Eighteen subjects were recruited for participation in this study. During psychophysical training, one subject was dismissed for low sensitivity and one for hypersensitivity to heat stimulation. A third subject was dismissed for falling asleep during the course of the meditation intervention. Fifteen healthy volunteers, six males and nine females (age range: 22–35 years, mean age: 26 years) completed the study. All subjects were right-handed, 13 were White, one Asian, and one Hispanic. Subjects gave written, informed consent recognizing that: 1) they would experience painful, heat stimuli, 2) all methods were clearly explained, and 3) they were free to withdraw from the study without prejudice. The Institutional Review Board of the Wake Forest University School of Medicine approved all procedures.
Overview of Experimental Procedures
An outline of experimental procedures is illustrated in . Subjects participated in a psychophysical training session to familiarize them with the noxious stimuli and psychophysical rating procedures (see Psychophysical Training). The first MRI session was conducted after psychophysical training but before meditation training (). In the first half of MRI session 1, subjects were instructed to close their eyes and reduce movement (rest condition). In the second half of both MRI sessions, subjects were instructed to “meditate by focusing on the changing sensations of the breath.” In MRI session 1, this condition was labeled as “attention to breath”-ATB, thereby providing a control (i.e., divided attention) for comparisons with the meditation condition after training. Subjects then participated in four days of mindfulness meditation training (see Mindfulness-Based Mental Training). After successful completion of meditation training, subjects returned for MRI session 2. Similar to MRI session 1, in the first half of MRI session 2, subjects were instructed to close their eyes and reduce movement (rest condition). Afterwards, subjects were instructed to “begin meditating by focusing on the changing sensations of the breath (“mindfulness-based attention to breath” meditation).
Figure 1 Experimental procedures across time. (First column). Psychophysical Training Session: Subjects first came in for psychophysical training. In this session, subjects were familiarized with visual analog scales, the range of thermal stimulation, and thermal (more ...)
All subjects were initially familiarized with thirty-two, 5s duration stimuli (35–49°C) in order to provide experience with the visual analog scales (VAS). After familiarizing subjects with the heat stimuli, they received five minutes and 55 seconds of stimulation, identical to the paradigm employed in subsequent fMRI experimental sessions. A 15 cm plastic sliding VAS scale was used to quantify pain intensity and unpleasantness [Paresian Novelty (Price et al., 1994
)]. The minimum rating was represented as “no pain sensation” or “not at all unpleasant,” whereas the maximum was designated with “most intense imaginable” or “most unpleasant imaginable.”
MRI Session 1
Subjects were positioned in the MRI scanner, a pulse oximeter was attached to each subject’s left index finger to assess heart rate, and a transducer was placed around the chest to gauge respiration rate. Noxious thermal stimuli were delivered to the posterior aspect of the right calf by a 16 × 16 mm2 TSA II thermal stimulator (MedocRamat Yishai, Israel). The “heat” condition consisted of thermal stimuli that were administered in alternating patterns of heat (49°C) and neutral (35°C) with 12-s durations at each temperature (5 min and 55s total duration per MRI series). The baseline temperature was maintained at 35°C, and stimulus temperatures were delivered with rise and fall rates of 6°C/s. The “neutral” series consisted of only neutral stimuli (35°C for 5 min and 55s). Thermal stimuli were counterbalanced across series. The heat probe was moved to a new location on the right calf after each series to reduce habituation. After each series, subject’s evaluation of pain intensity and unpleasantness were acquired with the VAS scale. They were instructed that their ratings should reflect the overall experience of the entire series.
In MRI session 1, four functional series (2 heat; 2 neutral) were separated by a structural acquisitions can. In the first half of the experiment, subjects were instructed to keep their eyes closed and restrict movement across conditions. After the structural image was obtained, subjects were instructed to “meditate by focusing on the changing sensations of the breath.”
Mindfulness-Based Mental Training
Mindfulness-based mental training was carried out in four separate, 20-minute sessions conducted by a facilitator with over ten years of experience leading similar meditation regimens. Subjects had no prior meditative experience and were informed that such training was secular and taught as the cognitive practice of Shamatha or mindfulness meditation. Each training session was held with one to three participants.
On mindfulness meditation training day 1, subjects were encouraged to sit with a straight posture, eyes closed and to focus on the changing sensations of the breath occurring at the tips of their nostrils. Instructions emphasized acknowledging discursive thoughts and feelings and to return their attention back to the breath sensation without judgment or emotional reaction whenever such discursive events occurred. On training day 2, participants continued to focus on breath-related nostril sensations and were instructed to “follow the breath,” by mentally noting the rise and fall of the chest and abdomen. The last ten minutes were held in silence so subjects could develop their meditative practice. On training day 3, the same basic principles of the previous sessions were reiterated. However, an audio recording of MRI scanner sounds was introduced during the last ten minutes of meditation to familiarize subjects with the sounds of the scanner. On the final training session (day 4), subjects received minimal meditation instruction but were required to lie in the supine position and meditate with the audio recording of the MRI sounds to simulate the scanner environment. Contrary to traditional mindfulness-based training programs, subjects were not required to practice outside of training.
Subjects also completed the Freiburg Mindfulness Inventory short-form (FMI), a 14-item assessment that measures levels of mindfulness, before psychophysical pain training and after MRI session 2. The FMI is a psychometrically validated instrument with high internal consistency (Cronbach alpha = 0.86) (Walach, 2006
). Statements such as “I am open to the experience of the present moment” are rated on a 5-point scale from 1 (rarely) to 5 (always). Higher scores indicate more skill with the mindfulness technique.
MRI Session 2
After successful completion of meditation training, subjects participated in MRI session 2. This session consisted of eight functional series (four heat; four neutral). After completion of the first four “rest” series, subjects were again instructed to “meditate by focusing on the changing sensations of the breath” at which point the anatomical scan was conducted. Subjects were instructed to meditate until the end of the experiment. Four minutes after the anatomical scan, functional acquisition was resumed, and four meditation series were obtained (). Evaluation of pain ratings and experimental procedures were identical to MRI session 1.
CBF images were acquired on a 1.5 T General Electric Twin-Speed LX Scanner with an 8-channel head coil (General Electric Medical Systems) pulsed arterial spin labeled MRI technique(PASL, Q2TIPS-FAIR) (Luh et al., 1999
). Scan parameters for the PASL Q2TIPS-FAIR acquisition are as follows: TR=2500 ms, TE=17.9 ms, TI=1700 ms, TI1=700 ms, TI2=1200 ms, FOV = 24 cm × 18 cm, matrix size = 64 × 48, slice thickness= 8 mm, slice gap =0 mm, number of slices=13, total number of volumes = 140, frequency direction = A/P, total scan time = 5 minute 55 seconds. A 2D Echo Planar Imaging trajectory with ramp sampling was used for image acquisition. PASL is sensitive to subject motion, which may lead to inaccurate CBF maps. To remove the influence of subject motion on CBF quantification, the Q2TIPS-FAIR time series data (volumes 8 through 140) was filtered to remove individual perfusion weighted images that may corrupt the final CBF map (Tan et al., 2009
). The first image in the PASL Q2TIPS-FAIR data was acquired with all inversion and saturation pulses turned off. This image was used as the M0 image to scale raw perfusion weighted images into a quantitative CBF map according to the General Kinetic Model (Buxton et al., 1998
). Volumes 2 through 7 were needed to establish a steady state prior and were not included when calculating the CBF maps.
After the rest condition, an accelerated (2x) T1 weighted Inversion Recovery 3D Spoiled Gradient Echo (IR-3DSPGR) structural scan was acquired (scan parameters: flip Angle = 12 degrees, TI= 600 ms, RBW=15.6 kHz, FOV=24×24, matrix size=240×240, slice thickness=1 mm, number of slices = 164, acceleration factor = 2, total scan time=3 minutes 57 seconds).
Statistical Analyses of Regional Signal Changes within the Brain
The functional image analysis package FSL [Functional Magnetic Resonance Imaging of the Brain (FMRIB) Software Library (Center for FMRIB, University of Oxford, Oxford, UK) was used for image processing and analyses. Functional data were movement corrected and spatially smoothed with a 8 mm full-width at half-maximum 3-D isotropic Gaussian kernel. Each functional image was scaled by its mean global intensity (intensity normalization). To minimize confounds arising from global CBF fluctuations, all subject’s functional images were registered to their structural data using a seven-parameter linear 3-D transformation and then transformed into non-linear standard stereotaxic space. High-resolution T-1 weighted images were used to classify brain regions of activation and normalize images to standard space.
Statistical analysis of regional signal changes was performed on each acquisition series (first level analyses) using a fixed effects general linear modeling approach (Wool rich et al., 2001
). Random effects analyses assessed activation across individuals. Z
(Gaussian zed T/F) statistic images were thresholded using clusters determined by z
>2.3 and a corrected cluster significance threshold of p
<0.05 (Worsley et al., 1992
A two-way Repeated Measures Analysis of Variance (RM ANOVA) was conducted on data from MRI session 1 in order to identify the main effects of pain and ATB. A similar RM ANOVA was conducted on data from MRI session 2 in order to identify main effects of pain and meditation, as well as their interaction. Paired comparisons were used to assess differences of main effects of pain and ATB/meditation between MRI sessions. These comparisons allowed us to differentiate brain mechanisms related to meditation (MRI session 2) from those associated with simple attention to breath (MRI session 1). To ensure that there was equal statistical power across sessions, we included only the first two volumes of the rest and meditation conditions in MRI session 2 were included to match the four volumes from MRI session 1.
In MRI session 2, we conducted a multiple regression analysis employing three regressors was conducted to assess the relationship between individual differences in meditation-induced pain reduction and brain activation. The first regressor was the mean effect of meditation vs. restin the presence of heat stimulation. Demeaned percent changes for each subject’s pain intensity ratings were entered as the second regressor, and demeaned percent changes for each subject’s unpleasantness ratings were entered as the third regressor. Intensity and unpleasantness regressors were orthogonal zed to each other in order to identify variability in brain activity uniquely related to each aspect of pain.
Analysis of Behavioral and Physiological Data
A RM ANOVA tested the effects of MRI session (1 and 2) between conditions (rest and ATB/meditation) on pain intensity and unpleasantness ratings (SPSS Inc). We also examined FMI scores before and after mindfulness meditation training.
RM ANOVAs also tested the effects of rest, ATB, and meditation across neutral and heat stimulation condition son heart rate, respiration rate, and global CBF across sessions (SPSS Inc). Due to equipment malfunction, we employed list wise deletion of six subjects for heart rate data and five subjects for respiration data in MRI session 1(Allison, 2002