Carnegie Mellon University Research Showcase @ CMU Department of Psychology Dietrich College of Humanities and Social Sciences 1-29-2016 Alterations in Resting-State Functional Connectivity Link Mindfulness Meditation With Reduced Interleukin-6: A Randomized Controlled Trial. J. David Creswell Carnegie Mellon University, creswell@cmu.edu Adrienne A. Taren University of Pittsburgh Emily K. Lindsay Carnegie Mellon University Carol M. Greco University of Pittsburgh Peter J. Gianaros University of Pittsburgh See next page for additional authors Follow this and additional works at: http://repository.cmu.edu/psychology Part of the Psychology Commons Published In Biological psychiatry, forthcoming. This Article is brought to you for free and open access by the Dietrich College of Humanities and Social Sciences at Research Showcase @ CMU. It has been accepted for inclusion in Department of Psychology by an authorized administrator of Research Showcase @ CMU. For more information, please contact research-showcase@andrew.cmu.edu. Authors J. David Creswell, Adrienne A. Taren, Emily K. Lindsay, Carol M. Greco, Peter J. Gianaros, April Fairgrieve, Anna L. Marsland, Kirk Warren Brown, Baldwin M. Way, Rhonda K. Rosen, and Jennifer Ferris This article is available at Research Showcase @ CMU: http://repository.cmu.edu/psychology/1401                                                                                                                                                                                                                                                                       J.D.  Creswell   SHORT TITLE: MEDITATION, BRAIN, IL-6 WORD COUNT: 3,860 (249 abstract) TABLES: 2 FIGURES: 3 SUPPLEMENTAL TEXT: 1 Alterations in Resting State Functional Connectivity link Mindfulness Meditation with Reduced Interleukin-6: A Randomized Controlled Trial In press, Biological Psychiatry J. David Creswell, PhD1§, Adrienne A. Taren, MD2, Emily K. Lindsay, MA1, Carol M. Greco, PhD3, Peter J. Gianaros, PhD4, April Fairgrieve, BS1, Anna L. Marsland, PhD4, Kirk Warren Brown, PhD5, Baldwin M. Way, PhD6, Rhonda K. Rosen, LCSW4, Jennifer L. Ferris, MA1 1 Department of Psychology and Center for the Neural Basis of Cognition, Carnegie Mellon University, 5000 Forbes Ave, Pittsburgh, PA, 15213 2 Department of Neuroscience and Center for the Neural Basis of Cognition, University of Pittsburgh, A210 Langley Hall, Pittsburgh, PA, 15260 3 Department of Psychiatry, University of Pittsburgh, 580 S. Aiken Ave, Pittsburgh, PA, 15232 4 Department of Psychology and Center for the Neural Basis of Cognition, University of Pittsburgh, 210 South Bouquet St, Pittsburgh, PA, 15260 5 Department of Psychology, Virginia Commonwealth University, 806 W Franklin St, Richmond, VA, 23284 6 Department of Psychology, Ohio State University, 1827 Neil Ave, Columbus, OH, 43210 § Corresponding Author: J. David Creswell, Associate Professor in Psychology and Center for the Neural Basis of Cognition, Carnegie Mellon University, 5000 Forbes Ave, Pittsburgh, PA, 15213, creswell@cmu.edu. (phone: 412-268-9182) Keywords: mindfulness meditation; functional connectivity; IL-6; unemployment; fMRI; stress 1                                                                                                                                                                                                                                                                         J.D.  Creswell   Abstract Background: Mindfulness meditation training interventions have been shown to improve markers of health, but the underlying neurobiological mechanisms are not known. Building on initial cross-sectional research showing that mindfulness meditation may increase default mode network (DMN) resting state functional connectivity (rsFC) with regions important in top-down executive control (dorsolateral prefrontal cortex, dlPFC), here we test whether mindfulness meditation training increases DMN-dlPFC rsFC, and whether these rsFC alterations prospectively explain improvements in interleukin-6 (IL6) in a randomized controlled trial. Method: Stressed job-seeking unemployed community adults (N=35) were randomized to either a 3-day intensive residential mindfulness meditation or relaxation training program. Participants completed a five-minute resting state scan before and after the intervention program. Participants also provided blood samples at pre-intervention and at 4-month follow-up, which were assayed for circulating IL-6, a biomarker of systemic inflammation. Results: We tested for alterations in DMN rsFC using a posterior cingulate cortex (PCC) seed-based analysis, and found that mindfulness meditation training, and not relaxation training, increased PCC rsFC with left dlPFC (p<.05, corrected). These pre-post training alterations in PCC-dlPFC rsFC statistically mediated mindfulness meditation training improvements in IL-6 at 4-month follow-up. Specifically, these alterations in rsFC statistically explained 30% of the overall mindfulness meditation training effects on IL-6 at follow-up. Conclusions: These findings provide the first evidence that mindfulness meditation training functionally couples the DMN with a region known to be important in top-down executive control at rest (left dlPFC), which in turn is associated with improvements in a marker of inflammatory disease risk. Abstract word count: 249 Trial Registration: The RCT is registered on clinicaltrials.gov (#NCT01628809) 2                                                                                                                                                                                                                                                                         J.D.  Creswell   Alterations in Resting State Functional Connectivity link Mindfulness Meditation with Reduced Interleukin-6: A Randomized Controlled Trial Mindfulness meditation training programs, which train receptive attention and awareness to one’s present moment experience, have been shown to improve a broad range of stress-related psychiatric and physical health outcomes in initial randomized controlled trials (e.g., depression relapse, anxiety, HIV-progression) (1–7). For example, recent well-controlled studies indicate that mindfulness meditation training may reduce markers of inflammation (C Reactive Protein, Interleukin-6 (IL-6), neurogenic inflammation) in stressed individuals (8–11). However, little is known about the neural mechanisms underlying the effects of mindfulness training on health among these individuals (12; 13). One possibility is that mindfulness meditation training alters resting state functional connectivity (rsFC) of brain networks implicated in mind wandering (the Default Mode Network, DMN) and executive control (the Executive Control Network, EC), which in turn improves emotion regulation, stress resilience, and stress-related health outcomes in at-risk patient populations (1; 14–16). Two lines of research support this hypothesis. First, a cross-sectional study (N=25) showed that advanced mindfulness meditation practitioners had increased functional connectivity of a key hub in the default mode network (DMN) (i.e., posterior cingulate cortex) with regions considered to be important in top down executive control (EC) (dorsolateral prefrontal cortex, dorsal ACC), both at rest and during a guided mindfulness meditation practice (17). This coupling of one’s DMN at rest with regions of the EC network may be important for emotion regulation and stress resilience effects, as greater activation and functional 3                                                                                                                                                                                                                                                                         J.D.  Creswell   connectivity of EC regions, such as the dlPFC, is associated with reduced pain, negative affect, and stress (18–21). A second line of research demonstrates initial links between alterations in DMN rsFC and psychiatric (e.g., Alzheimer’s Disease, schizophrenia) (22; 23) and physical (e.g., obesity, diabetes) (24; 25) health risks; for example, there is reduced rsFC of the posterior cingulate cortex and the dlPFC observed in schizophrenia patients relative to matched controls (26). Here we provide the first experimental test of whether an intensive 3-day mindfulness meditation training intervention (relative to a relaxation training intervention) alters DMN connectivity and circulating IL-6 in a high stress unemployed job-seeking community sample. IL-6 is an established clinical health biomarker that is elevated in high stress populations (27; 28) and is associated with elevated cardiovascular disease and mortality risk (29–31). Moreover, unemployment is a well-known chronic stressor that can foster a loss of control, helplessness, and financial setbacks (32)—and unemployment is associated with elevated inflammation (33). Building on initial crosssectional evidence (17), we hypothesized that mindfulness meditation training would increase rsFC between the DMN and regions implicated in attention and executive control (dlPFC and dACC). Moreover, we tested whether mindfulness meditation training (relative to relaxation training) decreased circulating IL-6 at 4-month follow up, and whether pre-post intervention increases in DMN-dlPFC rsFC mediated IL-6 improvements at 4-month follow-up. Method and Materials Participants 4                                                                                                                                                                                                                                                                         J.D.  Creswell   Thirty-five right-handed unemployed job-seeking community adults participated (see Table S1 for inclusion/exclusion study criteria) (see Figure S1 for Consort flowchart). Participants had moderate to high levels of job-seeking stress over the past month, scoring >5 on an adapted 4-item Perceived Stress Scale (34) (M=9.60, SD=2.35; for job-seeking stress measure see Table S1). After complete description of the study to the participants, written informed consent was obtained. Procedure Beginning four weeks before the 3-day training intervention, participants completed a baseline neuroimaging session, which included a 5-minute resting state scan where they passively viewed a fixation cross. After neuroimaging, participants were invited to a nearby residential retreat center where they provided a blood sample (for measurement of circulating IL-6) and were then randomized (via a random number generator by the study PI) to either a 3-day intensive mindfulness meditation training (N=18) or a matched 3-day relaxation residential retreat intervention (N=17). Posttreatment study personnel were blind to participant study condition (including personnel running the post-treatment MRI session and MRI data preprocessing). Participants returned for a neuroimaging assessment within two weeks of completing the 3-day intervention and completed an identical scanning procedure as at baseline (participants verbally confirmed they did not engage in meditation or relaxation activities during the resting state scan at both time points). At 4-month follow-up, participants were invited back to the retreat center where they provided a blood sample and completed a measurement battery. The measures described in this report are a subset of measures collected in this trial (see Supplementary Text). 5                                                                                                                                                                                                                                                                         J.D.  Creswell   Interventions We developed a 3-day residential mindfulness meditation retreat format from activities used in the Mindfulness-Based Stress Reduction (MBSR) program (35), called Health Enhancement through Mindfulness (HEM). Delivery of the HEM program in a structured residential retreat format improves compliance with training and reduces treatment attrition; greater experimental control is also afforded by offering a parallel matched relaxation training retreat (in a separate wing of the retreat center). Briefly, the HEM program consists of mindfulness training through body scan awareness exercises, sitting and walking meditations, mindful eating, mindful stretching, and discussion. We developed a structurally matched Health Enhancement through Relaxation (HER) program that included similar behavioral training activities (e.g., walking, stretching, and didactics) as HEM, but emphasized participation in these activities in a restful rather than a mindful way. The use of a structurally-matched active comparison group was designed to control for non-mindfulness specific factors such as positive treatment expectancies, group support, teacher attention, physical activity, and mental engagement. An hour-byhour outline of interventions is provided in Table S4. Neuroimaging Measures Image Acquisition and Preprocessing. Structural and functional images were acquired on a Siemens Verio 3T scanner using a 32-channel head coil. High-resolution T1-weighted gradient-echo images were acquired (TR=1800ms, TE=2.22ms, flip angle=9º, matrix size=256x256, number of slices=256, FOV=205mm, 0.8mm slices), GRAPPA acceleration factor PE=2, voxel size=0.8x0.8x0.8mm). Next, four functional echo-planar imaging runs were acquired, including a 300 second resting state scan 6                                                                                                                                                                                                                                                                         J.D.  Creswell   (TR=2000ms, TE=30ms, flip angle=79º, matrix size=64x64, number of slices=36, FOV= 205mm, 3.2mm thick slices EPI with rate 2 GRAPPA, voxel size=3.2mmx3.2mmx3.2mm). Three participants were excluded from fMRI data analyses due to neuroimaging session problems (at the baseline appointment: 1 participant reported sleeping, 1 had poor coverage, and 1 did not understand directions; at the post-treatment appointment: 1 participant reported sleeping, 1 did not understand directions, and 1 had poor coverage). Functional BOLD data were processed using SPM8 (Welcome Department of Cognitive Neurology, London, UK; implemented by MATLAB, MathWorks, Inc., Natick, MA, USA). Functional images were first realigned to the mean image of the first run and then smoothed with a 4mm FWHM Gaussian kernel. Data were then submitted to motion correction using the Art Repair utility (36), an interpolation-based motion correction program. The functional data was then normalized to the standard Montreal Neurological Institute (MNI) T1 template. Finally, the images were smoothed with a 7mm FWHM kernel, resulting in an overall FWHM smoothing of 8mm (36). IL-6 and Psychosocial Measures IL-6. Blood samples were collected (between the hours of 10am-noon) and processed (then frozen) in batch at baseline and at 4-month follow-up by technicians blinded to treatment conditions. IL-6 levels were determined from plasma in duplicate by high sensitivity quantitative sandwich enzyme immunoassay kit (R& D Systems, Minneapolis, MN) run according to manufacturer’s directions. Two participants had insufficient samples for IL-6 determination, and were excluded. 7                                                                                                                                                                                                                                                                         J.D.  Creswell   Psychosocial Measures. In order to evaluate whether the relaxation retreat program produced equivalent positive beliefs about its value (a placebo control) compared to the mindfulness retreat program, participants completed a 6-item measure of perceived positive treatment benefits at the conclusion of the 3-day retreat program using an adapted version of the Credibility/Expectancy Questionnaire (37) (study α=.87, sample item: “At this point, how much do you really feel that this therapy will help you reduce your stress symptoms?” 1=not at all to 9=very much). Re-employment status was assessed at 4-month follow-up; participants indicated whether they were unemployed, defined as having no job for more than 20 hours per week in the 4-month post-treatment period. Data Analysis Resting State Functional Connectivity Analysis. PCC-seeded resting state BOLD fMRI images were generated in the CONN toolbox, following the recommended CONN analysis procedures (35). Specifically, CONN implements several additional processing and 1st level analysis routines prior to rsFC analysis: CONN estimates an orthogonal time series using principal component analysis of the BOLD time series in each noise ROI (subject-specific white matter and csf masks). Structural MPRAGE images were segmented to define gray matter, white matter, and cerebrospinal fluid areas. BOLD signal from the subject-specific white matter and CSF masks, motion parameters (six dimensions), and the effect of rest (an average across the session) were used as regressors to account for further temporal confounding factors. CONN uses a component-based noise reduction (CompCor) that avoids regression of the global signal. A covariate for each subject’s head motion was entered at the first level. A band-pass 8                                                                                                                                                                                                                                                                         J.D.  Creswell   filter of 0.008-0.09 Hz was used. A hemodynamic response function was used to downweight the initial scans within each resting state block to minimize potential ramping effects. The PCC seed was anatomically defined using the Talaraich deamon database in the Wake Forest University (WFU) Pickatlas (38) centered on MNI: -4,-50,40. Seeded first-level maps in CONN were then submitted to a second-level full factorial analysis in SPM8 with two factors specified, time and group. To test study predictions in the brain, we specified a time-by-group spreading interaction contrast that tested for baseline to post-intervention increases in rsFC in the HEM program relative to no change in the HER program from baseline to post-intervention using contrast weights: [-1(pre,HEM), -1(pre,HER), 3(post, HEM), -1(post,HER)]. This t-contrast models the specific hypothesized differential group change from baseline to post-treatment. The strength of this approach (relative to testing for significant voxels using the more standard overall F-contrast, or just comparing the two groups at post-treatment only) is that it tests the specific prediction the mindfulness meditation program increased rsFC from baseline to post-treatment compared to no change in the relaxation group (as opposed to other types of interaction patterns that might be significant with an F-contrast analysis). Note that this approach compares the mindfulness group at post-treatment to the average of the other cells in this 2X2 design, testing the spreading interaction prediction (and not other interaction patterns, e.g., crossover interactions). Furthermore, we then plotted the parameter estimates from this spreading interaction contrast to visually confirm the specific interaction pattern (see Figure 1b and 1c). 9                                                                                                                                                                                                                                                                         J.D.  Creswell   For purposes of testing rsFC with the PCC in this study, two ROI masks were created using the WFU Pickatlas AAL atlas (each with dilation of 1mm): the first mask consisted of the left and right anterior cingulate, and the second ROI mask consisted of the left and right middle frontal cortex (based on 17). Cluster-level correction for multiple comparisons was obtained using a Monte Carlo simulations in AlphaSim (National Institute of Mental Health, Bethesda, MD). AlphaSim was first run on the anterior cingulate AAL-defined mask, with significant clusters (p<.05, corrected) defined as those involving k>49 contiguous voxels, each at p<.005. AlphaSim was then run on the middle frontal cortex AAL-defined ROI mask, with significant clusters (p<.05, corrected) defined as those involving k>82 contiguous voxels, each at p<.005. IL-6 and Head Motion Analysis. IL-6 values at baseline and 4-month follow-up were log transformed. Analyses adhered to intent-to-treat principles using mixed effect linear models (MLMs) conducted in SPSS 21.0 (IBM, Armonk, New York). All variables were modeled as fixed effects in models fit with a compound symmetric variancecovariance structure, using maximum likelihood estimation. These models included a condition factor (HEM vs. HER program), a time factor (baseline, 4-month follow-up), and their interaction, with F-statistics used to evaluate significant effects. All IL-6 MLMs included participant age as a covariate given the significant range in participant age in this sample (22-54 years old), as age is robustly associated with increasing IL-6 (27; 39). To test whether there were condition (and time×condition) differences in fast head motion (defined as TR-TR head motion greater than .25mm or .25 degrees in any plane), a hypothesis and condition-blind coder summed total high head motion TRs for each 10                                                                                                                                                                                                                                                                         J.D.  Creswell   participant’s baseline and post-treatment resting state scan, which was then tested in a MLM with total head motion as the outcome. PCC Resting State Functional Connectivity Mediation Analysis. Statistical mediation analyses were conducted following recommended procedures for testing intervening variable effects, using MLMs (40; 41). Specifically, rsFC cluster-level parameter estimates were extracted from the SPM8 group-level analyses and tested as mediating variables in a series of MLMs. The MLMs consisted of (MLM #1) testing a time×condition interaction effect on change in PCC rsFC (path a, see Figure 2); (MLM #2) testing for a significant effect of PCC rsFC on change in IL-6 (path b), when the PCC rsFC variable (baseline, post-training) was entered simultaneously along with the condition, time, and time×condition interaction variables; and (MLM #3) whether the original time×condition interactive effect on IL-6 (path c) was no longer statistically significant when the PCC rsFC effect was entered as a simultaneous predictor variable (path c’). Statistical mediation was defined as present when: 1) there were significant intervening variable paths (paths a and b), and 2) when the original time×condition interaction on IL-6 (path c’) was no longer significant when the change in rsFC variable was entered as a simultaneous predictor variable. Results Preliminary Analyses—Success of randomization and treatment program measures. There were no significant differences between the mindfulness (HEM) and relaxation (HER) groups on study baseline characteristics, indicating successful randomization (see Table 1). There were no significant group differences in fast head motion during the resting state scan period (F(1,35)= .002, p=.97) and no differential fast 11                                                                                                                                                                                                                                                                         J.D.  Creswell   head motion changes from baseline to the post-treatment fMRI resting state scan between groups (time×condition F(1,34)= .001, p=.98). The relaxation training program was an effective placebo control; there were no significant group differences in perceived treatment benefits at the conclusion of the 3-day retreat (independent samples t(31)=1.06, p=.30; HEM M=38.76, SE=2.19, HER M=42.03, SE=2.13). 97% of randomized participants completed the 3-day training programs and 97% were retained at the 4-month follow-up assessment (see Fig. S1 for CONSORT flowchart). These high retention rates may have reflected the high (and equivalent) treatment satisfaction reported by participants in both the mindfulness (HEM M=4.15, SE=.18) and relaxation (HER M=4.12, SE=.19) programs at follow-up (single item: “Would you recommend this program to other people you know?”: 1=not at all to 5=a great deal; independent samples t(32)=.11, p=.91). Despite our efforts to encourage home practice after the 3-day retreats (along with sending participants home with customized compact discs containing guided condition-specific mindfulness and relaxation exercises), participants did not complete much formal practice in the 4-month follow-up period. HEM participants reported using their home practice CD an average of 1.24 times per week (SD=1.28) over the last month (at the 4-month follow-up assessment), while HER participants reported using their home practice CD .38 times per week (SD=.86) over the last month (t(31)= 2.27, p=.03). (We also did not collect measures of home practice in the day(s) following the retreat program leading up to the post-treatment fMRI session, a study limitation.) Primary Analyses—Alterations in DMN resting state functional connectivity. Consistent with predictions, there was a significant pre-post intervention increase in rsFC 12                                                                                                                                                                                                                                                                         J.D.  Creswell   between the PCC (the DMN seed region) and left dlPFC (MNI: -22,52,10; k=111; Z=3.44, p<.05, corrected) in the mindfulness meditation training group relative to the relaxation training control group. As shown in Figure 1, there was no coupling of PCC with the left dlPFC (panel B) at baseline in both the mindfulness and relaxation groups, but at post-treatment the mindfulness (but not relaxation) group showed positive coupling of PCC and left dlPFC. Although the clusters did not survive multiple comparison correction, a homologue region of right dlPFC showed a similar effect (MNI: 26,44,34; k=24; Z= 3.22; see Figure 1 panel C), along with another cluster in left dlPFC (MNI: 30,42,38; k=77; Z=3.11). Contrary to predictions, mindfulness meditation training did not significantly alter rsFC of the PCC with dACC. We also observed no alterations in intraDMN rsFC; specifically, mindfulness meditation training did not decouple the PCC with other DMN nodes (e.g., PCC with ventromedial PFC) (42). We provide exploratory time×condition interaction results for PCC rsFC across the whole brain in Tables S2 and S3 (thresholded at uncorrected p<.005, k>50 voxels). Circulating IL-6. There were no significant baseline differences in log transformed IL-6 (or raw IL-6; see Table 1) between the mindfulness meditation and relaxation training groups (t(28)=1.04, p=.31). However, mindfulness meditation training, relative to relaxation training, reduced circulating levels of IL-6 at 4-month follow-up. A mixed effect linear model (controlling for participant age) indicated that mindfulness meditation training significantly reduced circulating log-transformed IL-6 from baseline (M=.13, SE=.07) to 4-month follow-up (M=.08, SE=.08) compared to increases in the relaxation training group (baseline M=-.06, SE=.08; 4-month follow-up M=.08, SE=.07) (time×condition interaction F(1,29)= 4.14, p=.05, d=.71) (see Table 2). 13                                                                                                                                                                                                                                                                         J.D.  Creswell   Changes in re-employment during the 4-month follow-up period could have explained these changes in IL-6, but this was not the case; there were equal rates of re-employment in the two groups at the 4-month follow-up assessment (48% of participants in both groups were re-employed at 4-month follow-up). Alterations in DMN rsFC mediate mindfulness meditation training improvements in IL-6. As shown in Figure 3, increases in pre-post intervention PCC-left dlPFC rsFC statistically mediated changes in circulating IL-6 from baseline to 4-month follow-up. Specifically, a mediation model indicated that the strength of the time×condition interaction effect on IL-6 (path c) was no longer significant when the change in the left dlPFC cluster (MNI:-22,52,10; k=111) connectivity predictor variable was entered simultaneously in a mixed effect linear model (see path c’ in Figure 2). Although the right dlPFC cluster did not survive multiple comparison correction (MNI: 26,44,34; k=24), it also marginally significantly mediated the IL-6 effects (also shown in Figure 3). Mindfulness meditation training alterations in rsFC observed in the second left dlPFC cluster (MNI:-30,42,38) did not mediate IL-6 effects. These findings in Figure 2 indicate that change in PCC-dlPFC connectivity accounts for 30% (left dlPFC MNI:-22,52,10) and 25% (right dlPFC MNI:26,44,34) of the overall mindfulness meditation (vs relaxation training) effect on reductions in IL-6 at 4-month follow-up. Conclusions There has been considerable recent interest in characterizing resting state neural networks such as the DMN, and understanding their role in health and disease (23). We report that mindfulness meditation training (compared to relaxation training without a mindfulness component) increased rsFC between the DMN (PCC) and left dlPFC. These 14                                                                                                                                                                                                                                                                         J.D.  Creswell   findings corroborate and experimentally extend cross-sectional findings in advanced mindfulness meditation practitioners (17), suggesting that mindfulness meditation training may couple one’s resting state DMN with regions implicated in executive control (dlPFC). These findings were specific to coupling of the PCC with dlPFC, and not with dACC (cf. 17). Notably, some work has shown that dlPFC and dACC are involved with dissociable executive control and salience processing networks, respectively (43); furthermore, whereas increased dACC connectivity is associated with anxiety, increased dlPFC connectivity is associated with behavioral improvements in executive task performance (e.g., attention task switching) (43). Although studies commonly report that the DMN is anti-correlated with regions in the executive control network (44; 45), some studies indicate positive PCC-dlPFC coupling during self-focused and process-oriented mental simulations (45–47) and during guided mindfulness meditation practice (17). These findings suggest that a processoriented focus on the self during meditation may shift intra-network DMN rsFC toward inter-network connectivity, coupling DMN with regions in the executive control network. Notably, the present results show that these alterations in DMN-EC rsFC can be maintained in the weeks following brief, intensive mindfulness meditation training. We also provide an initial indication in this RCT sample (N=35) that mindfulness meditation training reduces IL-6 from baseline to 4-month follow-up, compared to increases in IL-6 in the relaxation training control group. This finding suggests that mindfulness meditation training may decrease biomarkers of inflammatory disease risk in populations whom have elevated inflammation (such as stressed unemployed adults), although we note this was a small sample and this IL-6 finding should be treated with 15                                                                                                                                                                                                                                                                         J.D.  Creswell   some caution. But like this study finding, several studies show that mindfulness meditation training interventions reduce markers of inflammation in stressed populations (8–11). Moreover, the present work shows that pre-post training changes in PCC-dlPFC rsFC mediate these effects on IL-6. Currently very little is known about top-down neural modulation of peripheral inflammation in humans, despite a large number of epidemiologic studies linking peripheral circulating pro-inflammatory cytokines with cardiovascular morbidity and all-cause mortality (29–31; 48). Some initial studies in humans highlight reciprocal brain-peripheral inflammation links (49; 50), although the present study offers a novel top-down regulatory pathway for the modulation of circulating IL-6 in humans. Notably, studies have implicated activation and functional connectivity of dlPFC with improved executive control resources, and reduced pain, negative affect and stress (18–21; 43). We speculate that mindfulness meditation training coupled the brain’s DMN with regulatory areas of prefrontal cortex (dlPFC), which facilitated more effective emotion regulation and stress resilience in this high stress unemployed job-seeking sample, reducing circulating IL-6. Specifically, efferent projections from dlPFC might modulate medial PFC and subcortical cell groups known to trigger peripheral stress and inflammatory response cascades (51; 52). A small experimental literature provides initial support for this neural stress-buffering account of mindfulness meditation training (14), although more research is needed to evaluate these stress reduction mediated pathways. An important question going forward will be to evaluate whether these observed PCC-dlPFC rsFC changes reflect neuroplasticity in white matter connectivity after mindfulness meditation training. Several lines of evidence provide suggestive support: 16                                                                                                                                                                                                                                                                         J.D.  Creswell   anatomic studies in primate models have shown that measures of rsFC have high correspondence to underlying white matter architecture (53), and there are established white matter tracts linking PCC with dlPFC (54). Notably, an initial experimental study demonstrates that even brief meditation training (~11 hours) fosters neuroplasticity by increasing white matter connectivity (as measured by fractional anisotropy) of the anterior corona radiata (55). Unlike a previous study (17), mindfulness meditation training did not increase PCC-dACC rsFC in this sample. We speculate that one possibility for these divergent effects may be due to differences in study methodology. The previous study measured rsFC using 2-minute resting state periods at the beginning of each run, followed by 4.5 minute meditation practice periods (17). By contrast, the present study collected rsFC at the beginning of the fMRI session (immediately following the collection of the structural scans). Some neuroimaging evidence indicating that acute meditation practice activates the ACC (56; 57), and it may be that ACC activity during these 4.5 minute meditation practice periods had some residual carry-over into the rsFC scans in the previous study. Indeed, there is research showing that cognitive tasks have carry-over effects on PCCACC functional connectivity during subsequent rsFC periods (58). The present study design is less susceptible to potential task-based carry over effects, and provides an initial indication for non-task based PCC-dlPFC rsFC effects in the weeks following a 3-day mindfulness meditation training intervention. Conclusions Little is known about how behavioral interventions, such as mindfulness meditation training, can impact the brain and physical health. We provide the first well- 17                                                                                                                                                                                                                                                                         J.D.  Creswell   controlled evidence that 3-days of mindfulness meditation training increases rsFC of the DMN with neural regions important in executive control (left dlPFC), and that these rsFC changes statistically mediate improvements in circulating levels of IL-6 at follow-up. 18                                                                                                                                                                                                                                                                         J.D.  Creswell   Acknowledgements and Financial Disclosures The authors declare no conflicts of interest. J.D. Creswell and A.A. Taren had full access to the data and take responsibility for the integrity of the data and the accuracy of the data analysis. We would like to thank the research assistants in the Health and Human Performance laboratory, the blood draw nurses (Karen Foley, Diana Ross) and lab technician (Katarina Krajina), the Sisters at Kearns Spirituality Center for hosting the retreats, and the Scientific Imaging and Brain Research (SIBR) center. We also thank Erica Julson, Laura Pacilio, and Shinzen Young for help and feedback. The authors report no financial interests or potential conflicts of interest regarding this research. This research was supported by funding from the Pittsburgh Life Sciences Greenhouse Opportunity Fund, who played no role in the design and conduct of the study; collection management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication.   19                                                                                                                                                                                                                                                                         J.D.  Creswell   References     1.  Ludwig  DS,  Kabat-­‐Zinn  J  (2008):  Mindfulness  in  medicine.  J  Am  Med  Assoc.  300:   1350–1352.   2.  Brown  KW,  Ryan  RM,  Creswell  JD  (2007):  Mindfulness:  Theoretical  foundations   and  evidence  for  its  salutary  effects.  Psychol  Inq.  18:  211–237.   3.  Goyal  M,  Singh  S,  Sibinga  EM,  Gould  NF,  Rowland-­‐Seymour  A,  Sharma  R,  et  al.   (2014):  Meditation  programs  for  psychological  stress  and  well-­‐being:  a   systematic  review  and  meta-­‐analysis.  JAMA  Intern  Med.  174:  357–368.   4.  Segal  ZV,  Bieling  P,  Young  T,  MacQueen  G,  Cooke  R,  Martin  L,  et  al.  (2010):   Antidepressant  monotherapy  vs  sequential  pharmacotherapy  and   mindfulness-­‐based  cognitive  therapy,  or  placebo,  for  relapse  prophylaxis  in   recurrent  depression.  Arch  Gen  Psychiatry.  67:  1256.   5.  Teasdale  JD,  Segal  ZV,  Mark  J,  Ridgeway  VA,  Soulsby  JM,  Lau  MA  (2000):   Prevention  of  relapse/recurrence  in  major  depression  by  mindfulness-­‐based   cognitive  therapy.  J  Consult  Clin  Psychol.  68:  615–623.   6.  Creswell  JD,  Myers  HF,  Cole  SW,  Irwin  MR  (2009):  Mindfulness  meditation   training  effects  on  CD4+  T  lymphocytes  in  HIV-­‐1  infected  adults:  A  small   randomized  controlled  trial.  Brain  Behav  Immun.  23:  184–188.   7.  Goldin  PR,  Gross  JJ  (2010):  Effects  of  mindfulness-­‐based  stress  reduction  (MBSR)   on  emotion  regulation  in  social  anxiety  disorder.  Emotion.  10:  83–91.   8.  Creswell  JD,  Irwin  MR,  Burklund  LJ,  Lieberman  MD,  Arevalo  JMG,  Ma  J,  et  al.   (2012):  Mindfulness-­‐Based  Stress  Reduction  training  reduces  loneliness  and   20                                                                                                                                                                                                                                                                         J.D.  Creswell   21   pro-­‐inflammatory  gene  expression  in  older  adults:  A  small  randomized   controlled  trial.  Brain  Behav  Immun.  26:  1095–1101.   9.  Malarkey  WB,  Jarjoura  D,  Klatt  M  (2013):  Workplace  based  mindfulness  practice   and  inflammation:  A  randomized  trial.  Brain  Behav  Immun.  27:  145–154.   10.  Rosenkranz  MA,  Davidson  RJ,  MacCoon  DG,  Sheridan  JF,  Kalin  NH,  Lutz  A  (2013):   A  comparison  of  mindfulness-­‐based  stress  reduction  and  an  active  control  in   modulation  of  neurogenic  inflammation.  Brain  Behav  Immun.  27C:  174–184.   11.  Lengacher  CA,  Kip  KE,  Barta  MK,  Post-­‐White  J,  Jacobsen  P,  Groer  M,  et  al.  (2012):   A  Pilot  Study  Evaluating  the  Effect  of  Mindfulness-­‐Based  Stress  Reduction  on   Psychological  Status,  Physical  Status,  Salivary  Cortisol,  and  Interleukin-­‐6   Among  Advanced-­‐Stage  Cancer  Patients  and  Their  Caregivers.  J  Holist  Nurs.   30:  170–185.   12.  Hölzel  BK,  Lazar  SW,  Gard  T,  Schuman-­‐Olivier  Z,  Vago  DR,  Ott  U  (2011):  How   does  mindfulness  meditation  work?  Proposing  mechanisms  of  action  from  a   conceptual  and  neural  perspective.  Perspect  Psychol  Sci.  6:  537–559.   13.  Tang  Y-­‐Y,  Hölzel  BK,  Posner  MI  (2015):  The  neuroscience  of  mindfulness   meditation.  Nat  Rev  Neurosci.  16:  213–225.   14.  Creswell  JD,  Lindsay  EK  (2014):  How  does  mindfulness  training  affect  health?  A   mindfulness  stress  buffering  account.  Curr  Dir  Psychol  Sci.  23:  401–407.   15.  Mrazek  MD,  Smallwood  J,  Schooler  JW  (2012):  Mindfulness  and  mind-­‐ wandering:  Finding  convergence  through  opposing  constructs.  Emotion.  12:   442–448.                                                                                                                                                                                                                                                                         J.D.  Creswell   22   16.  Christoff  K,  Gordon  AM,  Smallwood  J,  Smith  R,  Schooler  JW  (2009):  Experience   sampling  during  fMRI  reveals  default  network  and  executive  system   contributions  to  mind  wandering.  Proc  Natl  Acad  Sci.  106:  8719–8724.   17.  Brewer  JA,  Worhunsky  PD,  Gray  JR,  Tang  Y-­‐Y,  Weber  J,  Kober  H  (2011):   Meditation  experience  is  associated  with  differences  in  default  mode   network  activity  and  connectivity.  Proc  Natl  Acad  Sci.  108:  20254–20259.   18.  Wager  TD,  Rilling  JK,  Smith  EE,  Sokolik  A,  Casey  KL,  Davidson  RJ,  et  al.  (2004):   Placebo-­‐Induced  Changes  in  fMRI  in  the  Anticipation  and  Experience  of  Pain.   Science.  303:  1162–1167.   19.  Lorenz  J,  Minoshima  S,  Casey  KL  (2003):  Keeping  pain  out  of  mind:  the  role  of   the  dorsolateral  prefrontal  cortex  in  pain  modulation.  Brain.  126:  1079– 1091.   20.  Goldin  PR,  McRae  K,  Ramel  W,  Gross  JJ  (2008):  The  Neural  Bases  of  Emotion   Regulation:  Reappraisal  and  Suppression  of  Negative  Emotion.  Biol   Psychiatry,  Stress,  Anxiety,  and  Post-­‐Traumatic  Stress  Disorder.  63:  577–586.   21.  Cisler  JM,  James  GA,  Tripathi  S,  Mletzko  T,  Heim  C,  Hu  XP,  et  al.  (2013):   Differential  functional  connectivity  within  an  emotion  regulation  neural   network  among  individuals  resilient  and  susceptible  to  the  depressogenic   effects  of  early  life  stress.  Psychol  Med.  43:  507–518.   22.  Greicius  M  (2008):  Resting-­‐state  functional  connectivity  in  neuropsychiatric   disorders:  Curr  Opin  Neurol.  24:  424–430.                                                                                                                                                                                                                                                                         J.D.  Creswell   23   23.  Anticevic  A,  Cole  MW,  Murray  JD,  Corlett  PR,  Wang  X-­‐J,  Krystal  JH  (2012):  The   role  of  default  network  deactivation  in  cognition  and  disease.  Trends  Cogn   Sci.  16:  584–592.   24.  Tregellas  JR,  Wylie  KP,  Rojas  DC,  Tanabe  J,  Martin  J,  Kronberg  E,  et  al.  (2011):   Altered  Default  Network  Activity  in  Obesity.  Obesity.  19:  2316–2321.   25.  Musen  G,  Jacobson  AM,  Bolo  NR,  Simonson  DC,  Shenton  ME,  McCartney  RL,  et  al.   (2012):  Resting-­‐State  Brain  Functional  Connectivity  Is  Altered  in  Type  2   Diabetes.  Diabetes.  61:  2375–2379.   26.  Zhou  Y,  Liang  M,  Jiang  T,  Tian  L,  Liu  Y,  Liu  Z,  et  al.  (2007):  Functional   dysconnectivity  of  the  dorsolateral  prefrontal  cortex  in  first-­‐episode   schizophrenia  using  resting-­‐state  fMRI.  Neurosci  Lett.  417:  297–302.   27.  Kiecolt-­‐Glaser  JK,  Preacher  KJ,  MacCallum  RC,  Atkinson  C,  Malarkey  WB,  Glaser  R   (2003):  Chronic  stress  and  age-­‐related  increases  in  the  proinflammatory   cytokine  IL-­‐6.  Proc  Natl  Acad  Sci.  100:  9090–9095.   28.  Steptoe  A,  Hamer  M,  Chida  Y  (2007):  The  effects  of  acute  psychological  stress  on   circulating  inflammatory  factors  in  humans:  A  review  and  meta-­‐analysis.   Brain  Behav  Immun.  21:  901–912.   29.  Harris  TB,  Ferrucci  L,  Tracy  RP,  Corti  MC,  Wacholder  S,  Ettinger  Jr  WH,  et  al.   (1999):  Associations  of  elevated  Interleukin-­‐6  and  C-­‐Reactive  protein  levels   with  mortality  in  the  elderly.  Am  J  Med.  106:  506–512.   30.  Pal  M,  Febbraio  MA,  Whitham  M  (2014):  From  cytokine  to  myokine:  the   emerging  role  of  interleukin-­‐6  in  metabolic  regulation.  Immunol  Cell  Biol.  92:   331–339.                                                                                                                                                                                                                                                                         J.D.  Creswell   24   31.  Danesh  J,  Kaptoge  S,  Mann  AG,  Sarwar  N,  Wood  A,  Angleman  SB,  et  al.  (2008):   Long-­‐Term  Interleukin-­‐6  Levels  and  Subsequent  Risk  of  Coronary  Heart   Disease:  Two  New  Prospective  Studies  and  a  Systematic  Review.  PLoS  Med.  5:   e78.   32.  Baum  A,  Fleming  R,  Reddy  DM  (1986):  Unemployment  stress:  Loss  of  control,   reactance  and  learned  helplessness.  Soc  Sci  Med.  22:  509–516.   33.  Hintikka  J,  Lehto  S,  Niskanen  L,  Huotari  A,  Herzig  K-­‐H,  Koivumaa-­‐Honkanen  H,  et   al.  (2009):  Unemployment  and  ill  health:  a  connection  through   inflammation?  BMC  Public  Health.  9:  410.   34.  Warttig  SL,  Forshaw  MJ,  South  J,  White  AK  (2013):  New,  normative,  English-­‐ sample  data  for  the  Short  Form  Perceived  Stress  Scale  (PSS-­‐4).  J  Health   Psychol.  18:  1617–1628.   35.  Kabat-­‐Zinn  J  (1990):  Full  catastrophe  living:  Using  the  wisdom  of  your  body  and   mind  to  face  stress,  pain,  and  illness.  New  York,  NY:  Delta.   36.  P.  Mazaika,  Whitfield-­‐Gabrieli  S,  Reiss  A  (2007):  Artifact  repair  for  fMRI  data   from  motion  clinical  subjects.  Presented  at  the  Organization  for  Human  Brain   Mapping  Annual  Conference.   37.  Devilly  GJ,  Borkovec  TD  (2000):  Psychometric  properties  of  the   credibility/expectancy  questionnaire.  J  Behav  Ther  Exp  Psychiatry.  31:  73–86.   38.  Maldjian  JA,  Laurienti  PJ,  Kraft  RA,  Burdette  JH  (2003):  An  automated  method   for  neuroanatomic  and  cytoarchitectonic  atlas-­‐based  interrogation  of  fMRI   data  sets.  Neuroimage.  19:  1233–1239.                                                                                                                                                                                                                                                                         J.D.  Creswell   25   39.  Wei  J,  Xu  H,  Davies  JL,  Hemmings  GP  (1992):  Increase  of  plasma  IL-­‐6   concentration  with  age  in  healthy  subjects.  Life  Sci.  51:  1953–1956.   40.  MacKinnon  DP,  Lockwood  CM,  Hoffman  JM,  West  SG,  Sheets  V  (2002):  A   comparison  of  methods  to  test  mediation  and  other  intervening  variable   effects.  Psychol  Meth.  7:  83–104.   41.  Baron  RM,  Kenny  DA  (1986):  The  moderator–mediator  variable  distinction  in   social  psychological  research:  Conceptual,  strategic,  and  statistical   considerations.  J  Pers  Soc  Psychol.  51:  1173–1182.   42.  Buckner  RL,  Andrews-­‐Hanna  JR,  Schacter  DL  (2008):  The  Brain’s  Default   Network.  Ann  N  Y  Acad  Sci.  1124:  1–38.   43.  Seeley  WW,  Menon  V,  Schatzberg  AF,  Keller  J,  Glover  GH,  Kenna  H,  et  al.  (2007):   Dissociable  Intrinsic  Connectivity  Networks  for  Salience  Processing  and   Executive  Control.  J  Neurosci.  27:  2349–2356.   44.  Fox  MD,  Snyder  AZ,  Vincent  JL,  Corbetta  M,  Essen  DCV,  Raichle  ME  (2005):  The   human  brain  is  intrinsically  organized  into  dynamic,  anticorrelated   functional  networks.  Proc  Natl  Acad  Sci  U  S  A.  102:  9673–9678.   45.  Spreng  RN  (2012):  The  Fallacy  of  a  “Task-­‐Negative”  Network.  Front  Psychol.  3:   145.   46.  Gerlach  KD,  Spreng  RN,  Madore  KP,  Schacter  DL  (2014):  Future  planning:  default   network  activity  couples  with  frontoparietal  control  network  and  reward-­‐ processing  regions  during  process  and  outcome  simulations.  Soc  Cogn  Affect   Neurosci.  nsu001.                                                                                                                                                                                                                                                                         J.D.  Creswell   26   47.  Spreng  RN,  Stevens  WD,  Chamberlain  JP,  Gilmore  AW,  Schacter  DL  (2010):   Default  network  activity,  coupled  with  the  frontoparietal  control  network,   supports  goal-­‐directed  cognition.  NeuroImage.  53:  303–317.   48.  Cesari  M,  Penninx  BWJH,  Newman  AB,  Kritchevsky  SB,  Nicklas  BJ,  Sutton-­‐Tyrrell   K,  et  al.  (2003):  Inflammatory  markers  and  onset  of  cardiovascular  events.   Circulation.  108:  2317–2322.   49.  Harrison  NA,  Brydon  L,  Walker  C,  Gray  MA,  Steptoe  A,  Critchley  HD  (2009):   Inflammation  Causes  Mood  Changes  Through  Alterations  in  Subgenual   Cingulate  Activity  and  Mesolimbic  Connectivity.  Biol  Psychiatry.  66:  407–414.   50.  Gianaros  PJ,  Marsland  AL,  Kuan  DC-­‐H,  Schirda  BL,  Jennings  JR,  Sheu  LK,  et  al.   (2014):  An  Inflammatory  Pathway  Links  Atherosclerotic  Cardiovascular   Disease  Risk  to  Neural  Activity  Evoked  by  the  Cognitive  Regulation  of   Emotion.  Biol  Psychiatry,  Mechanisms  of  Aging  and  Cognition.  75:  738–745.   51.  Irwin  MR,  Cole  SW  (2011):  Reciprocal  regulation  of  the  neural  and  innate   immune  systems.  Nat  Rev  Immunol.  11:  625–632.   52.  Jankord  R,  Zhang  R,  Flak  JN,  Solomon  MB,  Albertz  J,  Herman  JP  (2010):  Stress   activation  of  IL-­‐6  neurons  in  the  hypothalamus.  Am  J  Physiol  -­‐  Regul  Integr   Comp  Physiol.  299:  R343–R351.   53.  Wang  Z,  Chen  LM,  Négyessy  L,  Friedman  RM,  Mishra  A,  Gore  JC,  Roe  AW  (2013):   The  Relationship  of  Anatomical  and  Functional  Connectivity  to  Resting-­‐State   Connectivity  in  Primate  Somatosensory  Cortex.  Neuron.  78:  1116–1126.                                                                                                                                                                                                                                                                         J.D.  Creswell   54.  Vogt  BA,  Rosene  DL,  Pandya  DN  (1979):  Thalamic  and  Cortical  Afferents   Differentiate  Anterior  from  Posterior  Cingulate  Cortex  in  the  Monkey.   Science,  New  Series.  204:  205–207.   55.  Tang  Y-­‐Y,  Lu  Q,  Geng  X,  Stein  EA,  Yang  Y,  Posner  MI  (2010):  Short-­‐term   meditation  induces  white  matter  changes  in  the  anterior  cingulate.  Proc  Natl   Acad  Sci.  107:  15649–15652.   56.  Zeidan  F,  Martucci  KT,  Kraft  RA,  Gordon  NS,  McHaffie  JG,  Coghill  RC  (2011):   Brain  mechanisms  supporting  the  modulation  of  pain  by  mindfulness   meditation.  J  Neurosci.  31:  5540–5548.   57.  Hölzel  BK,  Ott  U,  Hempel  H,  Hackl  A,  Wolf  K,  Stark  R,  Vaitl  D  (2007):  Differential   engagement  of  anterior  cingulate  and  adjacent  medial  frontal  cortex  in  adept   meditators  and  non-­‐meditators.  Neurosci  Lett.  421:  16–21.   58.  Grigg  O,  Grady  CL  (2010):  Task-­‐Related  Effects  on  the  Temporal  and  Spatial   Dynamics  of  Resting-­‐State  Functional  Connectivity  in  the  Default  Network.   PLoS  ONE.  5:  e13311.         27                                                                                                                                                                                                                                                                         J.D.  Creswell   28   Figure  Legend   Figure  1.  Baseline  to  Post-­‐Treatment  Resting  State  Functional  Connectivity  in  the   Mindfulness  and  Relaxation  Training  Groups.  (A)  Left  dlPFC  cluster  (MNI:  -­‐22,52,10;   k=111,  p<.05  corrected).  that  showed  increased  rsFC  with  PCC  from  before  to  after   mindfulness  meditation  training  (HEM)  relative  to  relaxation  training  (HER).   Specifically,  a  time×condition  spreading  interaction  analysis  revealed  a  significant   cluster  in  left  dlPFC    (panels  A  and  B).  A  cluster  in  right  dlPFC  (MNI:  26,42,38;  k=24)   showed  the  same  pattern  of  effects  as  the  left  dlPFC  cluster  but  it  did  not  survive   correction,  thus  the  (panel  C)  right  dlPFC  rsFC  results  should  be  interpreted  with   caution.  .The  spreading  interaction  effects  that  mediate  IL-­‐6  effects  are  depicted  for   left  (B)  and  right  (C)  dlPFC.  Specifically,  mean  connectivity  strength  is  shown  for  the   mindfulness  (HEM)  and  relaxation  (HER)  training  groups  at  baseline  and  post-­‐ treatment.  Error  bars  depict  +/-­‐  1  standard  error.       Figure  2.  IL-­‐6  Mediation  analyses.  Increases  in  left  dlPFC  connectivity  (MNI:  -­‐ 22,52,10)  significantly  mediate  (panel  A)  the  time  × treatment  interaction  on   circulating  (log  transformed)  IL-­‐6.  Increases  in  right  dlPFC  (MNI:  26,42,38)   marginally  significantly  mediate  (panel  B)  IL-­‐6  effects.  Numbers  represent  b   coefficients  from  mixed  effect  linear  models,  with  parentheses  representing  b   coefficients  when  the  main  effect  and  time  × treatment  condition  interaction  terms   and  dlPFC  connectivity  parameter  estimates  are  entered  in  a  mixed  effect  linear   model  simultaneously.  ***p<.05;  **p=.05;  *p=.06                                                                                                                                                                                                                                                                           J.D.  Creswell   Table  I.  Baseline  Characteristics  of  Randomized  Participants.       Characteristic   Age        [mean  years  (SD)]   Gender            Male            Female   Ethnicity                Caucasian            African  American            Asian  American            Latino(a)            Native  American   Biracial            Other   Months  Unemployed   Education          No  high  school  degree        GED          High  school  degree          Technical  training          Some  college          Associate  degree          Bachelor’s  degree          Master’s  degree          MD/PhD/JD/PharmD   Body  Mass  Index   Cognitive  Impairment   (MMSE)   IL-­‐6  pg/mL   Post-­‐fMRI  Days  Elapsed   HEM   HER   Difference  Statistic   37.94   (10.96)     11   7     10   6   1   0   0   1   0   8.17(12.48)     1   1   1   3   4   2   2   3   1   27.15   (4.30)   29.39  (.70)   41.00  (9.55)   t(33)=  -­‐.48,  p=  0.64   1.81  (2.03)   5.06  (3.29)     χ2(1)=.24,  p=  0.63   9     8       χ2(5)=  6.37,  p=  0.27   13     2     0     1     0     0     1     10.58(20.31)      t(33)=  -­‐.43,  p=  0.67     χ2(8)=  8.43,  p=  0.39   0     0     2     2     3     0     7     3     0     26.44  (5.50)      t(32)=  .423,  p=  0.68   28.88  (1.15)   t(32)=  1.60,  p=  0.12   1.21  (.76)   5.29  (4.52)    t(28)=  1.03,  p=  0.31   t(32)=  -­‐.17,  p=.86     Notes:  Standard  deviation  values  are  provided  in  parentheses.  HEM=  3-­‐Day  Health   Enhancement  thru  Mindfulness  group;  HER=  3-­‐Day  Health  Enhancement  thru  Relaxation   group;  MMSE=  Mini-­‐Mental  State  Exam;  IL-­‐6=  Interleukin-­‐6;  Participants  indicated  their   ethnicity/gender  via  self-­‐report;  Post-­‐fMRI  Days  Elapsed=  number  of  days  elapsed  at  post-­‐MRI   session  from  the  conclusion  of  the  retreat  program.   29                                                                                                                                                                                                                                                                         J.D.  Creswell     Table  2.  Circulating  IL-­‐6,  by  Treatment  Group  and  Time     Mean HEM SE Mean HER SE Log-Transformed IL-6 Baseline .13 .07 -.06 .08 4-Month Follow-Up .08 .08 .08 .07 Baseline 1.87 .31 1.17 .32 4-Month Follow-Up 1.45 .32 1.41 .30 Raw IL-6   Notes:  Means  and  Standard  Errors  (SE)  from  IL-­‐6  MLMs,  with  Baseline  Age  as  a  covariate.  IL-­‐6=   Circulating  Interleukin-­‐6;  SE=  standard  error;  HEM=  Health  Enhancement  thru  Mindfulness  program;   HER=  Health  Enhancement  thru  Relaxation  program;  All  IL-­‐6  values  are  in  pg/mL.       30                                                                                                                                                                                                                                                                         J.D.  Creswell   31   Figure  1     A)             B)  Left  dlPFC  (MNI:  -­‐22,52,10)                                                              C)  Right  dlPFC  (MNI:26,44,34)                                                                                                                                                                                                                                                                                                 J.D.  Creswell   Figure  2       (A)     Left  dlPFC   Connectivity       Path  a:   -­‐.20***     Path  b:     (-­‐.43***)       Mindfulness  vs.   Relaxation  Training   Baseline-­‐4  Month     IL-­‐6   Path  c:    .190**   Path  c’:  (.133)             (B)       Right  dlPFC   Connectivity         Path  a:   -­‐.20***   Path  b:     (-­‐.34*)       Mindfulness  vs.   Relaxation  Training   Baseline-­‐4  Month     IL-­‐6   Path  c:    .190**   Path  c’:  (.143)                 32                                                                                                                                                                                                                                                                         J.D.  Creswell   33   Supplementary  Online  Information   Supplementary  Text     Detailed  Description  of  Mindfulness  (HEM)  and  Relaxation  (HER)  Programs     The  present  study  utilizes  two  strategies  for  increasing  the  scientific  rigor  of   meditation  research.  First,  we  have  developed  a  3-­‐day  condensed  retreat  format  for   teaching  mindfulness  meditation  training  entitled  the  Health  Enhancement  thru   Mindfulness  (HEM)  program  (which  is  based  on  exercises  and  didactics  taught  in   the  Mindfulness-­‐Based  Stress  Reduction  (MBSR)  program).  Delivery  of  the  HEM   program  in  a  retreat  format  ensures  a  consistent  and  strong  dose  of  mindful   awareness  practice  across  participants,  provides  a  relatively  controlled  external   environment  away  from  the  variety  of  stresses  and  supports  of  participants’  lives,   and  has  the  potential  to  minimize  drop-­‐outs.  Second,  we  have  designed  an  active   control  or  comparison  program,  Health  Enhancement  through  Relaxation  (HER)   that  is  structurally  similar  to  HEM,  but  avoids  cultivation  of  mindfulness  and  body   awareness.  The  use  of  a  structurally-­‐matched  active  comparison  group  helps  to   control  for  non-­‐specific  factors  such  as  group  support,  teacher  attention,  physical   activity,  and  mental  engagement.    Both  HEM  and  HER  consist  of  similar  behavioral   activities,  scheduled  on  an  hourly  basis  throughout  each  retreat  day,  such  as  sitting   quietly,  stretching,  slow  walking,  eating,  and  discussion.  However,  the  instructions   and  guidance  for  how  to  engage  in  the  activities  differ  in  systematic  ways.  In  the   HEM  program,  participants  are  instructed  to  train  their  attention  and  awareness  to   be  fully  present  to  each  of  these  activities,  learning  the  skill  of  mindful  awareness.  In   contrast,  HER  participants  engage  in  similar  guided  activities,  but  are  instructed  to   participate  in  these  activities  in  a  relaxed  and  restful  way,  to  let  their  minds  wander,   or  even  to  rest  or  sleep.     Common  features  of  the  HEM  and  HER  retreats  included:  individual  10   minute  interviews  with  the  teacher  on  Day  1  and  Day  3  in  order  to  build  rapport  and   also  assess  progress,  instructions  to  avoid  discussion  with  members  of  the  other   group  in  order  to  avoid  cross-­‐contamination  of  groups,  and  instructions  to  keep  up   with  practice.  CD’s  with  meditation  (HEM  group)  or  relaxation  exercises  (HER   group)  were  provided  for  home  use  following  the  retreat.       HEM  Program.  The  HEM  retreat  program  was  designed  as  an  adapted  and   condensed  version  of  the  8-­‐week  MBSR  program.  The  specific  content  of  the  HEM   program  was  developed  in  consultation  with  a  senior  teacher  at  the  UMass  Medical   School’s  Center  for  Mindfulness  (Melissa  Blacker,  MA).    On  retreat  day  1,  HEM   participants  were  welcomed  and  introduced  to  the  topic  of  mindfulness  meditation   for  stress  reduction,  and  oriented  to  the  program.    Following  participant   introductions,  the  group  engaged  in  a  mindful  eating  exercise,  and  the  body  scan   meditation  was  introduced  after  a  period  of  gentle  stretching.  Participants  were   encouraged  to  practice  mindful  awareness  of  eating  during  lunch.    Sitting   meditation,  walking  meditation,  and  mindful  movement  (gentle  hatha  yoga   postures)  were  also  introduced  on  retreat  Day  1.    After  each  formal  meditation                                                                                                                                                                                                                                                                         J.D.  Creswell   34   period,  participants  engaged  in  discussion  of  their  observations  about  themselves   and  the  practices.  During  all  interactions,  from  guided  meditations  to  discussions,   the  instructor  modeled  and  encouraged  attitudes  of  mindfulness  such  as  letting  go   of  judgment  and  expectations,  cultivating  self-­‐care,  patience,  and  friendly  curiosity   regarding  present  moment  experience.    In  the  evening  of  Day  1,  participants  were   asked  to  notice  pleasant  and  unpleasant  events  and  record  their  physical  sensations,   moods  or  emotions,  and  thoughts  in  their  retreat  notebook,  to  be  discussed  on  Day   2.      Participants  were  asked  to  continue  their  awareness  practice  by  remaining   relatively  quiet  after  the  group  sessions  ended  at  9PM.  During  Day  2,  which  began  at   7:30AM,  formal  mindfulness  meditation  practices  were  continued,  consisting  of   mindful  movement  (yoga),  sitting  meditation,  guided  and  unguided  walking   meditation,  and  meditation  on  various  objects  of  attention,  such  as  breath   sensations,  body  sensations,  sounds,  moods  or  emotions,  and  thoughts  and  the   spaces  between  thoughts.  Discussions  during  Day  2  included  exploration  of  pleasant   and  unpleasant  events,  stress  and  stress  physiology,  and  reacting  and  responding  to   stress.  The  structure  of  these  discussions  included  reflections  in  dyads  or  triads  as   well  as  the  entire  group.      Participants  were  asked  to  remain  quiet  during  the   evening  session  until  after  breakfast  on  Day  3.  The  intention  of  the  quiet  period  was   to  foster  self-­‐care  and  deepen  the  mindful  awareness  experienced  during  the  day’s   group  sessions.    On  retreat  Day  3,  formal  meditation  practices  continued,  with  the   addition  of  Choiceless  Awareness  and  Lovingkindness  or  Metta  meditation.     Building  from  the  previous  day’s  discussions  of  stress  reactivity  and  the  possibility   of  responding  to  stress  with  greater  flexibility  and  presence,  Day  3  discussions   focused  more  closely  upon  participants’  own  life  challenges,  and  their  plans  for   meeting  their  unemployment  and  job-­‐seeking  stress  with  mindfulness.  Participants   worked  individually,  in  dyads,  and  with  the  entire  group  to  develop  their  intentions   for  continuing  their  meditation  practice  and  for  applying  what  they  had  learned  at   the  retreat  to  their  daily  lives.    Closure  exercises  included  yoga  postures  or  favorite   stretches  taught  by  the  group  members,  and  sharing  of  good-­‐byes.           HER  Program.  The  HER  program  was  designed  to  maximize  similarity  in   activity  to  the  HEM  program  but  without  mindfulness  skill  development.  On  Day  1,   participants  were  oriented  to  the  HER  program  as  an  opportunity  to  rest  and   rejuvenate  and  gain  information  about  healthy  living.  They  had  the  opportunity  to   introduce  themselves  to  the  group  and  to  share  their  ways  of  managing  stress.  To   parallel  the  HEM  meditation  without  mindfulness  or  specific  instructions  about   awareness  of  body  sensations,  the  HER  instructor  led  imagery  exercises  focused  on   visualizing  a  stream  and  later  on  a  comfortable  and  safe  place.    To  parallel  the   dyadic  work  in  the  HEM  group,  the  HER  participants  worked  in  pairs  to  develop   scripts  for  pleasant,  relaxing  place  imagery,  and  guided  one  another  in  this   relaxation  exercise.  Discussions  throughout  Day  1  and  Day  2  included  topics  such  as   exercise  for  stress  management,  nutrition,  using  Michael  Pollan’s  Food  Rules,  and   Sleep  Hygeine.    To  match  the  mindful  yoga  of  the  HEM  group,  HER  participants   engaged  in  stretches  led  using  counting  or  joke-­‐telling  in  order  to  avoid  the   development  of  body  awareness  and  acceptance.  On  Day  2,  HER  retreatants   continued  to  engage  in  light  stretching,  took  walks,  engaged  in  drawing  and  imagery   exercises  involving  exploration  of  their  own  strengths  and  values.    Stress  and  Health                                                                                                                                                                                                                                                                         J.D.  Creswell   35   was  a  topic  for  discussion,  with  participants  working  in  dyads  as  well  as  the  whole   group  to  explore  what  they  experience  when  they  are  stressed.    Hobbies  and   interests  was  an  additional  topic  for  discussion.  On  Day  3,  HER  participants   continued  gentle  calisthenics  and  engaged  in  writing  about  what  they  were  learning   at  the  retreat.    Discussions  centered  on  stress  and  social  support,  and  participants   used  drawing  to  explore  their  own  social  circles  of  support,  and  discussed  this  in   dyads  and  the  entire  group.  Imagery  exercises  included  reflecting  on  the  future  and   where  participants  imagined  themselves  being  in  five  years.    The  group  closed  with   creation  of  and  sharing  of  personal  plans  for  a  healthy  daily  routine,  and  sharing  of   good-­‐byes.         Overview  of  Parent  Study  Measurement  Approach     The  present  report  focuses  on  testing  hypotheses  about  mindfulness  training   alterations  in  Default  Mode  Network  (DMN)  resting  state  functional  connectivity   (rsFC),  and  whether  these  training  neural  alterations  prospectively  predict  changes   in  the  health  biomarker  IL-­‐6  at  follow-­‐up.  These  neural  and  circulating  IL-­‐6   measures  are  part  of  a  broader  measurement  battery  in  this  study,  the  other   measures  we  aim  to  describe  in  other  published  reports.  Here  we  provide  an   overview  of  the  additional  measures  collected  in  this  study:     Baseline  Assessment:  psychosocial  and  individual  difference  self-­‐report   questionnaire  packet  (e.g.,  SF-­‐36,  Perceived  Stress  Scale,  Life  Experiences  Survey,   UCLA  Loneliness).     Baseline  MRI:  1  hour  functional  and  structural  neuroimaging  session  (high   resolution  structural  scan,  four  functional  scans:  resting  state  BOLD,  Multi-­‐Source   Interference  Task,  Affect  labeling,  and  Unemployment  Stress  Task).  At  the  end  of  the   fMRI  session  participants  completed  an  audio-­‐guided  mindfulness  meditation  task   (during  arterial  spin  labeling  perfusion  MRI).     Retreat  Assessment:  blood  sample  (pre-­‐retreat  prior  to  randomization),  palm  pilot   assessments  (5  times  per  day  during  all  retreat  days),  instructor  interviews  (during   Day  1  and  during  Day  3),  and  retreat  behavioral  tasks  (e.g.,  walking  speed,  implicit   tasks).     Post-­‐Treatment  MRI:  [same  measures  as  baseline  MRI]     4-­‐Month  Follow-­‐Up  Assessment  session  at  Retreat  Center:  blood  sample,  hair   sample,  psychosocial  self-­‐report  questionnaire  packet  [same  as  at  baseline,  plus   additional  program/study  evaluation  measures,  assessment  of  job  status  and  job-­‐ seeking  behaviors],  behavioral  task  measures  (implicit  tasks).                                                                                                                                                                                                                                                                               J.D.  Creswell   36   Table  S1.  Study  inclusion/exclusion  criteria.  Participants  were  required  to  meet  all  study   criteria  at  screening  to  be  enrolled  in  the  trial.     General   Inclusion:   • 24-­‐54  years   • Fluent  in  English   • Nonsmoker  (smoking  prohibited  at   retreat  center)   Exclusion:     • Cognitive  impairment  (<23  on  MMSE)   • Serious  mind-­‐body  practice   (>2x/week)   • Alcohol  Consumption  (>15   drinks/week)   • Recreational  Drug  Use  (past  month)   • Pregnancy  or  Breastfeeding   Health  &  Immune-­‐Related   Exclusion:   • Hospitalization  for  mental  or  physical   health  problem  (past  6  months)   • Chronic  disease  (e.g.,  cancer,  HIV,   diabetes,  heart  disease,  arthritis)     • Medication  use:   o Mental  health  problems   o Cholesterol-­‐lowering  meds   (Crestor,  statins)   o Immune-­‐related  treatment   (corticosteroids,  chemotherapy,   transplantation  medication)   fMRI  Criteria   Exclusion:   • Left-­‐handed   • Presence  of  metal  in  body   • Weight  >350  pounds   • • Claustrophobia   Neurological  disorders  (epilepsy,  spina   bifida,  stroke,  traumatic  brain  injury)   Stress  &  Unemployment   Inclusion:   • Unemployed  (working  <20   hours/week)   • Seeking  employment   • Having  trouble  finding  a  job   • Stressed:  total  score  >5  on  job-­‐seeking   PSS  (4  items  listed  at  right)   o 5-­‐point  Likert  scale  (0  –  never  to  4   –  very  often)             Adapted Job Seeking Stress Screening Items 1. In the last month, how often have you felt that you were unable to control the important things in your life? 2. In the last month, how often have you felt confident about your ability to handle your jobrelated problems? This could include your efforts at finding a job, or paying your bills. (reverse scored) 3. In the last month, how often have you felt that things were going your way in finding a job? (reverse scored) 4. In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?                                                                                                                                                                                                                                                                         J.D.  Creswell   Table  S2.  Whole-­‐brain  analyses  of  regions  displaying  increased  PCC-­‐seeded  resting  state   functional  connectivity  in  the  mindfulness  relative  to  the  relaxation  treatment  program  in  a   spreading  interaction  analysis  using  contrast  weights:  [-­‐1(pre,HEM),  -­‐1(pre,HER),  3(post,   HEM),  -­‐1(post,HER)]  (  (p<  0.005,  k>50).  Specifically,  regions  that  are  more  coupled  with  PCC   after  the  HEM  training  (relative  to  HER  training).       Region   Cluster  Size   Peak  MNI  Coordinates   Z   Right  Superior  Frontal  Gyrus   147   20  62  -­‐6   3.97   Left  Pallidum   72   -­‐20  -­‐4  2   3.68   Left  Insula   57   -­‐26  18  -­‐8   3.98   Left  Pallidum   62   -­‐6  0  -­‐6   4.09   Left  Thalamus   54   -­‐28  -­‐26  8   3.58   Left  Middle  Frontal  Gyrus   117   -­‐22  52  10   3.44   Left  Superior  Frontal  Gyrus   79   -­‐30  42  38   3.11   Right  Supramarginal  Gyrus   66   60  -­‐44  38   3.35           37                                                                                                                                                                                                                                                                         J.D.  Creswell   38   Table  S3.  Whole-­‐brain  analyses  of  regions  displaying  decreased  PCC-­‐seeded  resting  state   functional  connectivity  in  the  mindfulness  relative  to  the  relaxation  treatment  program  in  a   spreading  interaction  analysis  using  contrast  weights:  [1(pre,HEM),  1(pre,HER),  -­‐3(post,   HEM),  1(post,HER)]  (  (p<0.005,  k>50).  Specifically,  regions  that  are  less  coupled  with  PCC  after   the  HEM  training  (relative  to  HER  training).         Region   Cluster  Size   Peak  MNI  Coordinates   Z   Right  Inferior  Temporal  Cortex   54   46  2  -­‐44   3.23   Right  Fusiform  Gyrus   56   36  -­‐62  -­‐14   3.96   Left  Lingual  Gyrus   353   -­‐12  -­‐78  -­‐4   4.09   Left  Inferior  Frontal  Gyrus   55   -­‐46  32  0   3.33   Right  Insula   57   32  -­‐26  20   3.57   Right  Postcentral  Gyrus   235   48  -­‐4  28   4.55   Left  Precentral  Gyrus   91   -­‐52  -­‐12  36   3.34   Right  Paracentral  Lobule   82   4  -­‐34  64   3.43   Right  Precentral  Gyrus   464   10  -­‐6  80   4.36                                                                                                                                                                                                                                                                               J.D.  Creswell   39   Table  S4. A detailed overview of the 3-day Health Enhancement thru Mindfulness (HEM) and Health Enhancement thru Relaxation (HER) intervention programs.   Health Enhancement thru Mindfulness   Welcome and orientation to the guidelines for participation in the program; introduce the concept of mindfulness; centering exercise and individual introductions; mindful raisineating exercise; 45-minute body scan exercise (show alternative postures, and start with some stretching).   Depending on when program starts, raisin and body scan may be after lunch.       12:15- Mindful lunch (not silent) (1/2 hr in 1:45   cafeteria, 1 hr break) (3 student interviews)     1:45Continue intros, raisin exercise, body 3:15   scan as needed, depending on progress prior to lunch.     Discussion of body scan/morning, weave Conf in relevant ‘attitudes of mindfulness’ area   (non-striving, patience, don’t know mind) intro to sitting and postures, sitting meditation with AOB. Review definitions of mindfulness as indicated. Use Mountain meditation or other stabilizing imagery as indicated.   3:15Snack /tea (2 student interviews)   4:15     4:15Standing and walking   5:45   Mindful movement yoga, Intro awareness of pleasant events assignment   Chapel   (to discuss tomorrow)     10:0012:00     Conf area   5:457:30   7:30-9     Conf area   5:30-7:30     Day 1   10 12:15     chapel   Mindful dinner (2 student interviews)     Sit (~15 min). Walking. Seated body scan, followed by stretching. Use Mountain imagery as indicated.   Reflection and discussion of 1st today’s experiences, integrating attitudes of Health Enhancement thru Relaxation   Orientation, guidelines*, and introductions (pairs first, then big group)   How do you manage stress? (keep it somewhat light) (this may generate a list of diverse coping methods, and can be referred back to as the weekend goes on).   End session with stream imagery for relaxation.       12:00-1:30   Lunch (3 student interviews)     1:30-3:00     chapel   Brief disc of physical activity as method for stress reduction.   Stretching / exercise.   Discussion what keeps you from exercising, if you don’t--list   3:00-4:00   Snack / tea (2 student interviews)     Intro to safe place.   Demonstration of dyadic work--development of an image of being safe and at ease. Work in pairs, writing down, then guiding each other. (comfort and ease imagery)   Share feedback in large group   Dinner (2 student interviews)     Sleep Hygiene: talk about the connection between sleep and health. Ask them how they sleep (this could take a long time!) Give Sleep Hygiene handout and go over it 4:00-5:30     Conf area   7:30 – 9     chapel                                                                                                                                                                                                                                                                         J.D.  Creswell   mindfulness.   Orientation and instructions for continuing mindfulness practice during the later evening and bedtime, noting pleasant and also unpleasant events, (give calendars) returning wandering mind to present moment awareness. Lying down Body Scan.   40   briefly.   Nutrition (using Pollen material)   What is true; what is helpful here?     Reminder: maintain quiet this evening and in early morning (if you talk, keep it fairly quiet)     Relaxation exercise for sleep.   9:00-10:30   Quiet time (4-5 interviews)     10:30pm   Lights out   910:30   10:30 pm     Day 2   7am   7:308:15   Chapel   Quiet time (4-5 interviews)     Lights out   8:159:15   9:1510:45     Conf area   Mindful Breakfast (silence optional)   8:30-9:30   Brief sit (~15 minutes). Discussion of last evening and this morning’s mindfulness practice, including silence; dyads and then group disc/ mindful listening re: pleasant and unpleasant events. Begin disc of what makes an event pleasant/ unpleasant; topic of ‘stress/ stress physiology’   Walking meditation (unguided)   9:30-11     Chapel   Physical Exercise   ‘non-mindfulness’ version of yoga stretch/strengthening exercises.   Incorporate some HEP exercise activities.     Something to think about on your walk: what are your personal strengths?   11-11:30   Take a walk   Mindful stretch followed by sit. Continue disc of anatomy of stress, stress reactivity /automatic pilot and possibility of responding mindfully to stressful events. Use metaphors as indicated (e.g, 4 story building, waterfall and barrel, etc.)   Mindful lunch   11:3012:15     Chapel   Discussion of strengths and values. Who are your role models and what do you admire re: them?   Exercise: draw your personal ‘values and strengths’ coat of arms. Share ‘coat of arms’ in dyads and then large groups.   10:4511:15   11:1512     Conf area   12-1     Wake up (maintain silence)   Mindful stretch – on floor, followed by sit with AOB, AO body sensations. Prep for mindful breakfast.     Day 2   7am   7:30-8:30   Conf .     Wake up (maintain quiet)   Movement – gentle calesthenics   Reflective writing re: ideas sparked by yesterday’s and last night’s disc (what do they want to explore, what do they want to remember.)   Breakfast   12:15-1:15   Lunch     1-2:30   Sit with AOB, Body, AOSound, AO 1:15-3   Demonstration of development of imagery for Chapel   Thought (internal image/talk). Disc of feeling competent and effective, or strong.     participants own stress reactivity patterns Conf area   They break into dyads to do this.   (start with dyads) how do they know (if you don’t have a memory to use, can                                                                                                                                                                                                                                                                       J.D.  Creswell   2:303:30   3:305:30     Conf area   they are stressed and how do they take care of self. Disc mindful approach to emotions; problem focused and emotion focused coping. Lovingkindness meditation if appropriate. Introduce possibility of silent period (to start ~4:30pm).   Snack / tea     Chair yoga followed by sit. Continue discussion of stress reactivity and mindful responding as needed. Disc of upcoming silent time with rationale as a time to be with yourself, focus on yourself in a caring way. If group/individuals seem able to meet this, silence begins (~4:30pm) run this section like MBSR retreat: body scan, mindful movement, sitting, lovingkindness meditation (include choiceless awareness)   41   imagine one – like you are an actor in a scene)     Share feedback in large group   3-4   4-5:45     Chapel   5:307:15   7:15-9   Chapel   Dinner     Practice in silence. Variety of formal practices, include choiceless awareness. Poetry reading. Mountain or Lake meditations as indicated.   5:457:30pm   7:30-9   Conf area   910:30   10:30     Day 3   7am   7:308:15   Quiet time     Lights out       Wake up   Mindful stretching/yoga followed by brief body scan.   9-10:30   10:30     Day 3   7am   7:30-8:30   chapel   Snack / tea     Stress physiology   Stress and Health: JKZ handout,   Charting arousal/time and symptom development—the arousal curve and what symptoms shows up at which level of arousal. Do this as a group discussion. What do they experience when they are stressed..   Do this in dyads first, and then expand to the bigger group. This should take approximately 1 hour.     Go back to Pollen material if did not finish it yesterday.   Earlier today we did imagery exercises around our strengths. What personal strengths do you have? Hobbies and interests – have a go around about favorite hobbies and interests that help them to feel comfortable / good. Create a group list of ideas.   Dinner     Disc – how do I have / make fun?   Demonstration and then guided imagery for fun, breaking into dyads to develop them.   Humorous video     Quiet time     Lights out       Wake up   Movement – gentle calesthenics - narrative writing – free association or free drawing                                                                                                                                                                                                                                                                         J.D.  Creswell   Conf area   8:159:15   9:1510   chapel   10-11   11-12     chapel   121:30   1:30-3     Conf area   3-3:45   3:455:30   chapel   5:307:30     42   What am I learning/ do I want to take with me?   Mindful breakfast (silence)   8:30-9:30   Breakfast   Continue formal practices (guided). 9:30-10:30   Disc of stress and social support. Support can Seeing meditation. Participants remain in Conf area   come from the living or the deceased. Even just silence.   a small thing from memory can be powerful (Nazi example)   Draw circles of intimacy/ mandala of social support system   Walking unguided. Mindfulness in 10:30Quiet Rest or take a walk (3 interviews)   motion (3 interviews)   11:30     Sit. Awareness Exercises for breaking 11:30Discussion of support system /mandalas in silence (dyads, etc). Group discussion of 12:15   dyads and in big group. Brief relaxation silence. If we have not engaged in Conf area   focusing on feeling competent and connected, silence, continue discussion of mindful Supported.   approaches for self care and responding to stress.     Mindful Lunch (optional silence) (3 12:15-2:00   Lunch (3 interviews)   interviews)       Continue formal practices. Include 2:00-3:30   Obstacles to applying who you have learned and Mountain meditation as well as other how you are going to work with them     formal practices already introduced. Chapel     Mindfulness in everyday life disc.   Guided reflections for the future – where do you Obstacles to applying what you have imagine yourself 5 years from now? Who is learned and how you are going to work there? What are you doing?   with them. Formal practice – body scan. Discuss in small and large groups   Development of personal ‘action plans’     Break / tea (2 interviews)   3:30-4:15   Break / tea (2 interviews)       Practice. group led yoga stretches if 4:15-5:30   Making a personal plan for your self: creation of appropriate. Discussion of personal plans   daily routine, relaxation, etc. Final for continuing practice. Final comments Conf area   comments/closing exercises.   and closure exercises. Dyads and whole Drawing—this group and this time here.   group.   Final study procedures (5 interviews)   5:30-7:30   Final study procedures (5 interviews)                                                                                                                                                                                                                                                                                             J.D.  Creswell   43   Enrollment   Figure  S1.  Consort  Flowchart      Excluded  at  screening  (n=  661)                      -­‐Unreachable  (n=  216)   Interested  in  program  (n=  89)                                      -­‐-­‐Not   Employed  (n=  63)                    -­‐Heavy  smoker  (n=  58)   health  medication  (n=  38)                                    -­‐-­‐Mental   Low  stress  score  (n=  36)                        -­‐Metal  concerns  (n=  20)                  -­‐Not  looking  for  a  job  (n=  19)                  -­‐Not  available  for  retreat  (n=  17)                  -­‐Age  (n=  16)                  -­‐Left-­‐handed  (n=  16)                  -­‐Psychologist/Psychiatrist  (n=  13)                  -­‐Cholesterol  medication  (n=  13)                  -­‐Mind/body  practice  (n=  6)                  -­‐Chronic  Disease  (n=  5)                  -­‐Other  (n=  36)   Assessed  for  Eligibility   (N=  763)     Baseline  Assessment  (n=  62)              Eligible  But  Did  Not  Attend  Baseline   Assessment    (n=  40)                  -­‐Did  not  return  phone  calls  (n=  14)                  -­‐Did  not  show  up  to  appt  (n=  14)                  -­‐Cancelled  appointment  (n=  4)                  -­‐Not  available  on  retreat  dates  (n=  4)                  -­‐Received  job  offer  (n=  2)                  -­‐No  longer  interested  (n=  1)                  -­‐Hospitalized  (n  =  1)               Excluded  At  Baseline  (n=  7)                  -­‐Metal  concerns  (n=  4)                  -­‐Not  technically  unemployed  (n=   1)                  -­‐Mental  health  medication  (n=  1)                  -­‐Unable  to  attend  retreat  (n=  1)      Pre-­‐Retreat  fMRI    (n=  50)     Did  Not  Complete  fMRI  (n=  5)    Did  Not  Attend  fMRI  (N=  5)                  -­‐Did  not  return  phone  calls  (n=  2)                  -­‐Received  job  offer  (n=  2)                  -­‐Did  not  show  up  (n=  1)                                        -­‐Metal  concerns  (n=  2)   Allocation                    -­‐Claustrophobia  (n=  2)                  -­‐Unable  to  scan  (n  =1)   Randomized   (n  =  35)       Allocated  to  3-­‐Day  M indfulness  Retreat  (n  =   18)                          Treatment          3-­‐Day  M indfulness  Retreat  sample  (n=  18)                Post-­‐Test    Did  Not  Attend  R etreat  (N=  10)                  -­‐Did  not  return  phone  calls  (n=  1)                  -­‐Received  job  offer  (n=  3)                  -­‐Did  not  show  up  at  retreat  center  (n=                      3)                  -­‐Sick  (n=  2)                  -­‐Abnormal  scan  (n=  1)             Did  not  complete   retreat  (n=  1)   Allocated  to  3-­‐Day  Relaxation  Retreat  (n  =  17)                  3-­‐Day  Relaxation  Retreat  sample  (n=17)       Did  not  complete   retreat  (n=0)      R   etained  at  Post-­‐fMRI  (n=17)              Treatment  sample  retained  (n=  17)              Treatment  sample  dropouts  (n=  1)                        Retained  at  Post-­‐fMRI  (n=17)                                Treatment  sample  retained  (n=  17)                  Treatment  sample  dropout  (n=  0)                        Retained  at  4-­‐Month  Follow-­‐Up  (n=  17)              Treatment  sample  retained  (n=  17)              Treatment  sample  dropouts  (n=  0)                        Retained  at  4-­‐Month  Follow-­‐Up  (n=  17)                                Treatment  sample  retained  (n=  17)                  Treatment  sample  dropout  (n=  0)