Conflicts of interest The authors have no conflicts of interest to declare. Acknowledgment The authors would like to thank the anonymous reviewers for their thorough feedbacks. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.concog.2014.12.008. References Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology, 53(1), 27–51. http://dx.doi.org/10.1146/ annurev.psych.53.100901.135231. Anonymous (2012). Udana: Exclamations (Thanissaro Bhikkhu, Trans.). Valley Center, CA: Metta Forest Monastery. Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10(2), 125–143. http://dx.doi.org/10.1093/clipsy.bpg015. Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control. Current Directions in Psychological Science, 16(6), 351–355. http:// dx.doi.org/10.1111/j.1467-8721.2007.00534.x. Bergeron, N., & Schneider, B. H. (2005). Explaining cross-national differences in peer-directed aggression: A quantitative synthesis. Aggressive Behavior, 31(2), 116–137. http://dx.doi.org/10.1002/ab.20049. Bodhi, B. (2000). A comprehensive manual of Adhidhamma. Seattle: BPS Pariyatti. Bonate, P. L. (2000). Analysis of pretest-posttest designs. Boca Raton, Florida: Chapman & Hall/CRC. Boot, W. R., Simons, D. J., Stothart, C., & Stutts, C. (2013). The pervasive problem with placebos in psychology: Why active control groups are not sufficient to rule out placebo effects. Perspectives on Psychological Science, 8(4), 445–454. http://dx.doi.org/10.1177/1745691613491271. Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007a). Addressing fundamental questions about mindfulness. Psychological Inquiry, 18(4), 272–281. http:// dx.doi.org/10.1080/10478400701703344. Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007b). Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry, 18(4), 211–237. http://dx.doi.org/10.1080/10478400701598298.
134
C. Yusainy, C. Lawrence / Consciousness and Cognition 33 (2015) 125–134
Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality-social, clinical and health psychology. Psychological Bulletin, 92(1), 111–135. http://dx.doi.org/10.1037/0033-2909.92.1.111. Converse, P. D., & DeShon, R. P. (2009). A tale of two tasks: Reversing the self-regulatory resource depletion effect. Journal of Applied Psychology, 94(5), 1318–1324. http://dx.doi.org/10.1037/a0014604. DeWall, C. N., Baumeister, R. F., Stillman, T., & Gailliot, M. T. (2007). Violence restrained: Effects of self-regulation and its depletion on aggression. Journal of Experimental Social Psychology, 43(1), 62–76. http://dx.doi.org/10.1016/j.jesp.2005.12.005. Duckworth, A. L., & Kern, M. L. (2011). A meta-analysis of the convergent validity of self-control measures. Journal of Research in Personality, 45(3), 259–268. http://dx.doi.org/10.1016/j.jrp.2011.02.004. Erisman, S. M., & Roemer, L. (2010). A preliminary investigation of the effects of experimentally induced mindfulness on emotional responding to film clips. Emotion, 10(1), 72–82. http://dx.doi.org/10.1037/a0017162. Finkel, E. J., DeWall, C. N., Slotter, E. B., Oaten, M., & Foshee, V. A. (2009). Self-regulatory failure and intimate partner violence perpetration. Journal of Personality and Social Psychology, 97(3), 483–499. http://dx.doi.org/10.1037/a0015433. Fix, R. L., & Fix, S. T. (2013). The effects of mindfulness-based treatments for aggression: A critical review. Aggression and Violent Behavior, 18(2), 219–227. http://dx.doi.org/10.1016/j.avb.2012.11.009. Friese, M., Messner, C., & Schaffner, Y. (2012). Mindfulness meditation counteracts self-control depletion. Consciousness and Cognition, 21(2), 1016–1022. http://dx.doi.org/10.1016/j.concog.2012.01.00. Giancola, P. R., & Parrott, D. J. (2008). Further evidence for the validity of the Taylor aggression paradigm. Aggressive Behavior, 34(2), 214–229. http:// dx.doi.org/10.1002/ab.20235. Hagger, M. S., Wood, C., Chris, S., & Chatzisarantis, N. L. D. (2010). Ego-depletion and the strength model of self-control: A meta-analysis. Psychological Bulletin, 136(4), 495–525. http://dx.doi.org/10.1037/a0019486. Hayes, A. F. (2012a). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from