The protective effects of brief mindfulness meditation training

Consciousness and Cognition 33 (2015) 277–285

Contents lists available at ScienceDirect

Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog

The protective effects of brief mindfulness meditation training Jonathan B. Banks ⇑, Matthew S. Welhaf, Alexandra Srour Division of Social and Behavioral Sciences, Nova Southeastern University, United States

a r t i c l e

i n f o

Article history: Received 15 August 2014

Keywords: Working memory Mind wandering Mindfulness meditation Stress Training

a b s t r a c t Mindfulness meditation has gained a great deal of attention in recent years due to the variety of physical and psychological benefits, including improved working memory, decreased mind wandering and reduced impact of stress on working memory. The current study examined a 1-week at home mindfulness meditation intervention compared to an active control intervention. Results suggest that mindfulness meditation does not increase working memory or decrease mind wandering but does prevent stress related working memory impairments. Mindfulness meditation appears to alter the factors that impair working memory such that the negative impact of mind wandering on working memory was only evident at higher levels of negative affect. The use of cognitive mechanism words in narratives of stressful events did not differ by condition but predicted poorer working memory in the control condition. The results support the use of an at home mindfulness meditation intervention for reducing stress-related impairments. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Mindfulness meditative practices focus on experiencing present-moment awareness without attempts to suppress, judge, or emotionally react to any stimuli (Kabat-Zinn, 1994). A wide variety of practices can be considered mindfulness meditation. However, these practices can be classified in two major groups that attempt to improve awareness through increases in focused attention and through open monitoring of sensations without reactivity to the sensations or thoughts (Davidson & Goleman, 1977). Mindfulness meditation techniques designed to improve focused attention include instructing participants to focus on their breath or bodily sensation. Techniques designed to improve open monitoring strategies focus on improving the individual’s ability to experience thoughts without attempts to suppress or react to unwanted thoughts (Lutz, Slagter, Dunne, & Davidson, 2008). Mindfulness based meditation impacts a variety of cognitive functions, including reductions in the impact of repetitive thoughts (Feldman, Greeson, & Seville, 2010), decreases in emotional interference on cognitive tasks (Ortner, Kilner, & Zelazo, 2007), increases in sustained attention (Morrison, Goolsarran, Rogers, & Jha, 2014), mood and attention (Baer, 2009; Jha, Krompinger, & Baime, 2007), working memory (Jha, Stanley, Kiyonaga, Wong, & Gelfand, 2010; Mrazek, Franklin, Phillips, Baird, & Schooler, 2013), and decreases in mind wandering (Morrison et al., 2014; Mrazek et al., 2013). Of specific interest to the current study are changes in working memory (WM) and mind wandering. Impairments in WM have been shown to occur following psychosocial stressors (Oei, Everaerd, Elzinga, Van Well, & Bermond, 2006) and self-reported negative life stress (Klein & Boals, 2001). Psychological stress related WM impairments may be due to intrusive thoughts regarding the negative life event (Klein & Boals, 2001). Intrusive thoughts or a more general ⇑ Corresponding author at: Division of Social and Behavioral Sciences, Nova Southeastern University, 3301 College Avenue, Fort Lauderdale, FL 33314, United States. Fax: +1 954 262 3760. E-mail address: [email protected] (J.B. Banks). http://dx.doi.org/10.1016/j.concog.2015.01.016 1053-8100/Ó 2015 Elsevier Inc. All rights reserved.

278

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

form of task-unrelated thoughts (TUTs), mind wandering, impairs performance on WM (Klein & Boals, 2001) and sustained attention tasks (Banks, Tartar, & Welhaf, 2014; McVay & Kane, 2009). Negative affect may be responsible for impaired WM following a psychological stressor. Stressful events result in increases in negative affect (Stawski, Sliwinski, Almeida, & Smyth, 2008). Increases in negative affect increase mind wandering (Smallwood, Fitzgerald, Miles, & Phillips, 2009) and impair WM (Curci, Lanciano, Soleti, & Rime, 2013). It is possible the impact of mind wandering on WM is greatest when negative affect is high. Interventions that prevent increases in negative affect or alter the impact of mind wandering on working memory may prevent impairment in working memory from occurring. Mindfulness meditation may help prevent stress related decreases in WM. Specifically, mindfulness meditation that aims to improve open monitoring of thought may provide a greater reduction in stress related decreases in WM. An 8-week brief mindfulness meditation training with an active duty military cohort during a stressful pre-deployment period, demonstrated improvements in WM in individuals that practiced the meditation intervention most (Jha et al., 2010). Cognitive benefits resulting from mindfulness meditation interventions may be due to reductions in mind wandering. A 2-week mindfulness training with college students evidenced increases in WM, GRE reading comprehension scores and decreases in TUTs during the GRE task compared to participants in a control condition who completed a nutrition class (Mrazek et al., 2013). WM performance improvements were mediated by reductions in TUTs. Similar increases in sustained attention and decreases in TUTs were observed following a 7-week training, but no increases in WM occurred (Morrison et al., 2014). Although prior work (Mrazek et al., 2013) has provided important clues as to the mechanism by which mindfulness meditation improves cognitive functioning, the control group did not sufficiently control for participant expectancy. Controlling participant expectancy is critical for studies examining cognitive benefits of mindfulness meditation because the intervention stresses focus of attention and minimizing distractions. The use of an active control reduces participant expectancy concerns only when expectation of improvement is similar across conditions (Boot, Simons, Stothart, & Stutts, 2013). The current study will address this concern by using an active control group that produces similar participant expectations. The current study focused on examining the impact of a mindfulness meditation training (MMT) intervention on WM and mind wandering. To this end, we examined the impact of an acute 15-min intervention on WM and TUTs, in comparison to a relaxation training (RT) intervention designed to generate equivalent expectancy. The two interventions are similar in that they both include focused attention meditation practices that instructed the individual to focus on their breath during the training. However, the MMT extends beyond the breathing techniques and includes open monitoring meditation training. Specifically, participants were instructed to notice instances of mind wandering in a non-judgmental manner and redirect attention rather than suppress the thought. We hypothesized that MMT condition participants would evidence increases in WM and reductions in TUTs following the 15-min intervention but similar changes would not occur in the RT condition. We were also interested in examining the impact of a 1-week MMT completed at home. We hypothesized MMT condition participants would experience further increases in WM and decreases in TUTs, while similar changes would not occur in the RT condition. Finally, we examined the ability of a brief mindfulness meditation to prevent the negative impact of stress on WM. Participants engaged in a brief psychological stressor involving writing about a personal negative life event. We hypothesized that MMT condition participants would not experience changes in WM or TUTs following the psychological stressor. However, we hypothesized that RT condition participants would experience decrements in WM and increases in TUTs following the stress manipulation. Specifically, we hypothesized that following the stressor, impairments in WM would be predicted by rates of TUTs and negative affect in the RT but not MMT condition. The written narratives produced during the stressor will be examined to determine if word use predicts changes in WM. Narratives about stressful events include a greater percentage of cognitive mechanism words (e.g. words about cause and insight) than narratives about less stressful events (Boals & Klein, 2005). It has been argued that higher levels of cognitive mechanism word use reflects an active search for meaning (Klein & Boals, 2010) and an attempt to engage in a meaning making process (Boals, Banks, Hathaway, & Schuettler, 2011). We believe the search for meaning will result in impaired WM due to intrusive thoughts about the event rather than focus on a subsequent task. Therefore, we hypothesized that rates of cognitive mechanism words would be negatively related with WM following the writing in the RT but not the MMT condition. 2. Methods 2.1. Participants Eighty meditation-naive undergraduate students from Nova Southeastern University participated for partial course credit and were randomly assigned to the MMT condition (n = 40) or RT condition (n = 40). Sixteen participants failed to return within the required 10-days between sessions and two participant’s Session 2 data was lost due to computer malfunction. Resulting in a sample of 62 participants at Session 2 (48 Female) with 34 in the MMT condition (M = 21.91 years old, SD = 7.89; 25 Female; 17 Caucasian, 10 Hispanic, 6 African–American, 1 other) and 28 in the RT condition (M = 19.82 years old, SD = 4.21; 22 Female; 13 Caucasian, 7 Hispanic, 5 African–American, 2 Asian–Pacific Islander, 1 other). Participants who failed to return did not differ from those who returned on any Session 1 measures, p’s > .05. Procedures were approved by the Nova Southeastern University Institutional Review Board. Two exclusion criteria were applied to the current sample.

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

279

Participants could not have prior meditation experience and had to be current students at Nova Southeastern University. All participants reported being students and no prior meditation experience. 2.2. Measures 2.2.1. Automated Operation Span Task (AOSPAN; Unsworth, Heitz, Schrock, & Engle, 2005) To assess WM, participants completed the AOSPAN. During the AOSPAN participants verify the accuracy of a solution to a math problem (e.g. (2 ⁄ 5) + 3 = ?; 7) and are presented with a capital letter (out of 12 possible letters). Following each trial, of 3–7 verification-letter pairs, participants identified, via mouse click, letters in the trial in serial order from a grid of 12 possible letters. The AOSPAN was scored by summing the number of letters recalled in correct serial position (Conway et al., 2005). 2.2.2. Thought probes (McVay & Kane, 2009) 15 Thought probes were inserted into the AOSPAN following letter recall grids to measure task-unrelated thoughts (TUTs). Participants responded to the prompt, ‘‘What were you just thinking about,’’ by selecting one of the response options: ‘‘a. Task-related thoughts’’ pertaining to the current task of letter recall, ‘‘b. Task-related evaluative thoughts-positive’’ ‘‘c. Task-related evaluative thoughts-negative,’’ ‘‘d. Task-unrelated thoughts, neutral content,’’ ‘‘e. Task-unrelated thoughts, positive content,’’ and ‘‘f. Task-unrelated thoughts, negative content,’’ and ‘‘g. Task-unrelated thoughts, about the writing task.’’ Percentage of TUTs was calculated by summing the number of off-task response choices (options d–g) selected and dividing by the number of probes. 2.2.3. Dundee Stress State Questionnaire (DSSQ; Matthews et al., 1999) The DSSQ is a retrospective measure used to assess subjective states in response to stressful environments. Participants completed the thinking content (TC) scale, retrospective measure of TUTs (DSSQTUTs), and the thinking style (TS) scales, including Self-focused Attention, Self-esteem, Concentration, Control and Confidence (Matthews et al., 2002). Participants responded to questions in reference to thoughts during the AOSPAN (e.g. ‘‘I thought about personal worries’’) on a 5 point Likert-type scale, from 1 (Never) to 5 (Very Often). The Thinking Content and Thinking Styles demonstrated reasonable internal consistency in the current sample at all time points (a = .75–.88 and a = .82–.90, respectively). 2.2.4. Five-Facet Mindfulness Questionnaire (FFMQ; Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006) The 39-item self-report measure of mindfulness is composed of five subscales: observing, describing, acting with awareness, non-judging of inner experience, and non-reactivity to inner experience. Participants responded to questions (e.g. ‘‘I watch my feelings without getting lost in them’’) on a 5-point Likert-type scale, from 1 (never or very rarely true) to 5 (very often or always true). Internal consistency in the current sample was acceptable at both time points for all subscales (a = .64–.91). 2.2.5. Positive and negative affect schedule (PANAS; Watson, Clark, & Tellegen, 1988) The 20-item questionnaire measures positive and negative affect. Participants rated their current mood in relation to items on a 5-point Likert-type scale, from 1 (Very slightly or not at all) to 5 (Extremely). Separate scores for positive and negative affect were calculated, with higher scores indicated greater affect. This scale yields high inter-item correlations, in the current sample for positive affect (ranging from a = .90–.94) and negative affect (a = .80–.84). 2.2.6. Acceptance and Action Questionnaire (AAQ: Bond et al., 2011) The 7-item AAQ was used to assess psychological flexibility and self-regulation. Participants rated statements regarding their level of psychological flexibility (e.g. ‘‘worries get in the way of my success’’) on a 7-point Likert-type scale from 1 (Never true) to 7 (Always true). Scores were calculated by summing items; with higher scores indicate poorer psychological flexibility and self-regulation. The AAQ has been shown to have high reliability (.81–.87; Bond et al., 2011) and demonstrated high internal consistency (a = .86–.87) in the current sample at both time points. 2.2.7. Cognitive mechanism words Word use in participants’ narrative of the stressful event was examined using the Linguistic Inquiry and Word Count Program 2007 (LIWC2007; Pennebaker, Booth, & Francis, 2007). The LIWC program calculates percentage of cognitive mechanisms category words in the narrative. 2.2.8. Expectancy and compliance Two expectancy items were completed. Items asked participants, ‘‘To what extent [do/did] you expect that your performance on [a task or your ability to focus/the working memory task] [is likely to change/changed] as a result of participating in the study?’’ Participants responded on a 7-point Likert-type scale, ranging from 1 (I expected it to decrease significantly) to 7 (I expected it to increase significantly). Items were completed by a subset of the sample (Expectancy Session 1, n = 65; Expectancy Session 2, n = 50). No differences were found between conditions on either expectancy item, p’s > .05.

280

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

Two clarity and compliance questions (Feldman et al., 2010) were administered. The clarity prompt asked participants, ‘‘To what extent did you feel that the taped instructions were clear enough for you to understand what you were being asked to do.’’ The compliance prompt asked participants ‘‘To what extent did you follow the taped instructions.’’ Participants responded on a 7-point Likert-type scale, ranging from 1 (Not at All) to 7 (A Great Extent). In general, participants found the instructions clear (M = 6.47, SD = 0.74) and were able to follow them (M = 6.06, SD = 1.10), with no differences observed by condition, p’s > .05. 2.2.9. Visual analogue scale (VAS; Gaab, Rohleder, Nater, & Ehlert, 2005) The visual analogue scale was used to assess subjective stress. Participants indicated their response by marking a line with end points of ‘‘Not at all anxious’’ to ‘‘Very anxious.’’ Responses were calculated by measuring (centimeters) distance from the ‘‘Not at all anxious’’ end to the participant’s mark, with higher scores indicating greater anxiety. Items in this scale included ‘‘The past situation was stressful for me’’ and ‘‘I found the past situation to be a challenge.’’ 2.2.10. Demographics Participants completed a demographics questionnaire, assessing age, sex, and ethnicity. 2.2.11. Mindfulness Meditation Training (MMT) The MMT was based on previously used training scripts (Feldman et al., 2010) that were based on scripts used in mindfulness-based cognitive therapy (Segal, Williams, & Teasdale, 2002). The goal of the MMT, consistent with open monitoring meditation training, was to promote the awareness and acceptance of thoughts and emotions, rather than suppressing unwanted thoughts. Participants were instructed to focus their attention on breathing techniques and to notice when they experience any instance of mind wandering in an accepting and non-judgmental manner. Following the instance of mind wandering, the training encouraged the participant to return the focus of attention to the breathing techniques rather than actively suppressing the mind wandering. The training was delivered via a 15-min audio recording that guided participants to become aware of thoughts and emotions from moment-to-moment without focusing on any single thought or emotion (Feldman et al., 2010). 2.2.12. Relaxation Training (RT) The RT included progressive muscle relaxation and body-scan exercises used in prior studies as active control techniques (Feldman et al., 2010; Ortner et al., 2007). The 15-min RT audio recording guided participants through progressive muscle relaxation and body-scan techniques that directed attention to the body, and progressively tensing and relaxing each body part starting from the feet and ending at the head. The RT also includes instruction regarding focusing on breathing techniques to assist in the muscle relaxation and body-scan techniques. However, unlike the MMT, participants were not given any instruction regarding the occurrence of mind wandering. 2.2.13. Writing stressor To induce psychological stress participants nominated a current personal negative life event. Following the nomination, participants completed a 10-min expressive writing description of the event (Boals et al., 2011). Participants received the following instructions: You will now have ten minutes to write about the event that you just nominated. In your writing, I’d like you to really let go and explore your deepest emotions and thoughts about this specific issue. You may write about more than one aspect of the event. Some people find it best to explore their emotions and then try to come to an understanding of the event(s). Do your best to try to ‘‘tie it all together’’ at the end of your writing. We ask that you do not include any information that could identify you as we want this writing to remain confidential and for you to feel free to be able to write things you would not want tied to you. Only the researchers involved in this project will have access to read your essays. Therefore, your confidentiality is assured and your name will never be linked to anything you write. The important thing is that you really let go and dig down to your very deepest emotions and thoughts about the negative event and explore them in your writing. You have ten minutes to write about the nominated event. 2.3. Procedure Upon arrival, participants were randomly assigned to condition and completed informed consent forms. Participants completed self-report measures, including the AAQ, PANAS, FFMQ, and Expectancy item. Participants then completed the AOSPAN (with thought probes; AOSPAN Time 1) and DSSQ (TC and TS, Time 1) as baseline measures. Participants listened to the 15-min guided training recording based on condition. Following the training, participants completed the AOSPAN (with thought probes; AOSPAN Time 2) and DSSQ (TC, Time 2). Participants were given an audio recording of the training based on condition and instructed to practice the training at home in a quiet, private room at least four times during the week before the second session. Participants completed a daily time log to indicate the date and duration of each session at home. No differences were found between conditions on the number of practice sessions completed (M = 4.62 sessions, SD = 1.42), p > .05.

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

281

Seven to ten days following the first session (M = 7.06 days, SD = 0.90), participants returned and listened to the training recording to provide a brief ‘booster’. Participants then completed the AAQ, PANAS, FFMQ, and AOSPAN (with thought probes; AOSPAN Time 3) followed by the DSSQ (TC and TS, Time 3). Participants were then given 10 min to complete the writing stressor task. Finally, participants completed the AOSPAN (with thought probes; AOSPAN Time 4), DSSQ (TC, Time 4), PANAS, VAS, Expectancy, compliance, clarity and demographics measures. 3. Results To examine the first hypothesis, that mindfulness meditation training would increase WM and decrease mind wandering following the brief exposure in Session 1, three ANOVAs were conducted. No effects of time, condition, or interactions were found on the AOSPAN or TUTs, all p’s > .05. An effect of time was found on the DSSQ-TC-TUTs, F (1, 76) = 19.21, p < .0001, partial g2 = .25, but no effects of condition or interaction, p’s > .05. DSSQ-TC-TUTs decreased from the start of Session 1 (M = 11.42, SD = 5.32) to the end of Session 1 (M = 10.03, SD = 4.28). To examine the second hypothesis, that a 1-week mindfulness meditation training would increase WM and decrease mind wandering, scores on the AOSPAN, TUTs and DSSQ-TC-TUTs were compared from baseline measurement in Session 1 (Time 1) to initial measurement in Session 2 (Time 3). As seen in Fig. 1, a significant increase in AOSPAN scores was observed, F (1, 60) = 8.79, p < .01, partial g2 = .15, but no effects of condition or interaction were found, p’s > .05. No effects of time, condition, or interaction were observed for the TUTs or DSSQ-TC-TUTs, all p’s > .05. To examine changes on self-report measures of mindfulness, the AAQ, FFMQ, and DSSQ-TS, scores from Session 1 to Session 2 were compared. As seen in Table 1, a significant effect of time and a trend for a time by condition interaction, but no effect of condition, p > .05, was observed on the AAQ. Decreases in AAQ scores were observed in the MMT condition, t (34) = 3.94, p < .001, d = .40, 95% CI [1.12, 3.51] but not RT condition, t (27) = 0.63, p > .05, d = .008, 95% CI [ 1.12, 2.12]. A trend for a time by condition interaction was found for the DSSQ-TS-Control and Confidence subscale. MMT and RT conditions did not differ during Session 1, p > .05, however, during Session 2 DSSQ-TS-Control and Confidence scores were higher in the MMT condition than RT condition, t (61) = 2.60, p < .05, d = .66, 95% CI [1.05, 8.02]. Significant time effects were observed on the FFMQ-Act with Awareness subscale, FFMQ-Nonjudging subscale, DSSQ-TSSelf-esteem subscale, and DSSQ-TS-Self-focused Attention subscale. However, no time by condition interactions were observed on any of these measures, p’s > .05. No effects of time, condition or interactions were observed on the FFMQObserving, FFMQ-Describing, FFMQ-Nonreactivity, or DSSQ-TS-Concentration subscales, p’s > .05. The stress manipulation was effective at increasing negative affect, F (1, 60) = 17.27, p < .0001, partial g2 = .29, and decreasing positive affect. F (1, 60) = 8.60, p < .01, partial g2 = .14. No effects of condition or time by condition interactions were observed for positive or negative affect, p’s > .05. VAS item ratings did not differ by condition, p’s > .05, suggesting an equivalent perceived stress in each condition. Further, no differences were found by condition on cognitive mechanism word use, p > .05. To test the hypothesis the stress manipulation would impact the RT but not MMT condition, AOSPAN, TUTs, and DSSQ-TCTUTs prior to and following the stressor were examined. An effect of time, F (1, 60) = 6.53, p < .05, partial g2 = .34, condition, F (1, 60) = 6.72, p < .05, partial g2 = .11, and time by condition interaction, F (1, 60) = 6.53, p < .05, partial g2 = .34, were found for the AOSPAN. As seen in Fig. 1, the MMT and RT conditions did not differ prior to the stressor, p > .05, but AOSPAN scores were higher in the MMT condition following the stressor (M = 62.47, SD = 9.77) than RT condition (M = 52.29, SD = 17.06), F (1, 61) = 8.68, p < .01, partial g2 = .13. Increases in task-unrelated thoughts were observed on the DSSQ-TC-TUTs, F (1, 60) = 8.30, p < .01, partial g2 = .14, and a trend for a similar effect on TUTs, F (1, 60) = 3.34, p = .07, partial g2 = .06. No effects of condition or interactions were observed on the DSSQ-TC-TUTs or TUTs, p’s > .05. As hypothesized a negative relationship was found between TUTs and AOSPAN in the RT condition following the stressor, r (28) = .42, p < .05, but not the MMT condition, r (34) = .08, p > .05. These results suggest the MMT appears to have altered the relationship between WM and TUTs. To determine predictors of change in AOSPAN following the stress manipulation, we conducted a series of hierarchical multiple regression analyses predicting AOSPAN Time 4 scores, controlling for AOSPAN Time 3 scores. As seen in Table 2,

Fig. 1. AOSPAN scores by condition as a function of time. Error bars indicate SEM and asterisks indicate a significant difference between the conditions (p < .05).

282

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

Table 1 Self-report measures of mindfulness across session by condition. MMT Session 1 M (SD) AAQ FFMQ-Act FFMQ-Nonjudging FFMQ-Observing FFMQ-Describing FFMQ-Nonreactivity DSSQ-TS-Control and Confidence DSSQ-TS-Self-focused DSSQ-TS-Self-Esteem DSSQ-TS-Concentration + ** ****

15.97 (5.70) 25.37 (6.79) 27.63 (5.70) 26.11 (4.63) 27.74 (6.53) 23.40 (4.39) 25.60 (6.46) 16.54 (6.56) 13.29 (5.72) 16.89 (3.20)

RT Session 2 M (SD)

Session 1 M (SD)

13.66 (5.14) 26.86 (6.58) 28.91 (6.44) 26.26 (5.25) 27.71 (8.07) 22.63 (4.93) 26.86 (6.97) 13.46 (5.32) 16.49 (5.15) 17.09 (4.42)

Session 2 M (SD)

16.93 (5.48) 26.68 (5.78) 26.71 (6.67) 24.25 (4.55) 26.82 (6.24) 19.68 (4.88) 24.11 (6.48) 14.89 (5.64) 15.52 (3.43) 16.74 (4.36)

F (effect of time)

16.43 (6.15) 27.35 (4.97) 28.61 (7.26) 23.93 (5.42) 26.46 (4.80) 19.61 (4.56) 22.32 (6.74) 13.79 (5.71) 18.00 (3.29) 15.82 (6.17)

Partial

g2

F (time  condition)

Partial

g2

+

8.52

**

.14

3.51

.06

9.13

**

.15

1.27

–

.19

0.41

–

0.04

–

0.27

–

0.19

–

0.14

–

0.84

–

0.58

–

0.18

–

3.76+

.06

.13

1.70

–

.37

0.32

–

–

0.91

–

11.36

**

7.67** 22.21

****

0.44

p < .07. p < .01. p < .0001.

the first step included only AOSPAN Time 3. Cognitive mechanism words, post-stressor positive and negative affect, TUTs, and an interaction between TUTs at Time 4 and post stressor negative affect were entered in step 2. Cognitive mechanism words and negative affect predicted poorer AOSPAN performance in the RT condition but not the MMT condition. Positive affect predicted AOSPAN performance in the MMT condition. The interaction between TUTs and negative affect was a significant predictor in the MMT but not the RT condition. As seen in Fig. 2, the impact of TUTs on AOSPAN at Time 4 was plotted for a hypothetical participant in the MMT condition, 1 SD above and below the mean negative affect. AOSPAN scores declined as TUTs increased when negative affect was high but not when it was low. 4. Discussion The current study examined the impact of mindfulness meditation on WM and mind wandering. Results suggest MMT did not improve WM or decrease mind wandering compared to the RT, following the training in Session 1 or as a result of the

Table 2 Hierarchical regression models for AOSPAN performance at Time 4. B

SE

t

b

F

Adjusted R2

Mindfulness meditation training condition Step 1: AOSPAN T3 Step 2: AOSPAN T3 TUT T4 Cognitive Words PANAS-NEG PANAS-POS TUT  PANAS-NEG Interaction

0.83 0.75 6.54 0.38 0.25 0.20 2.79

0.15 0.12 4.02 0.36 0.16 0.09 0.84

.69 .62 .16 .11 .16 .22 .34

5.42**** 6.03**** 1.63 1.06 1.56 2.18* 3.31**

29.38**** 14.09****

.46 .71

Relaxation training condition Step 1: AOSPAN T3 Step 2: AOSPAN T3 TUT T4 Cognitive Words PANAS-NEG PANAS-POS TUT  PANAS-NEG Interaction

1.15 1.35 3.17 1.67 0.90 0.18 1.87

0.15 0.16 7.83 0.59 0.42 0.17 2.16

.84 .99 .04 .28 .21 .10 .09

7.79**** 8.45**** 0.40 2.85** 2.14* 1.03 .86

60.64**** 17.03****

.70 .78

* ** ****

p < .05. p < .01. p < .0001.

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

283

Fig. 2. The relationship between working memory and Task-Unrelated Thoughts as a function of negative affect in the Mindfulness Meditation Condition. Lines depict slopes for a hypothetical participant at the mean and 1 SD above and below the mean for negative affect following the stress manipulation.

1-week training. We were interested in determining if MMT prevented stress related changes in mind wandering and WM. Decreases in WM were observed following the stressor in the RT but not MMT condition. MMT did not reduce mind wandering following the stressor. The lack of improvements in WM and decreases in mind wandering as a result of the MMT are important due to the differences in control groups between the current study and prior work (Morrison et al., 2014; Mrazek et al., 2013). The active control group in the current study produced equivalent expectation of changes in WM. Ensuring equivalent expectations are critical for active control groups to ensure that any effect of training is not due to participant expectancy (Boot et al., 2013). In light of the contradictory findings, the benefits of mindfulness meditation must be examined cautiously. It is important to note the duration of the current intervention, 1 week, was shorter than prior work (Morrison et al., 2014; Mrazek et al., 2013). Therefore, a longer intervention could produce improved WM and decreased TUTs. The MMT condition demonstrated greater improvements in one self-report measure of mindfulness, the AAQ, than RT condition, suggesting the MMT did alter mindfulness. Despite the successful use in prior work (Feldman et al., 2010) it is possible the delivery of the intervention, via audio-recording, caused the differences in results. Prior training studies have relied on a trained meditation leader to guide participants through the intervention in small groups (Jha et al., 2010; Morrison et al., 2014; Mrazek et al., 2013). One benefit to this approach is that the meditation leader can adjust the delivery of the intervention based on the individuals responses and training programs make focus on different techniques during each training session, providing for a gradual building of the techniques required to engage in mindfulness meditation. While we do not argue that this may be beneficial, indeed, training programs using this approach may provide more robust effects, we were interested in the efficacy of a brief at home intervention. Following similar logic as work examining a self-administered, mild exposure therapy (Boals, Banks, & Hayslip, 2012), we wanted to provide an alternative method of training that could work for individuals who were unable or unwilling to attend a meditation training lead by an instructor. These results provide additional support for the prophylactic effects of mindfulness following stress (Jha et al., 2010), such that impairments in WM were observed in the RT but not MMT condition. Although MMT did not reduce mind wandering frequency compared to the RT, MMT altered the relationship between TUTs and task performance. TUTs were correlated with AOSPAN task performance following the stressor in the RT but not MMT condition. The stressor produced similar changes in both conditions on negative and positive affect and similar ratings on the VAS measure of perceived stress. However, working memory did not decrease as a result of these changes in the MMT condition, suggesting the intervention altered the impact of these factors. Reduced reaction to repetitive thoughts following a mindfulness intervention, demonstrated previously (Feldman et al., 2010), appears to be responsible for protective effects of mindfulness interventions. The decreases reactivity of TUTs is critical for cognitive functioning because the suppression of TUTs may be responsible for impaired cognitive performance. Suppression of TUTs consumes cognitive resources required for ongoing task performance (Klein & Boals, 2001; Wegner, 1994). The open monitoring techniques included in the MMT encourage individuals to redirect attention following TUTs rather than suppressing TUTs (Kabat-Zinn, 1994). The nonjudgmental acceptance of TUTs may be responsible for increases in executive control and emotion regulation (Teper, Segal, & Inzlicht, 2013) and may be the critical component of protective effects of mindfulness meditation. Cognitive mechanism words provide a novel way to examine this hypothesis. Cognitive word use did not differ between conditions in the writing task since the goal of the writing is to think about the event. However, higher rates of cognitive mechanism words, indicating greater engagement in meaning making (Boals et al., 2011), should require more effort to suppress during the AOSPAN, resulting in poorer AOSPAN performance. Cognitive mechanism words predicted poorer AOSPAN task performance in the RT condition, supporting this hypothesis. Further, rates of cognitive mechanism word use did not predict change in AOSPAN performance in the MMT condition, reflecting an acceptance of TUTs and lack of attempts to actively suppress the thoughts. This supports the idea that acceptance and lack of suppression of TUTs may be responsible for the protective effects of mindfulness meditation. Thus, mindfulness meditation training that employs only focused attention techniques, such as the breathing techniques used in the RT condition are unlikely to produce the same benefits observed in the current study using open monitoring techniques.

284

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

The impact of negative affect on WM replicated prior findings (Curci et al., 2013), such that negative affect predicted poorer AOSPAN performance following the stressor in the RT condition. Further, higher positive affect resulted in positive changes in AOSPAN performance in the MMT condition, consistent with prior findings (Yang, Yang, & Isen, 2013). The only predictor for poorer AOSPAN performance following the stressor in the MMT condition was the interaction between TUTs and negative affect. Although mindfulness meditation should prevent the impact of TUTs and negative affect (Feldman et al., 2010), the relationship may be present because of the short duration of the intervention. The 1-week training appears to have been sufficient to reduce the impact of TUTs at low to moderate levels but not the highest levels of negative affect, suggesting the impact of mind wandering is greatest when negative affect is high. Longer training studies have shown greater benefits for those that engage in the most practice (Jha et al., 2010). Future research should determine if the impact of the interaction between TUTs and negative affect on WM occurs with increased training time. The current study has several limitations worth noting. First, due to the at home nature of the intervention it is not possible to ensure that participants completed all of the training sessions. However, the MMT intervention did produce the expected protective results suggesting participants completed the training. Further, examining this alternative delivery of the MMT is helpful due to the ability to easily deliver it to a larger population. Second, the lack of a waitlist control group means that increases in WM across sessions must be interpreted with caution. It is not possible to determine if increases are due to practice effects or beneficial effects of the RT. Finally, the duration of the training period, 1 week, and each training session, 15 min, may have limited the impact of the intervention on working memory and mind wandering. Training studies demonstrating an effect on working memory and mind wandering have used 7–8 week durations (Jha et al., 2010; Morrison et al., 2014). These studies have also relied on longer single training session, ranging from 20 min sessions (Morrison et al., 2014), 45 min sessions (Mrazek et al., 2013) to studies that include day long retreats (Jha et al., 2010). However, improvements following mindfulness meditation interventions of as little as 8 (Mrazek, Smallwood, & Schooler, 2012), 15 (Feldman et al., 2010), and 20 min (Zeidan, Johnson, Diamond, David, & Goolkasian, 2010) have been demonstrated. Future studies should determine if a longer training period would produce increases in working memory or decreases mind wandering observed previously. Understanding the effect of mindfulness meditation on WM and mind wandering is critical given the importance of preventing stress related impairment in academic and real world settings. Current findings suggest a novel approach to delivering a brief mindfulness meditation intervention to a wide audience. Although methodological differences between the current study and prior work (Morrison et al., 2014; Mrazek et al., 2013) exist, current results suggest that further investigation of the impact of mindfulness mediation on WM is warranted. Reduced reaction to mind wandering, rather than reduced mind wandering appears to be responsible for the prophylactic benefits of mindfulness meditation. Acknowledgment We would like to thank Michael Kane for assistance with the study design. References Baer, R. A. (2009). Self-focused attention and mechanisms of change in mindfulness-based treatment. Cognitive Behaviour Therapy, 38(S1), 15–20. http:// dx.doi.org/10.1080/16506070902980703. Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self- report assessment methods to explore facets of mindfulness. Assessment, 13, 27–45. http://dx.doi.org/10.1177/1073191105283504. Banks, J. B., Tartar, J. L., & Welhaf, M. (2014). Where’s the impairment: An examination of factors that impact sustained attention following a stressor. Cognition and Emotion, 28, 856–866. http://dx.doi.org/10.1080/02699931.2013.857643. Boals, A., Banks, J. B., Hathaway, L. M., & Schuettler, D. (2011). Coping with stressful events: Use of cognitive words in stressful narratives and the meaningmaking process. Journal of Social and Clinical Psychology, 30, 378–403. http://dx.doi.org/10.1521/jscp.2011.30.4.378. Boals, A., Banks, J. B., & Hayslip, B. (2012). A self-administered, mild form of exposure therapy for older adults. Aging & Mental Health, 16, 154–161. Boals, A., & Klein, K. (2005). Word use in emotional narratives about failed romantic relationships and subsequent mental health. Journal of Language and Social Psychology, 24, 252–268. http://dx.doi.org/10.1177/0261927X05278386. Bond, F. W., Hayes, S. C., Baer, R. A., Carpenter, K. M., Guenole, N., Orcutt, H. K., et al (2011). Preliminary psychometric properties of the acceptance and action questionnaire – II: A revised measure of psychological flexibility and experiential avoidance. Behavioral Therapy, 42, 676–688. http://dx.doi.org/10.1016/ j.beth.2011.03.007. Boot, W. R., Simons, D. J., Stothart, C., & Stutts, S. (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. Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm, O., & Engle, R. W. (2005). Working memory span tasks: A methodological review and user’s guide. Psychonomic Bulletin and Review, 12(5), 769–786. http://dx.doi.org/10.3758/BF03196772. Curci, A., Lanciano, T., Soleti, E., & Rime, B. (2013). Negative emotional experiences arouse rumination and affect working memory capacity. Emotion, 13(5), 867–880. http://dx.doi.org/10.1037/a0032492. Davidson, R., & Goleman, D. J. (1977). The role of attention in meditation and hypnosis: A psychobiological perspective on transformations of consciousness. International Journal of Clinical and Experimental Hypnosis, 25, 291–308. Feldman, G., Greeson, J., & Seville, J. (2010). Differential effects of mindful breathing, progressive muscle relaxation, and loving kindness meditation on decentering and negative reactions to repetitive thoughts. Behavior Research and Therapy, 48, 1002–1011. http://dx.doi.org/10.1016/j.brat.2010.06.006. Gaab, J., Rohleder, N., Nater, U. M., & Ehlert, U. (2005). Psychological determinants of the cortisol stress response: The role of anticipatory cognitive appraisal. Psychoneuroimmunology, 30, 599–610. http://dx.doi.org/10.1016/j.psyneuen.2005.02.001. Jha, A., Krompinger, J., & Baime, M. J. (2007). Mindfulness training modifies subsystems of attention. Cognition, Affective, and Behavioral Neuroscience, 7, 109–119. http://dx.doi.org/10.3758/CABN.7.2.109. Jha, A., Stanley, E. A., Kiyonaga, A., Wong, L., & Gelfand, L. (2010). Examining the protective effects of mindfulness training on working memory capacity and affective experience. Emotion, 10, 54–64. http://dx.doi.org/10.1037/a0018438.

J.B. Banks et al. / Consciousness and Cognition 33 (2015) 277–285

285

Kabat-Zinn, J. (1994). Mindfulness-based interventions in context: Past, present, & future. Clinical Psychology: Science and Practice, 10(144), 156. http:// dx.doi.org/10.1093/clipsy.bog016. Klein, K., & Boals, A. (2001). The relationship between life event stress and working memory capacity. Applied Cognitive Psychology, 15, 565–579. http:// dx.doi.org/10.1002/acp.727. Klein, K., & Boals, A. (2010). Coherence and narrative structure in personal accounts of stressful experiences. Journal of Social and Clinical Psychology, 29, 256–280. http://dx.doi.org/10.1521/jscp.2010.29.3.256. Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008). Attention regulation and monitoring in meditation. Trends in Cognitive Science, 14, 163–169. Matthews, G., Joyner, L., Campbell, S. E., Gilliland, K., Huggins, J., & Falconer, S. (1999). Validation of a comprehensive stress state questionnaire: Towards a state big three? Personality Psychology, 7, 335–350. http://dx.doi.org/10.1177/154193120404801107. Matthews, G., Campbell, S. E., Falconer, S., Joyner, L. A., Huggins, J., Gilliand, K., et al (2002). Fundamental dimensions of subjective state in performance settings: Task engagement, distress, and worry. Emotion, 2, 315–340. McVay, J., & Kane, M. (2009). Conducting the train of thought: Working memory capacity, goal neglect, and mind wandering in an executive-control task. Journal of Experimental Psychology, 35, 196–204. http://dx.doi.org/10.1037/a0014104. Morrison, A. B., Goolsarran, M., Rogers, S. L., & Jha, A. P. (2014). Taming a wandering attention: Short-form mindfulness training in student cohorts. Frontiers in Human Neuroscience, 7, 1–12. http://dx.doi.org/10.3389/fnhum.2013.00897. Mrazek, M., Franklin, M., Phillips, D. J., Baird, B., & Schooler, J. W. (2013). Mindfulness training improves working memory capacity and GRE performance while reducing mind wandering. Psychological Science, 5, 776–781. http://dx.doi.org/10.1177/0956797612459659. Mrazek, M. D., Smallwood, J., & Schooler, J. W. (2012). Mindfulness and mind wandering: Finding convergence through opposing constructs. Emotion, 12, 442–448. http://dx.doi.org/10.1037/a0026678. Oei, N. Y. L., Everaerd, W., Elzinga, B., Van Well, S., & Bermond, B. (2006). Psychosocial stress impairs working memory at high loads: An association with cortisol levels and memory retrieval. Stress, 9, 133–141. http://dx.doi.org/10.1080/10253890600965773. Ortner, C., Kilner, S., & Zelazo, P. (2007). Mindfulness meditation and reduced emotional on a cognitive task. Motivation and Emotion, 31, 271–283. http:// dx.doi.org/10.1007/s11031-007-9076-7. Pennebaker, J. W., Booth, R. J., & Francis, M. E. (2007). Linguistic inquiry and word count: LIWC 2007. Austin, TX: LIWC. Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-based cognitive therapy for depression: A new approach to preventing relapse. NY: Guilford Press. Smallwood, J., Fitzgerald, A., Miles, L. K., & Phillips, L. H. (2009). Shifting moods, wandering minds: Negative moods lead the mind to wander. Emotion, 9(2), 271–276. http://dx.doi.org/10.1037/a0014855. Stawski, R. S., Sliwinski, M. J., Almeida, D. M., & Smyth, J. M. (2008). Reported exposure and emotional reactivity to daily stressors: The roles of adult age and global perceived stress. Psychology and Aging, 23(1), 52–61. http://dx.doi.org/10.1037/0882-7974.23.1.52. Teper, R., Segal, Z. V., & Inzlicht, M. (2013). Inside the mindful mind: How mindfulness enhances emotion regulation through improvements in executive control. Current Directions in Psychological Science, 22(6), 449–454. http://dx.doi.org/10.1177/0963721413495869. Unsworth, N., Heitz, R., Schrock, J., & Engle, R. W. (2005). An automated version of the operation span task. Behavior Research Methods, 37, 498–505. http:// dx.doi.org/10.3758/BF03192720. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. http://dx.doi.org/10.1037/022-3514.51.6.1063. Wegner, D. (1994). Ironic processes of mental control. Psychological Review, 101, 34–52. http://dx.doi.org/10.1037/0033-295X.101.1.34. Yang, H., Yang, S., & Isen, A. M. (2013). Positive affect improves working memory: Implications for controlled cognitive processing. Cognition and Emotion, 27(3), 474–482. http://dx.doi.org/10.1080/02699931.2012.713325. Zeidan, F., Johnson, S. K., Diamond, B. J., David, Z., & Goolkasian, P. (2010). Mindfulness meditation improves cognition: Evidence of brief mental training. Consciousness and Cognition, 19, 597–605. http://dx.doi.org/10.1016/j.concog.2010.03.014.