Progressive muscle relaxation

Progressive muscle relaxation (PMR) is a somatic relaxation practice developed by Edmund Jacobson in the early 20th century, with its clinical standard established by Bernstein and Borkovec in 1973. It is among the most extensively studied somatic interventions, carrying a Grade A recommendation for migraine prophylaxis.^[1] It alters consciousness through interoceptive contrast — systematic tension and release of muscle groups that downregulates sympathetic and HPA-axis arousal.

The practitioner sits or reclines and works through a fixed sequence of skeletal muscle groups, tensing each for five to seven seconds before releasing for roughly ten seconds. The central instruction is to attend to the contrast between tension and release as the muscle softens, learning to recognize and let go of residual tension. Sessions range from ten minutes in abbreviated practice to thirty to forty minutes in the full sixteen-group protocol.

A typical session lasts ten to forty minutes depending on the protocol used. At introductory levels, practitioners report heaviness and warmth in the limbs, slowing of breath, mild drowsiness, and a noticeable softening of habitual postural tension. At established levels, release-only practice produces rapid drops into a low-arousal state that can be deployed in stressful situations within seconds. Restlessness, intrusive thoughts, transient muscle cramping during the tension phase, and paradoxical anxiety during release are common.

The primary documented adverse effect is relaxation-induced anxiety — a paradoxical spike in arousal during release that appears in roughly a third of participants in early studies and is more pronounced in generalized anxiety and depression.^[2] Active musculoskeletal injuries, acute back pain, and post-surgical recovery are standard contraindications to the tension phase. PMR is among the most extensively studied somatic interventions, with trials supporting use in sleep, anxiety, hypertension, and migraine prophylaxis.^[3]

Origins

Edmund Jacobson (1888–1983), an American physician and physiologist trained at Harvard, Cornell, and Rush Medical College, developed progressive muscle relaxation from observations he first presented at Harvard in 1908. He began treating patients with tension-related disorders by 1915. His central thesis was that chronic skeletal muscle tension co-occurs with — and amplifies — anxiety and stress, and that systematically reducing muscle tension produces measurable decreases in autonomic arousal.

Jacobson's foundational publications span three decades. "The Technic of Progressive Relaxation" (1924) offered the first formal description of the didactic method. The 1929 monograph Progressive Relaxation provided a full experimental account with electromyographic data.^[4] You Must Relax (1934) made the method accessible to lay readers. Jacobson's use of early EMG equipment to quantify muscle tension gave PMR an empirical grounding unusual for its era.

The original protocol was demanding. Training could extend across fifty or more sessions over several months, working through muscle groups with methodical attention to sub-threshold tension. The format was clinically rigorous but impractical for broad adoption.

Clinical adaptation

Douglas A. Bernstein and Thomas D. Borkovec standardized the method for clinical dissemination in their 1973 manual Progressive Relaxation Training: A Manual for the Helping Professions.^[5] Their protocol covered sixteen muscle groups with a five-to-seven second tension phase followed by a ten-second release phase, completing a session in roughly thirty to forty minutes. This version became the clinical and research standard.

Subsequent abbreviations — to seven muscle groups, then four — allowed delivery in outpatient cognitive-behavioural therapy and group settings. Lars-Göran Öst extended the progressive abbreviation further in 1987 with "Applied Relaxation," designed to reach full relaxation in twenty to thirty seconds for in-situation use during phobia and generalized anxiety treatment.

Contemporary context

From the 1970s onward, PMR was integrated into behavioural medicine, pain management, oncology supportive care, cardiac rehabilitation, and sleep medicine. It appears as a component within CBT protocols, in MBSR-adjacent programs, and as a standalone intervention manual. Transmission has proceeded through clinical training and published manuals rather than lineage or apprenticeship. The American Psychological Association's Division 12 lists relaxation training, including PMR, as an empirically supported treatment for headache and chronic pain. No central credentialing body governs delivery; nurses, physiotherapists, occupational therapists, and clinical psychologists are the most common providers.

Technique

The practitioner sits in a supportive chair with the spine vertical or reclines on a flat surface with head and knees gently supported. Loose clothing and a quiet room are standard. Eyes may be closed or left softly open. The body needs to be positioned so that each muscle group can be tensed and released without strain to neighbouring areas.

The method is a fixed sequence of skeletal muscle contractions and releases. The practitioner brings attention to a single muscle group — for example the dominant forearm — tenses it to roughly two-thirds of maximum strength for five to seven seconds, then releases the contraction completely for approximately ten seconds while attending to the resulting sensation of release.^[5] The central principle is contrast: the tension phase makes the release perceptible, and repeated cycles train the practitioner to recognize and let go of residual tension that would otherwise go unnoticed. The Bernstein and Borkovec sequence covers sixteen groups, moving from hands and forearms through biceps, face, neck and throat, chest and shoulders, abdomen, thighs, calves, and feet. The most common departure from the method is contracting to maximum effort or holding tension to the point of cramp — both of which obscure the release contrast and increase the risk of musculoskeletal complaint.

Progression is protocol-driven rather than experiential. At the introductory level, the practitioner works through the full sixteen-group sequence in roughly thirty to forty minutes, ideally daily during the acquisition phase. Once the tension-release contrast is internalized, the sequence shortens — to seven groups, then four — and the tension phase can be omitted entirely, leaving a release-only practice of fifteen to twenty minutes. At the established level, Applied Relaxation condenses the response to twenty to thirty seconds for use in stressful situations; this skill typically requires eight to twelve weeks of prior practice.

Two clinical variants share the underlying tension-release structure. Bernstein and Borkovec's protocol is the dominant research and outpatient standard. Öst's Applied Relaxation builds on it as a phased curriculum culminating in in-situation deployment for anxiety and phobic disorders. Self-guided audio versions distributed by health systems and CBT publishers are adapted directly from the Bernstein and Borkovec script and represent the most common form of contemporary lay delivery.

Setting & Preparation

PMR has minimal physical requirements. A quiet room with low light, a supportive chair or recliner that allows the head and limbs to be loosened, and roughly thirty minutes of uninterrupted time are sufficient for the full protocol. Loose clothing is preferable. The practitioner does not need any equipment, though a recorded script — widely available through clinical and health-system sources — is commonly used during the acquisition phase. No specialist supervision is required, and the practice is regularly delivered as self-guided audio. A facilitator is helpful when PMR is taught within a clinical course of treatment.

A quiet room with low ambient noise, comfortable temperature, and a supportive chair or flat surface is adequate. Loose clothing helps. The session is best scheduled when the practitioner is not acutely fatigued and has not eaten a large meal in the prior hour. A recorded script during the first weeks of practice helps anchor the sequence and pacing; once the protocol is memorized, the recording is no longer needed. Practising at the same time each day shortens the acquisition phase.

PMR engages two interacting systems: an interoceptive system that builds discrimination between tension and relaxation, and an autonomic system that downregulates sympathetic and HPA-axis activity during the release phase.

Interoceptive contrast

The tension–release cycle is a deliberate generator of interoceptive contrast. By bringing a muscle group to a defined level of contraction and then releasing it completely, the practitioner produces a sharp difference in afferent signal that is easier to detect than the slow drift of resting tension. Repeated cycles train discrimination of sub-threshold tension that would otherwise go unnoticed. An fMRI investigation (n = 11) found that PMR progressively decreased BOLD activity in the superior frontal gyrus, inferior frontal gyrus, and posterior cingulate cortex, with session-by-time interactions in the putamen, anterior cingulate cortex, postcentral gyrus, and insula.^[6] The insular and postcentral involvement is consistent with engagement of the interoceptive cortex during the practice.

Autonomic and HPA downregulation

Reductions in afferent signalling from relaxed muscles correspond to measurable shifts in autonomic state. Multiple studies have documented decreased salivary cortisol following PMR sessions, consistent with downregulation of hypothalamic-pituitary-adrenal axis activity [UNCITED — verify: salivary cortisol reductions following PMR sessions]. High-frequency heart rate variability (a vagal-tone index) increases and the LF/HF ratio decreases following practice in high-stress samples, indicating a shift toward parasympathetic dominance [UNCITED — verify: HRV shifts following PMR practice in high-stress samples]. EEG studies in chemotherapy populations report increased theta-band activity in posterior scalp regions and decreased frontal beta, a pattern broadly associated with reduced cortical arousal [UNCITED — verify: EEG theta increase and beta decrease following PMR in chemotherapy populations]. A controlled study using polysomnography found that PMR before a daytime nap increased N3 slow-wave sleep, providing objective evidence of an effect on sleep architecture rather than only on subjective quality or sleep latency.^[7]

Jacobson's original theoretical claim that muscle relaxation decreases hypothalamic discharge has been partially supported by neuroendocrine measurement, although direct hypothalamic imaging evidence in humans is absent.

PMR produces a coherent low-arousal state characterized by physical heaviness, slowed respiration, reduced postural tension, and a quieted internal narrative. The practice is somatic rather than perceptual: there are no characteristic visual or auditory alterations, and the cognitive content of the experience remains close to ordinary waking awareness. Practitioners commonly report increased body awareness and a clearer sense of where habitual tension is held.

Effects scale with the protocol used rather than with cumulative practice. Introductory work through the full sixteen-group sequence produces the most pronounced acute relaxation in newcomers, with reports of warmth in the limbs, mild drowsiness, and softening of jaw, shoulder, and forehead tension. Established practitioners using release-only or applied formats access a shallower but more rapidly reached relaxation that can be deployed in stressful situations. Across populations and protocols, controlled trials document moderate-to-large reductions in self-reported anxiety and improvements in sleep quality.^[8][3]

Unsought effects are routine. Restlessness and intrusive thinking during long sessions, transient muscle cramping during the tension phase, and a paradoxical anxiety surge during release are all documented. Trauma-history populations may find inwardly directed somatic attention activating rather than calming, paralleling concerns documented in the meditation literature.

Adverse Effects

No large-scale epidemiological survey of adverse effects specific to PMR exists. The adverse-event profile is thinner than the meditation literature, which likely reflects the routine clinical context in which PMR is delivered rather than the absence of risk.

Contraindications

Substance Interactions

No published case reports, prospective studies, or controlled trials document interactions between PMR and psychoactive substances. PMR is a non-pharmacological intervention and the literature does not address combined-use risks.

Two considerations are inferential rather than evidence-based. PMR consistently down-regulates the sympathetic branch and raises parasympathetic tone, a direction of action similar to that of benzodiazepines, sedative-hypnotics, and alcohol; combined use with central nervous system depressants could theoretically deepen sedation or potentiate hypotensive effects. Acute use with stimulants is unlikely to produce harm but may reduce the practice's effectiveness, as sympathetic activation works against the relaxation response. Neither consideration has been tested clinically.

Habituation

PMR has no documented habituation profile, no evidence of compulsive use, and no withdrawal pattern. The practice does not produce a reward state strong enough to drive repetition for its own sake. Daily practice during the acquisition phase is the recommended pattern; later, practitioners typically reduce frequency once the response can be deployed reliably in situations of need. There is no clinical literature describing dependency on PMR as a problem requiring intervention.

If Distress Arises

If a muscle cramps during the tension phase, release immediately and skip the rest of that cycle for the affected group. If localized pain persists across multiple sessions, stop tensing that group and use release-only attention there instead.

If anxiety or panic-like arousal spikes during the release phase — the classic relaxation-induced anxiety presentation — do not force the practice to continue. Open the eyes, sit up, and orient to the room. Naming three nearby objects, attending to the contact of the feet with the floor, or stepping outside for fresh air all serve as grounding. RIA is a recognized response pattern, not a sign of doing the practice incorrectly; subsequent sessions can be shortened and the tension phase softened.

If intrusive thoughts or distressing memories surface, the safest response is to stop the session and return to ordinary activity. Persistent surfacing of distressing material warrants consultation with a mental health professional before continuing intensive practice.

PMR does not have a formal stage model. Progression is practical: the practitioner moves from a long, fully embodied sequence to shorter and more readily deployable versions as the tension-release contrast is internalized.

The introductory phase covers the full Bernstein and Borkovec sequence of sixteen groups over roughly thirty to forty minutes, practised daily during acquisition.^[5] Two sessions per day during the first week are commonly recommended for skill consolidation. Once the tension-release contrast is internalized, release-only practice — focusing attention on softening without preceding contraction — can be used with the sequence abbreviated to seven or four groups. At the established level, Öst's Applied Relaxation condenses the response further: practitioners learn to access a recognizable relaxation state within twenty to thirty seconds for use in real-world situations, typically after eight to twelve weeks of prior practice. No teacher validation or certification marks transitions between levels; progression rests on self-report of proficiency and clinical judgement.

PMR has one of the broader controlled-trial bases among somatic relaxation practices. Anchor evidence spans sleep, anxiety, depression, hypertension, cancer-related distress, and migraine prophylaxis.

A 2024 systematic review searched five databases through March 2023 and included 46 studies from 16 countries covering more than 3,402 adults, finding consistent reductions in stress, anxiety, and depression and additive effects when PMR was combined with other interventions.^[3]

A 2026 systematic review and meta-analysis of 31 RCTs (n = 2,277) reported that PMR significantly improved sleep quality (SMD −1.74; 95% CI −2.14 to −1.34), reduced anxiety (SMD −1.11; 95% CI −1.69 to −0.53), and improved quality of life (SMD 1.32; 95% CI 0.53 to 2.12); sensitivity analyses confirmed consistency despite high heterogeneity.^[8]

In cancer populations, a 2022 review found significant effects on anxiety, pain, and quality of life but not on fatigue;^[9] a later review found potential benefit on cancer-related fatigue with evidence quality rated moderate to very low.^[10] A pooled analysis of four trials (n = 227) in mild-to-moderate COVID-19 inpatients found a large reduction in anxiety (SMD −1.35; 95% CI −2.38 to −0.32).^[11]

A 2025 network meta-analysis of 182 RCTs for hypertension and prehypertension found PMR reduced systolic blood pressure by approximately 7.46 mmHg relative to passive comparators at short-term follow-up of three months or less.^[12] A quasi-randomized trial in migraine (n = 16) found six weeks of PMR significantly reduced monthly migraine frequency and normalized contingent negative variation amplitudes, a cortical electrophysiological marker of migraine susceptibility;^[1] the US Headache Consortium assigns PMR a Grade A recommendation for migraine prophylaxis, the same grade as beta-blockers.

Limitations

Heterogeneity is consistently described as high across PMR meta-analyses, reflecting diverse populations, delivery formats, protocol durations, and outcome instruments. Sensitivity analyses in the largest sleep and mental health review confirmed direction-of-effect consistency despite this heterogeneity,^[8] but headline effect sizes should be read as ranges rather than point estimates.

Blinding is structurally constrained: PMR cannot be sham-controlled in the way a pharmacological agent can, so expectancy and demand effects cannot be cleanly partitioned from active mechanism. Comparators range from waiting list and usual care to other behavioural interventions, and meta-analytic pooling across these comparators inflates apparent effect sizes against weaker controls.

Durability is poorly characterized. The hypertension network meta-analysis found that blood-pressure effects attenuated at longer follow-up, with few trials extending beyond three months.^[12] Similar follow-up gaps exist in the sleep and anxiety literature. No systematic adverse-event survey of the scale of the Britton meditation-adverse-experiences literature exists for PMR, and the adverse-event profile is therefore weaker than direct comparison with meditation outcomes would suggest. Trauma-history populations, individuals with active arrhythmias, and pediatric samples are underrepresented across the trial base.