Running Head: TM-SIDHI BRAIN PATTERNS1

NOTICE: this is the author’s version of a work that was accepted for publication in International Journal of Psychophysiology. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in International Journal of Psychophysiology, 2011, 81, 198–202

Comparison of Coherence, Amplitude, and eLORETA Patterns

during Transcendental Meditation and TM-Sidhi Practice

Fred Travis1

1Center for the Brain, Consciousness and Cognition

Maharishi University of Management

1000 North 4th Street, Fairfield, IA 52557

Abstract

This random-assignment study compared coherence, amplitude, and eLORETA patterns during practice of the Transcendental Meditation (TM) and the TM-Sidhi programs. The TM technique involves systematic transcending of contents of experience to a state of pure consciousness. The TM-Sidhi program involves sanyama—the simultaneous experience of dhārānā (fixity), dhyāna (transcending) and samādhi (pure consciousness). Thirty-two channel EEG was recorded from experienced TM subjects randomly assigned to two consecutive 10-min TM sessions or to a 10-min TM session followed by 10-min TM-Sidhi practice. Compared to TM practice, TM-Sidhi practice was characterized by higher frontal alpha1 and beta1 amplitudes, and eLORETA-identified sources of alpha1 EEG in right-hemisphere object recognition areas including the right parahippocampus gyrus, right fusiform gyrus, lingual gyrus, and inferior and medial temporal cortices. These cortical areas are involved in specific/holistic representation of words. The observed brain patterns support the descriptions of sanyama as including both specificity (sutras or verses), as suggested by higher frontal beta1 EEG amplitude and by eLORETA sources in right-hemisphere object-recognition areas, and holistic experience (pure consciousness) as suggested by higher frontal alpha1 EEG amplitude. These EEG patterns fit the complex description of sanyama.

Key Words: TM technique; TM-Sidhi program; coherence; eLORETA, word recognition

Comparison of Coherence, Amplitude, and eLORETA Patterns

during Transcendental Meditation and TM-Sidhi Practice

1.Introduction

Research has described two features of consciousness: the level of consciousness (graded levels of being awake or asleep), and the contents of consciousness (inner thoughts, feelings, and perception of outer objects) (Koch & Tsuchiya, 2007; Tsuchiya & Adolphs, 2007). These two features of consciousness are intertwined during ordinary waking experiences. Consequently, most scientists beginning with William James concluded that consciousness cannot be experienced without an object (James, 1890/1951; Natsoulas, 1997). However, these two features of consciousness can be separated during meditation practices, allowing exploration of conscious contents and levels of consciousness.

Three meditation categories have been described that are distinguished by cognitive processes and EEG patterns (Travis & Shear, 2010). The first two categories, Focused Attention and Open Monitoring, include both contents of consciousness and levels of consciousness. In Focused Attention meditations, the level of consciousness is intertwined with contents—theobject of sustained focus completely fills awareness. In Open Monitoring meditations, the level of consciousness begins to be separated (mindful) from changing contents,objects of experience such as body states, thoughts, feelings or breath. The third category, Automatic Self-Transcending, includes meditations designed to transcend the procedures of the meditation. These techniques minimize the contents of consciousness and so allow exploration of levels of consciousness devoid of content.

The Transcendental Meditation™ (TM™) technique is in the third category of meditations. TM practice is a process of transcending, which involves appreciating a mantra at ‘‘finer” levels in which the mantra becomes increasingly secondary in experience and ultimately disappears, while self-awareness becomes more primary (Maharishi Mahesh Yogi, 1969; Travis & Pearson, 2000). This state is described as “pure consciousness” in which consciousness is open to itself (Maharishi Mahesh Yogi, 1994). TM practice has been characterized by 1) EEG patterns—higher frontal and central alpha power (Banquet, 1973; Dillbeck & Bronson, 1981; Hebert, Lehmann, Tan, Travis, & Arenander, 2005; Travis & Wallace, 1999), and higher frontal alpha coherence (Dillbeck & Bronson, 1981; Gaylord, Orme-Johnson, & Travis, 1989; Levine, 1976; Travis & Arenander, 2006; Travis et al.; Travis, Tecce, & Guttman, 2000); 2) physiological patterns—lower breath rate, lower skin conductance and lower plasma lactate (Dillbeck & Orme-Johnson, 1987); 3) MEG patterns—source localization of MEG activity in medial prefrontal and anterior cingulate cortices (Yamamoto, Kitamura, Yamada, Nakashima, & Kuroda, 2006); 4) eLORETA source localization—sources of alpha1 activity in midline frontal and parietal cortices that are part of the default mode network (Travis et al., 2010); and 5) patterns of cerebral metabolic rate in a pilot PET study—higher frontal and parietal activity and lower thalamic activity, compared to eyes-closed rest (Newberg et al., 2006).

While TM practice has been extensively investigated, no studies have compared brain patterns during TM with those during the advanced TM program, the TM-Sidhi program. The TM technique involves transcending; in contrast, the TM-Sidhi program involves sanyama—the simultaneous processes of dhārānā (fixity), dhyāna (transcending) and samādhi (pure consciousness) (Maharishi Mahesh Yogi, 1978). (The TM-Sidhi program is described in more detail below in the procedure.) Since TM-Sidhi practice involves both changing objects of attention (dhārānā), transcending(dhyāna), and the experience of pure consciousness(samādhi) then, compared to TM practice, one might expect heightened alpha1 activitycharacteristic of Automatic Self-Transcending and heightened beta1 activity characteristic of active processing, during TM-Sidhi practice.

This random-assignment study compared EEG amplitude and coherence during Transcendental Meditation and TM-Sidhi practice in theta2 through gamma frequency bands. It also investigated eLORETA patterns during these two practices. eLORETA was developed at the KEY Institute for Brain-Mind Research at the University of Zurich (Pascual-Marqui, Michel, & Lehmann, 1994) to compute the 3-D intracerebral distribution of sources of scalp-recorded electrical potentials (Pascual-Marqui, 2002). Two refinements of this method have been released: first, sLORETA (standardized Low Resolution Electromagnetic Tomography), which uses standardized current density to calculate intracerebral generators, and recently eLORETA (exact Low Resolution Electromagnetic Tomography), which does not require standardization for correct localization (Pascual-Marqui, 2002, 2007). Both sLORETA and eLORETA are argued to have low resolution but zero localization error even in the presence of measurement and biological noise (Pascual-Marqui, 2007). The current implementations of sLORETA and eLORETA uses a realistic head model calculated by Fuchs (Fuchs, Kastner, Wagner, Hawes, & Ebersole, 2002), and electrode coordinates provided by Jurcak (Jurcak, Tsuzuki, & Dan, 2007).

The hypothesis tested in this study was that compared to TM practice, TM-Sidhi practice would be characterized by higher levels of alpha1and beta1 amplitude and coherence. No predictions were made about eLORETA sources, since only one paper has reported results from eLORETA during TM practice.

2. Material and Methods

2.1. Subjects

Twenty-six subjects volunteered to participate in this study—12 men and 14 women—average age 49.0 ± 14.5 yrs, who had been practicing the TM technique for 25.6 ± 11.6 yrs and the TM-Sidhi program for 19.4 ± 10.8 yrs. Subjects were randomly assigned to practice TM for two consecutive 10-min sessions (TM-Only Group), or to practice TM for 10 minutes followed by 10-min practice of the TM-Sidhi program (TM-Sidhi Group). We used a between-design, because TM-Sidhi practice always follows TM practice. Thus, it would not have been appropriate to use a within design with counterbalanced TM and TM-Sidhi sessions.

2.2. Procedure

Subjects came in individually for their EEG measurement in the early afternoon. Thirty-two EEG active-sensors were applied according to the 10/10 system using the BIOSEMI ActiveTwo amplifier and acquisition software ( Potentials at the left and right ear lobes were also measured for calculating an averaged-ears reference offline. EEG was recorded for two ten-minute periods and stored for analysis off line.

2.3. Meditation Practices

2.3.1. The Transcendental Meditation technique

TheTM practice is a mental procedure practiced for 20 minutes sitting with eyes-closed. During TM, one allows the attention to experience ‘‘finer” levels of a mantra—the mantra becomes increasingly secondary in experience and self-awareness becomes increasingly primary (Maharishi Mahesh Yogi, 1969; Travis & Pearson, 2000). In the process of transcending, attention moves from the ordinary thinking level to the least excited state of consciousness— consciousness without content, called pure consciousness (Maharishi Mahesh Yogi, 1969; Travis & Pearson, 2000).

Unlike most mantra meditations, the mantras used in TM practice are used for their sound value and not for any possible meanings. Also, unlike most mantra meditations, TM is not a process of concentration—keeping the mantra in awareness or continued mental rehearsal of the mantra. Rather, TM is a process of automatic transcending (see (Travis et al., 2010; Travis & Shear, 2010) and (Cahn & Polich, 2006) for a discussion of the concept of automatic transcending.)

2.3.2. The TM-Sidhi Program

The TM-Sidhi program was developed by Maharishi Mahesh Yogi from the Yoga Sutras of Patanjali. It is learned after many months of TM practice. While TM practice is a process of transcending, the TM-Sidhi practice involves the procedure of sanyama—the simultaneous processes of dhārānā (fixity on a sutra or phrase),dhyāna (transcending on that sutra) and samādhi (pure consciousness) (Maharishi Mahesh Yogi, 1978). Patanjali, the author of the Yoga Sutras predicted effects of practicing sanyama on different sutras or phrases. The TM-Sidhi practice includes a subset of these sutras. TM-Sidhi program is intended to connect aspects of the individual personality, such as senses, intuition, and emotions, with pure consciousness.

2.4. Data Selection

The first 30-sec artifact-free periods were selected within the first minute of the TM and the TM-Sidhi sessions. Previous research reports that brain patterns in the first minute of TM practice are similar to those in the middle and end of the TM session (Travis & Wallace, 1999). Thus, brain patterns in the beginning of these sessions should be representative of brain patterns during each practices.

2.5 Data Analyses

The data were analyzed with Brain Vision Analyzer. The 30-sec artifact-free data were re-referenced to averaged left and right ears, to compare with previous TM research, digitally filtered in a 2.0 - 50 Hz band pass filter with a 48 dB roll off, and fast Fourier transformed in 2-s epochs, using a Hanning window with 10% onset and offset. EEG amplitude was calculated from 2.0 - 50 Hz at the 32 recording sites. Coherence, the absolute value of the cross-correlation function in the frequency domain, was calculated for the 496 possible combination pairs of 32 recording sites.

2.5.1 Amplitude analysis

Brain Vision Analyzer can output either peak-to-peak amplitude (sqrt(real2 + imaginary2)) or power (real2 + imaginary2) values from the FFT. This paper reports amplitude values, since they are more normally distributed—skewness was between 1 and -1for amplitudes in all frequency bands. Amplitude estimates were grouped into seven frontal (AF3, F3, FC1, Fz, AF4, F4, FC2) and seven parietal (PO3, P3, CP1, Pz, P4, CP2, PO4) spatial averages and averaged into six frequency bands—theta2 (5-7.5 Hz), alpha1 (8-10 Hz), alpha2 (10.5-12.5 Hz), beta1 (16-20 Hz), beta2 (20.5-30 Hz), and gamma bands (30.5-50 Hz). Changes in frontal and parietal power have been most often reported in the literature during TM practice.

2.5.2 Coherence analysis

Coherence estimates were averaged into two spatial averages: seven frontal coherence pairs (AF3-AF4, F3-F4, FC1-FC2,AF3-F3,AF4-F4,AF4-FC2,AF3-FC1), and six anterior-posterior coherence pairs (AF3-P3, AF3-PO3, F3-PO3, AF4-P4, AF4-PO4, F4-PO4) in the same six frequency bands. Changes in frontal and anterior-posterior coherence have been most often reported in the literature during TM practice.

2.5.3 eLORETA

Exact low resolution electrotomography (eLORETA) was used to explore cortical sources of surface EEG. The 30-sec artifact-free periods were exported in ASCII format for eLORETA analysis. The steps of eLORETA analysis include: 1) computing EEG cross-spectra from the raw recordings using 2-sec windows with the same six frequency bands used in the spectral analysis; 2) computing cortical generators of surface oscillatory activity from the cross-spectra; and 3) calculating t-test differences between conditions foreach cortical voxel normalized by voxel averages. Normalizing by voxel in eLORETA analysis is similar to relative power in spectral analysis.

2.6. Statistical Analysis

Repeated-measure MANOVAs were used to test group differences in amplitude and coherence. In both analyses, there were three within factors: pre/post, spatial average (frontal/parietal) and frequency band (theta2 through gamma) in the amplitude analysis; and pre/post, spatial average (frontal/anterior-posterior) and frequency band (theta2 through gamma) in the coherence analysis. Power statistics are reported as partial eta squared (η2 ), the power statistic derived from F-tests using SPSS. Partial eta squared is the variance accounted for, similar to r2.

The SPM statistical software in eLORETA was used to conduct repeated-measure MANOVAs of voxel differences between groups. A voxel in eLORETA was considered significantly different if it and its six nearest neighbors (top, bottom, sides, front and back) differed significantly at the p < .0003 (two-tailed alpha level) between the two conditions. The eLORETA output program specifies the Brodmann areas (BA) identified with significant activation levels.

3.0 Results

3.1 Baseline Comparisons

Two separate MANOVAs were used to assess possible group differences in amplitude and coherence estimates during the initial 10-min TM sessions. These analyses yielded no significant omnibus groups differences, or differences within any specific frequency band during the initial TM practice in the two groups (Power: Wilks’ Lambda F(12, 13) = 1.4, ns; Coherence: Wilks’ Lambda F(17,8) = 1.3, ns). There were also no significant differences in sources of EEG in the eLORETA analysis of the pretest TM sessions.

3.2 Comparison of Peak-to-Peak Amplitude Estimates during TM and TM-Sidhi Practice

The repeated-measure MANOVA of frontal and parietal amplitude-averages yielded significant three-way prepost x frequency x spatial-average interactions (Wilks’ Lambda F(5,20) = 4.38, p = .007). Thus, two repeated-measure MANOVAs were conducted—one with frontal amplitude-averages in the six frequency bands as variates, and the other with parietal amplitudes-averages as variates. Significant three-way pre/post x frequency x group interactions were found in the analysis of frontal amplitudes (Greenhouse-Geisser F(2,50) = 4.1, p = .021), but none in parietal amplitudes. (all interactions F(2,50) < 1.0). Analysis within individual frequency bands yielded significant prepost x group interactions in frontalalpha1 amplitude (Wilks’ Lambda F(1,24) = 5.4, p = .029, η2 = .18)and in frontal beta1 amplitude (Wilks’ Lambda F(1,24) = 4.46, p = .045, η2 = .16).

Figure 1 presents the amplitude spectra averaged in seven frontal sensors during the second TM practice (dotted line) in the comparison group and practice of the TM-Sidhi program (solid line) in the experimental group. The frontal averages are presented since they were significant differences between groups in frontal averages. Notice higher amplitude in the 7.5-10 Hz band and in the16-20Hz band during TM-Sidhi practice.

3.3 Comparison of Coherence Estimates

The omnibus F-test of coherenceyieldedsignificant frequency x spatial-average interactions(Wilks’ Lambda F(5,20) =22.8, p .0001). Therefore separate MANOVAs were conducted within frontal and anterior/posterior averages. Neither of these analyses yielded significant prepost x group interactions.

3.4 Comparison of eLORETA images

eLORETA analysis yielded significantly stronger sources of alpha1 EEG for TM-Sidhi practice compared to TM practice in the right and left parahippocampal gyri (BA 19), the lingual gyrus (BA 18), the right fusiform gyrus (BA 37), and right inferior and medial temporal cortices (BA 20,21). These are part of right-hemisphere object-recognition areas. There were no significant sources that were stronger during TM compared to TM-Sidhi practice and no significant sources in other frequency bands. eLORETA sources in the alpha1 band from z= -45 to z = 45 are presented in Figure 2. A white area indicates a significantly stronger generator of surface alpha1 EEG during TM-Sidhi practice compared to TM.

3.5 Post Hoc Analysis

Amplitude analysis was only done for frontal and parietal averages, since power in these brain areas have been reported to be different during TM practice compared to other conditions. Since strong sources in temporal cortices were identified in the eLORETA analysis, a post hoc analysis was conducted with temporal amplitudes to investigate possible differences in temporal amplitude during TM-Sidhi practice compared to TM practice. EEG was averaged in left and right temporal/occipital leads including T7, CP5, P8, PO3, O1 and the homologous sensors on the right side in the same six frequency bands used in other analyses. MANOVAs revealed no significant initial group differences or repeated measure group differences in either left or right temporal amplitude averages in any frequency band (all Wilks' Lambda F(12,13) < 1.0).

4.0 Discussion

The data partially supported the research hypotheses. During TM-Sidhi practice, frontal alpha1and beta1 amplitudes were higher, but coherence was not. In addition, strong sources of alpha1 EEG were identified in the eLORETA analysis in right-hemisphere object-recognition areas—parahippocampal, fusiform, and lingual gyri, and in inferior and medial temporal cortices.

EEG changes in frontal alpha1 power and coherence have been most often reported in the literature as distinguishing eyes-closed rest and TM practice (Travis et al., 2010; Travis & Wallace, 1999). Alpha activity is usually interpreted as cortical idling (Pfurtscheller, Stancak, & Neuper, 1996)—higher alpha power in sensory and motor cortices inversely correlates with lower cerebral metabolic rate (Oakes et al., 2004). However, higher alpha in frontal cortices has been reported during tasks involving internal focus. This is called paradoxical alpha, and is seen during internally directed attentional tasks such as imagining a tune compared to listening to the same tune (Cooper, Burgess, Croft, & Gruzelier, 2006). Higher frontal alpha1 power during TM-Sidhi practice may be an instance of paradoxical alpha, reflecting internally directed attention to changing values of the sutra. Cortical activity in the beta1 band is associated with cortico-thalamic feedback loops modulating attention (Wrobel, 2000). Higher frontal alpha1 and beta1 activity during TM-Sidhi practice, compared to TM, supports the description of sanyama as involving internally directed attention in contrast to the process of effortless transcending during TM practice.