Relaxed Alertness in Novice and Advanced Meditators – A Neurophysiological and Psychological Study of Isha Yoga Practices

The study investigates how Isha Yoga practices influence neurophysiological brain oscillations and psychological outcomes in practitioners with differing expertise (advanced, novice) versus meditation-naïve controls. Motivated by extensive evidence that meditation supports relaxed alertness and benefits attention, emotion regulation, and well-being, but noting limited EEG research on Yogic practices (especially pranayama and deconstructive non-dual practices like Shoonya), the authors examined EEG spectral power changes during sukha kriya pranayama, breath-watching, and Shoonya meditation. They hypothesized differential oscillatory dynamics across practices, greater power increases during attentional practices (pranayama, breath-watching), trait increases (e.g., frontal midline theta) at rest in advanced meditators, state relaxed alertness (increased low- and high-frequency power) during pranayama and breath-watching in meditators but not controls, and better psychometric outcomes (lower stress and distress; higher well-being and non-attachment) and deeper states of consciousness (non-duality) in advanced meditators.

Prior EEG studies link meditation to increased alpha, theta, and gamma power indicative of relaxed alertness. Low-frequency oscillations (delta, theta, alpha) reflect stability/relaxation, while high-frequency oscillations (beta, gamma) relate to alertness and perceptual clarity. Research has focused more on attentional and constructive practices than deconstructive non-dual practices. Pranayama is a preparatory yogic breath practice associated with rapid stress reductions and benefits for cardiopulmonary health, brain function, and mental health, yet EEG studies on pranayama are relatively few. Isha Yoga comprises practices such as Shambhavi Mahamudra Kriya, Shoonya, and Samyama; prior studies report cardiovascular, neurophysiological, immune, neurological, and mental-health benefits in Isha practitioners. Shoonya is a non-dual awareness practice; non-duality involves a background awareness prior to conceptualization, with attenuated self–other boundaries, and has been reported across traditions. The phenomenological matrix model of meditation (object orientation, dereification, meta-awareness and secondary features) offers a framework to integrate phenomenology and neural data.

Design: Cross-sectional study including three groups: advanced Isha meditators (ADV, n = 42), novice Isha meditators (NOV, n = 33), and meditation-naïve controls (CNT, n = 28).

Participants: Recruited in Karnataka, India (meditators via Isha Foundation networks; controls via local community). ADV had completed at least one Samyama retreat; NOV had not. Controls had no prior yoga/meditation. ADV mean age 35.57 (SD 6.81) years; mean lifetime Isha practice 5507.80 (SD 2897.44) hours. NOV mean age 31.66 (SD 7.64) years; mean lifetime practice 1637.24 (SD 1126.61) hours. CNT mean age 31.14 (SD 6.38) years. Groups matched on age, education, marital status, socioeconomic status. Inclusion/exclusion criteria in Supplementary Tables S1–S2. Ethics approval (NIMH/DO/ETHICS SUB-COMMITTEE MEETING/2018); written informed consent obtained; no monetary compensation.

Procedure: EEG recorded continuously across four sessions: sukha kriya pranayama (alternate nostril breathing), breath-watching, a cognitive task (visual oddball; ERPs not reported here), and Shoonya meditation. Each session had 4-min rest blocks (alternating 1-min eyes-closed/open). A washout after Rest2 preceded the cognitive task; participants completed the Meditation Depth Questionnaire (MEDEQ) during washout and again after Shoonya. Both meditators and controls performed pranayama, breath-watching, and Shoonya-like instructions; meditators used their traditional approach; order not counterbalanced due to practice sequencing requirements.

EEG acquisition and preprocessing: Geodesic EEG System 300 (128-channel HydroCel nets), dimly lit, sound-attenuated room, 25°C, 40–60% humidity. Online reference near Cz; ground near AFz; NetAmps300 amplifier; 24-bit; 1 kHz sampling. Data imported into EEGLAB (MATLAB R2021b), resampled to 250 Hz, average referenced, high-pass >0.5 Hz, low-pass <80 Hz, 50 Hz notch. Artifact Subspace Reconstruction (ASR) to remove bad channels/segments (threshold 5 SD). Mean % data removed: ADV 11.19 (1.33–45.16), NOV 12.61 (1.3–66.7), CNT 11.21 (0.38–57.17). ICA (infomax); ICLabel (90% threshold) to remove muscle, ocular, ECG artifacts. Mean ICA components removed: ADV 3.14 (0–21), NOV 2.41 (0–21), CNT 3.99 (0–25). Spherical spline interpolation of channels.

Measures:

• Perceived Stress Scale (PSS-10); ω = 0.93; α = 0.78.
• PHQ-4 (anxiety/depression): ω total 0.85; α total 0.65; ω anxiety 0.76; ω depression 0.55.
• WHO-5 Well-Being Index: α 0.83; ω 0.88.
• Non-attachment Scale (NAS-7): ω 0.87; α 0.77.
• Meditation Depth Questionnaire (MEDEQ, 30 items; clusters: hindrances, relaxation, concentration, essential qualities, non-duality); α 0.96; ω 0.99; subscales ω, α > 0.8.
• Stanford Sleepiness Scale (SSS): categorized into Alert vs Not Alert.

EEG spectral analysis: Power spectral density (2 s epochs; Hamming window; 0.5 Hz resolution) for delta (1–4 Hz), theta (4–8), alpha (8–12), beta (13–30), gamma (30–80) during rest and practices. Statistics via LIMO toolbox: robust t with 2000 permutations; multiple comparisons across electrodes controlled using threshold-free cluster enhancement (TFCE) and permutation-derived FWER. Alpha < 0.05.

Psychometrics/statistics: RStudio; nonparametric Kruskal–Wallis with Dunn post-hoc when assumptions unmet; Welch’s ANOVA with Games–Howell post-hoc otherwise; Holm adjustment for multiple comparisons; effect sizes with 95% CIs; Pearson correlations. Significance p < 0.05.

Resting state (trait effects): Advanced meditators vs controls showed higher frontal midline theta power and lower centro-occipital gamma; advanced vs novice showed globally heightened theta and alpha and increased fronto-midline/central beta. Novice vs controls showed lower alpha (left posterior) and localized lower centro-occipital gamma. Baseline differences were more pronounced between advanced and novice than between advanced and controls.

Pranayama (state effects, within-group): Controls increased anterior delta and frontal midline theta; no major changes in other bands. Meditators (both NOV and ADV) showed increases across all bands; NOV changes were global, ADV more localized. Between-group: NOV > CNT in posterior theta, frontal/central beta, and frontal/central/occipital gamma; ADV > CNT in limited frontal/occipital gamma; little difference between ADV and NOV (trend toward higher NOV state change).

Breath-watching (within-group): Controls showed increased anterior delta/theta, decreased occipital alpha, and no major beta/gamma changes. NOV and ADV showed global increases in delta, theta, and beta; little change in alpha/gamma. Between-group: no significant differences, with trends toward higher alpha/beta/gamma in meditators.

Shoonya (within-group): Controls showed global decreases across all bands; NOV showed global reductions, especially delta, beta, gamma; ADV showed no significant change from rest (sustained power). Post vs Pre Shoonya rest: ADV no differences; NOV localized reductions across bands; CNT widespread decreases.

Psychometrics (group differences; Kruskal–Wallis):

• Perceived stress: χ² = 12.39, p < 0.001, η²ordinal = 0.12; post-hoc ADV < CNT (pHolm < 0.001).
• WHO well-being: χ² = 20.33, p < 0.001, η²ordinal = 0.20; post-hoc ADV > CNT (pHolm < 0.001), ADV > NOV (pHolm < 0.05), NOV > CNT (pHolm < 0.05).
• Non-attachment: χ² = 18.42, p < 0.001, η²ordinal = 0.18; post-hoc ADV > CNT (pHolm < 0.001), NOV > CNT (pHolm < 0.001).
• PHQ-4 (mental distress): χ² = 18.09, p < 0.001, η²ordinal = 0.18; post-hoc ADV < CNT (pHolm < 0.001), NOV < CNT (pHolm < 0.05).

Correlations with lifetime Isha practice hours (n=103):

• Perceived stress: r = −0.36, t(101) = −3.84, p < 0.001.
• Mental distress (PHQ-4): r = −0.35, t(101) = −3.71, p < 0.001.
• WHO well-being: r = 0.36, t(101) = 3.85, p < 0.001.
• Non-attachment: r = 0.35, t(101) = 3.76, p < 0.001.

Meditation depth (MEDEQ): Breath-watching depth differed by group (Welch’s F(2,44.17) = 20.59, p < 0.001, ω² = 0.45); post-hoc ADV > CNT (pHolm < 0.001), ADV > NOV (pHolm < 0.05), NOV > CNT (pHolm < 0.001); correlation with lifetime hours r = 0.47, t(68) = 4.37, p < 0.001. Shoonya depth: Welch’s F(2,39.75) = 26.98, p < 0.001, ω² = 0.55; post-hoc ADV > CNT (pHolm < 0.001), ADV > NOV (pHolm < 0.05), NOV > CNT (pHolm < 0.001); correlation with lifetime hours r = 0.51, t(64) = 4.70, p < 0.001. Subscales showed ADV had lower hindrances and higher relaxation, concentration, essential qualities, and non-duality; NOV exceeded CNT on several subscales.

Stanford Sleepiness Scale: Before breath-watching, alert: ADV 95%, NOV 84%, CNT 80%; after breath-watching: ADV 78%, NOV 72%, CNT 52%. Before Shoonya: ADV 81%, NOV 90%, CNT 92%; after Shoonya: ADV 97%, NOV 93%, CNT 62%.

Findings support the hypotheses that Isha Yoga practices produce distinct trait and state neurophysiological signatures and are associated with better psychological outcomes. Advanced meditators showed trait increases in frontal midline theta, alpha, and beta at rest, consistent with relaxed alertness and sustained internalized attention. During attentional practices (pranayama, breath-watching), meditators exhibited increases across low- and high-frequency bands indicative of simultaneous relaxation and alertness; controls primarily showed low-frequency increases and signs of drowsiness during breath-watching (reduced occipital alpha). In the deconstructive practice Shoonya, advanced meditators maintained stable oscillatory power across conditions (suggesting trait stabilization), whereas novices and controls exhibited global power reductions and increased hindrances/drowsiness. Converging self-report evidence (MEDEQ depth and non-duality, SSS alertness) aligns with EEG markers, indicating clearer, more stable, less effortful meditative states in experienced practitioners. Psychometric advantages (lower stress and distress; higher well-being and non-attachment) and their correlations with lifetime practice suggest dose–response relationships between Isha Yoga engagement and mental health benefits. Mapping results onto the phenomenological matrix highlights higher dereification, meta-awareness, clarity, and stability among advanced practitioners, especially during breath-watching and Shoonya.

Isha Yoga practices elicit distinct neurophysiological states and traits. Both novice and advanced meditators demonstrated relaxed alertness during pranayama and breath-watching, while advanced meditators exhibited trait-like stability and deeper non-dual phenomenology during Shoonya. Meditation proficiency was associated with superior mental health outcomes and greater meditation depth. These results advance understanding of yogic mind–body practices, particularly pranayama and non-dual Shoonya, and integrate neural and phenomenological perspectives. Future research should use longitudinal designs, control for sequencing effects, incorporate neurophenomenological and experiential sampling methods, and disentangle contextual moderators (e.g., personality, motivation) to refine mechanistic accounts and generalizability.

Cross-sectional design precludes causal inference. Session order could not be counterbalanced due to traditional sequencing, potentially introducing order effects. Session length may have induced fatigue, especially in controls. Unmeasured contextual factors (e.g., personality, motivation) may have influenced outcomes. No neurophenomenological/experience-sampling measures were collected to directly link subjective states with neural dynamics.