The Impact of Twin Hearts Meditation on Mood, Cognitive Functioning, and EEG Dynamics

Eighteen experienced Arhatic Yoga practitioners and 18 novice meditators were recruited for this study. Experienced meditators were recruited from a weeklong Arhatic Yoga Spiritual Retreat (July 20-28, 2024) in Bovina, NY. Inclusion to the experienced meditator sample included a minimum of 1,000 hours practice of THM as well as a current meditation practice. Inexperienced meditators were recruited in Eugene, OR, via social media and community flyers. Inclusion criteria for this sample included less than 10 hours of lifetime meditation practice. Compensation for the inexperienced group was a $30 check or paypal credit.

Screening of inclusion and exclusion criteria was completed via digital forms. Those deemed eligible completed the informed consent process, and provided more demographic information for sample characteristics.

Experimental Design

The experiment was an unblinded, control group comparison (2 x 2 factorial design) of experienced versus novice meditators groups with multimodal measurements (subjective mood questionnaire, cognitive tasks, neural/ EEG) taken before and after completion of a Twin Hearts Meditation practice. Both experienced and meditation naïve subjects underwent a structured research protocol consisting of screening, a pre-meditation assessment, a guided Twin Hearts Meditation session, and a post-meditation assessment. Sessions lasted about 90 min. The procedure was as follows:

Screening and Participant Selection. Participants completed an online screening and demographic questionnaire. Those meeting inclusion and exclusion criteria were invited to participate in the research study. Eligible subjects who indicated interest were scheduled for a laboratory-based session. Informed consent procedures were provided upon arrival to the lab space ahead of conducting the remaining protocol procedures.

Pre-Meditation Assessment. After obtaining informed consent, subjects completed a series of neural, mental and cognitive baseline measures. Psychological questionnaires included the Multidimensional Psychological Flexibility Inventory abbreviated scale (MPFI - Abbreviated Flexibility), and the Tellegen Absorption Scale (to assess baseline mental traits). These two measures were only given once for assessing baseline trait differences between experimental meditator samples. However, the following measures were given before and after meditation:

Participants next completed the Brunel Mood Scale. Then subjects were prepared for EEG to measure neural activity in a 3-minute resting state recording, as well as during their completion of a P300 auditory oddball task. They then completed the cognitive-behavioural task set, that included the Flanker task (to evaluate response inhibition and attention control) and Trail Making Tests A (to assess processing speed, simple attention) \& B (complex attention/executive function).

Twin Hearts Meditation Session. Participants then listened to a pre-recorded 21-minute guided Twin Hearts Meditation session via headphones with eyes closed. Participants were instructed to simply follow along with the instructions provided in the meditation.

Post-Meditation Assessment. Following the meditation, subjects completed the post-meditation EEG baseline recording. Additional mental given: i ) Depth of Meditation Scale (to assess subjective meditation experience), ii) Brunel Mood Scale (to measure post-meditation mood state). This was followed by re-administration of the Cognitive-Behavioural measurement tasks (P300 auditory oddball, Flanker, Trail Making Test A \& B)

Description of Psychological Questionnaire-based Measurements

Demographic questionnaire. This survey asks basic questions regarding age, sex, history of mental health diagnoses, and meditation history. Data was used from this survey to characterize overall and Experienced versus Inexperienced meditator sample characteristics.

Tellegren Absorption Scale (TAS). The Tellegen Absorption Scale is a 34-item multi-dimensional measure that assesses imaginative involvement and the tendency to become mentally absorbed in everyday tasks. Each item on the scale requires a 'true' or 'false' response. Scores for the TAS and its subscales reflect the number of questions marked 'true,' with higher scores indicating higher absorption. Tellegen reported an internal reliability of .88 and a 30-day test-retest reliability of .91. Kihlstrom reported a test-retest reliability of .85.

The TAS is divided into five subscales: altered states of consciousness (TAS-ACS; "I sometimes 'step outside' my usual self and experience an entirely different state of being"), aesthetic involvement in nature (TAS-AIN; "I like to watch cloud shapes change in the sky"), imaginative involvement (TAS-II; "I am able to wander off into my thoughts while doing a routine task and actually forget that I am doing the task, and then find a few minutes later that I have completed it"), Synesthesia (TAS-SYN; "I find that different odors have different colors"), and extrasensory perception or ESP (TAS-ESP; "I can often somehow sense the presence of another person before I actually see her/him"). A high score on a specific subscale indicates a greater tendency to experience that particular type of absorption, with the TAS Total score reflecting an overall trait of responsive absorption to stimuli, tasks, and surroundings.

Psychological Flexibility Scale (PFS). The Psychological Flexibility Scale (PFS) is a shorter global composite of the Multidimensional Psychological Flexibility Inventory. The 12-item PFS is comprised of the first two items of each of the MPFI subscales that subjects respond to using on a six-point Likert scale to rate the accuracy of statement related to acceptance, present moment awareness, self as context, defusion, values, and committed action. Higher scores indicate higher mental flexibility comprised by the item pairs in each domain.

Depth of Meditation Experience. The Meditation Depth Questionnaire is a 30-item self-report measure designed to assess the depth of meditative experience across five progressive stages. These stages include: (1) Hindrances, which captures difficulties such as boredom, impatience, and trouble with concentration (reverse-coded for consistency); (2) Relaxation, reflecting feelings of inner peace and calmness; (3) Concentration, describing detachment from thoughts, deep insight, and mental clarity; (4) Essential Qualities, encompassing emotions like love, gratitude, and connectedness; and (5) Non-Duality, representing the dissolution of cognitive processes and a sense of unity. The questionnaire is given post-meditation to evaluate subjective depth of experience, with higher scores indicating greater meditative depth across individual subscales and overall. The Hindrance scale was reverse coded, to be in line with other scales and total score, so that higher values indicate more favorable experiences overall and according to subscales.

Brunel Mood Scale. The Brunel Mood Scale is a 32-item self-report measure designed to assess current mood states. It consists of eight unipolar mood dimensions: anger, tension, depression, vigor, fatigue, confusion, happiness, and calmness. Participants respond to the prompt, "How do you feel right now?" by rating each mood descriptor on a 5-point Likert scale (0 = not at all, 1 = a little, 2 = moderately, 3 = quite a bit, 4 = extremely), indicating the intensity of their momentary emotional experience. As well as producing subscale scores, an overall mood score, referred to as a Total Mood Disturbance (TMD) score, can be calculated by summing the scores for Tension, Depression, Anger, Fatigue and Confusion and then subtracting the Vigor score. Higher TMD scores indicate greater overall mood disturbance, reflecting higher levels of negative emotions inversely correlated with lower positive affect. Conversely, lower scores suggest better mood balance and emotional well-being.

Description of Cognitive-Behavioral Measures

Trails A \& B. Trails A \& B is a cognitive assessment measurement consisting of two tasks, Trail A and Trail B. Trail A measures visual processing speed and simple attention, requiring subjects to connect numbers in sequential order across a display sheet. For Trail B subjects must complete a task similar to Trail A, but instead alternate between numbers and letters intermixed across the display, in ascending order. Trail B is considered a measure of complex attention and working memory, and of executive function including inhibition and divided attention. Trails A \& B were given digitally using the WAVi Software Platform under the supervision of a researcher. Instructions were provided via computer and researcher. Performance metrics included the number o

Statistical Analysis Statistical analyses were conducted using SPSS. Prior to hypothesis testing, data were screened for missing values, outliers, normality, and homogeneity of variance assumptions. Variables were evaluated for normality evaluating for skewness and kurtosis values between +2/-2 and +7/-7 respectively. Levene's tests of homogeneity were also conducted, p < 0.05. A combination of ANOVA, t-testing, and correlational analyses were used for hypothesis testing. All statistical tests were two-tailed, with an alpha level set at .05. Effect sizes and 95% confidence intervals (±1.96 SEM) are reported where appropriate.

The primary approach for hypothesis testing involved running 2 (Experimental Group: Experienced, Inexperienced) X 2 (Time: Pre-meditation, Post-meditation) repeated-measures ANOVAs (rm-ANOVA) for the Brunel Mood Scale, Flanker performance measures, Trails A\&B performance measures, P300 and Resting state EEG measures of interest. The stepwise Holm-Bonferroni correction (Holm, 1979) was applied to control for family-wise error of questionnaire, cognitive/attentional task, and EEG variable sets initially investigated by ANOVA sets, and applied for follow-up on pairwise comparisons where significant interaction effects were found.

One-way ANOVA and independent t-tests were used to investigate demographic, baseline and single time point comparisons between groups (e.g., TAS, depth of meditation experience). Chi-Square testing was conducted for applicable frequency-based values of the demographic variables.

In addition, Pearson's correlation analyses were conducted in planned and ad hoc approaches. Planned correlations included [ ]. Unplanned/exploratory correlations included assessing the relationship of variables discovered to be statistically significant in ANOVA results, with other variables that might help further describe or inform interpretation of a result. Correlation coefficients were interpreted following Cohen's guidelines, where values of 0.10, 0.30, and 0.50 were considered small, medium, and large, respectively. To control for multiple comparisons, a Benjamini-Hochberg false discovery rate (FDR) correction was applied (I have not yet applied this and may not - need specified correlation sets before deciding)

Data Cleaning and Outlier Handling Outlier detection was performed using a combination of visual inspection (boxplots, Q-Q plots), and statistical thresholds (interquartile method). Extreme outliers were removed as single data points, or winzored with imputation. The decision for complete omission versus imputation was dependent on the type and amount of missing or omitted data. Imputation was selectively and sparingly used for resting state EEG, whereby unrealistic outlier values (suspected of being artefact, missed on built-in software filtering and cleaning) were removed, then re-estimated with regression imputation based on group-time point averages and proximal electrode site values.