𝐃𝐞𝐛𝐮𝐧𝐤𝐢𝐧𝐠 ‘𝐒𝐭𝐨𝐫𝐞𝐝 𝐓𝐫𝐚𝐮𝐦𝐚’ 𝐢𝐧 𝐅𝐚𝐬𝐜𝐢𝐚: 𝐀 𝐍𝐞𝐮𝐫𝐨𝐬𝐜𝐢𝐞𝐧𝐜𝐞‑𝐁𝐚𝐬𝐞𝐝 𝐄𝐱𝐩𝐥𝐚𝐧𝐚𝐭𝐨𝐫𝐲 𝐍𝐚𝐫𝐫𝐚𝐭𝐢𝐯𝐞 𝐑𝐞𝐯𝐢𝐞𝐰

𝐃𝐞𝐛𝐮𝐧𝐤𝐢𝐧𝐠 ‘𝐒𝐭𝐨𝐫𝐞𝐝 𝐓𝐫𝐚𝐮𝐦𝐚’ 𝐢𝐧 𝐅𝐚𝐬𝐜𝐢𝐚: 𝐀 𝐍𝐞𝐮𝐫𝐨𝐬𝐜𝐢𝐞𝐧𝐜𝐞‑𝐁𝐚𝐬𝐞𝐝 𝐄𝐱𝐩𝐥𝐚𝐧𝐚𝐭𝐨𝐫𝐲 𝐍𝐚𝐫𝐫𝐚𝐭𝐢𝐯𝐞 𝐑𝐞𝐯𝐢𝐞𝐰

𝐀𝐛𝐬𝐭𝐫𝐚𝐜𝐭

Massage produces reproducible short‑term autonomic and peripheral effects — modest increases in HF‑HRV and small reductions in heart rate and blood pressure — that are typically transient and heterogeneous across studies (Diego & Field, 2009; Roura et al., 2021). Although peripheral mechanotransduction can alter tissue biology and, under chronic loading, contribute to structural adaptation, most evidence for remodeling derives from in vitro, animal, or chronic‑loading contexts; typical therapeutic massage applies lower strain and shorter duration than conditioning paradigms used in mechanobiology studies. Persistent pain and trauma‑related symptoms are better explained by central nervous system plasticity, in which altered large‑scale networks interact with autonomic and endocrine responses to shape interoception and threat appraisal. Massage is best conceptualized as a bottom‑up modulator that influences interoception, autonomic tone, and nociceptive processing, creating short‑term conditions favorable for cognitive, behavioral, and exposure‑based therapies within multimodal care.

𝐊𝐞𝐲𝐰𝐨𝐫𝐝𝐬: massage therapy; autonomic nervous system; mechanotransduction; pain neuroscience; interoception; trauma

𝐈𝐧𝐭𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧

This narrative review evaluates the claim that trauma or autobiographical memories are literally stored in peripheral tissues such as fascia or muscle, and it tests that claim against current mechanobiology, autonomic physiology, interoception, and systems‑level neuroscience. Rather than adjudicating therapeutic efficacy alone, the paper asks a mechanistic question: do the anatomical structures and cellular processes in connective tissue provide a plausible substrate for episodic or affective memory encoding?

To answer this, I synthesize evidence across biological scales: cellular mechanotransduction and extracellular matrix adaptation; peripheral–spinal–brain pathways that mediate interoception and nociception; and large‑scale neural networks that implement episodic memory, salience tagging, and predictive processing. The goal is practical and translational: clarify what hands‑on therapies can reliably change (autonomic state, interoceptive precision, motor output) and what they cannot (representational episodic storage), and to offer clinicians precise language and evidence‑based guidance for trauma‑informed practice.

𝐌𝐞𝐭𝐡𝐨𝐝𝐬 𝐚𝐧𝐝 𝐒𝐜𝐨𝐩𝐞

This article is an explanatory narrative review aimed at evaluating the mechanistic plausibility of the claim that autobiographical trauma is “stored” in fascia. Sources were selected to illustrate mechanistic constraints and representative clinical findings across mechanobiology, pain neuroscience, interoception, autonomic physiology, predictive processing, and trauma neurobiology. Priority was given to systematic reviews, randomized controlled trials for clinical autonomic outcomes, and foundational mechanistic papers. Because mechanistic disproof relies on biological constraints (anatomy, cellular signaling, network computation) as well as clinical data, randomized trials are used primarily to characterize effect sizes for autonomic and symptomatic change rather than to establish mechanistic encoding. This review is not a systematic review or meta‑analysis and does not attempt exhaustive study identification; inclusion criteria emphasized peer‑reviewed mechanistic and clinical work relevant to the central question, and non‑peer‑reviewed case reports and non‑English sources were excluded. Search transparency: search terms included “mechanotransduction,” “fascia memory,” and “massage HRV”; searches were performed in PubMed and Google Scholar through December 2025.

𝐒𝐨𝐦𝐚𝐭𝐢𝐜 𝐌𝐞𝐦𝐨𝐫𝐲 𝐂𝐥𝐚𝐢𝐦𝐬 — 𝐇𝐢𝐬𝐭𝐨𝐫𝐢𝐜𝐚𝐥 𝐋𝐢𝐧𝐞𝐚𝐠𝐞 𝐚𝐧𝐝 𝐏𝐞𝐫𝐬𝐢𝐬𝐭𝐞𝐧𝐜𝐞

The claim that the body “stores” trauma has a clear intellectual lineage that includes somatic‑experiencing, Reichian body‑memory traditions, and contemporary fascia‑as‑storage metaphors. These traditions converge on the idea that peripheral tissues (muscle, fascia) can retain episodic or affective content that is later released by touch or manipulation. The claim persists because bodily sensations reliably accompany emotion, clients often report dramatic affective shifts during hands‑on care, and metaphors about “holding” or “releasing” tension are easy to communicate in clinical settings. These experiential and communicative features create strong intuitive plausibility even when mechanistic evidence is lacking.

Modern neuroscience and mechanobiology, however, contradict the literal interpretation of somatic‑memory narratives. Interoception and trauma network models show that subjective bodily feeling and affective memory are produced by distributed brain systems (insula, amygdala, hippocampus, prefrontal cortex) that integrate peripheral signals with prior beliefs and salience tagging (Craig, 2009; Lanius et al., 2015). Pain‑science work explains that persistent symptoms more plausibly arise from central sensitization, altered descending control, and predictive‑processing biases rather than from peripheral episodic stores (Moseley & Butler, 2015). Mechanobiology demonstrates that ECM and fibroblast adaptations are mechanical and transcriptional responses (YAP/TAZ, stiffness changes) not representational encodings of events (Discher et al., 2009; Paszek et al., 2005; Iskratsch et al., 2014). Together these constraints show why somatic‑memory narratives fail as a literal biological account even while they remain psychologically and culturally resonant.

𝐊𝐞𝐲 𝐓𝐚𝐤𝐞𝐚𝐰𝐚𝐲𝐬 — 𝐌𝐲𝐭𝐡 𝐯𝐞𝐫𝐬𝐮𝐬 𝐌𝐞𝐜𝐡𝐚𝐧𝐢𝐬𝐦

• 𝐌𝐲𝐭𝐡 — 𝐋𝐨𝐜𝐚𝐭𝐢𝐨𝐧: Emotions or memories are “trapped” in fascia or muscle knots.

• 𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: Autobiographical memories are encoded in distributed brain networks (amygdala, hippocampus, cortex) (Craig, 2009).

• 𝐌𝐲𝐭𝐡 — 𝐌𝐞𝐜𝐡𝐚𝐧𝐢𝐬𝐦: Massage “releases” toxins or emotional energy.

• 𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: Massage transiently modulates peripheral sensory input and autonomic state (↑HF‑HRV), providing interoceptive feedback that shapes threat appraisal without extracting metabolites or encoding autobiographical memory (Craig, 2009; Diego & Field, 2009; Roura et al., 2021).

• 𝐌𝐲𝐭𝐡 — 𝐑𝐨𝐥𝐞 𝐨𝐟 𝐁𝐨𝐝𝐲: The body is a storage vessel for past events.

• 𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: The body supplies interoceptive signals that influence how the brain appraises safety versus threat (Craig, 2009; Lanius et al., 2015).

• 𝐌𝐲𝐭𝐡 — 𝐂𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐆𝐨𝐚𝐥: Force tissue to “release” memories.

• 𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: The clinical aim is to reduce autonomic arousal and support central reappraisal; massage is a supportive tool within multimodal care (Moseley & Butler, 2015).

𝐌𝐚𝐬𝐬𝐚𝐠𝐞 𝐭𝐡𝐞𝐫𝐚𝐩𝐲, 𝐚𝐮𝐭𝐨𝐧𝐨𝐦𝐢𝐜 𝐦𝐨𝐝𝐮𝐥𝐚𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐩𝐚𝐢𝐧 𝐬𝐜𝐢𝐞𝐧𝐜𝐞

Massage commonly produces small, short‑term autonomic changes. Controlled trials and systematic reviews report modest increases in HF‑HRV and small reductions in heart rate and blood pressure immediately after moderate‑pressure massage (Diego & Field, 2009; Monteiro et al., 2025 (in press); Roura et al., 2021; Thadanatthaphak et al., 2024; Van Dijk et al., 2020). Effect sizes and directions vary by technique, body region, and population. Typical short‑term blood‑pressure reductions reported after moderate‑pressure massage are small — often only a few mmHg — and HF‑HRV changes are generally modest and transient; however, effect sizes vary substantially by technique, timing, population, and measurement method, and overall certainty is low to moderate (Diego & Field, 2009; Roura et al., 2021; Van Dijk et al., 2020).

𝐐𝐮𝐚𝐧𝐭𝐢𝐭𝐚𝐭𝐢𝐯𝐞 𝐜𝐨𝐧𝐭𝐞𝐱𝐭

Individual trials sometimes report statistically significant acute changes, but high heterogeneity in protocols (pressure, duration, control groups) and outcome measures often precludes reliable meta‑analytic pooling; when pooling is possible, effects are typically small to moderate and short‑lived (Roura et al., 2021; Monteiro et al., 2025 (in press)).

𝐋𝐢𝐦𝐢𝐭𝐚𝐭𝐢𝐨𝐧𝐬 𝐚𝐧𝐝 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲 𝐨𝐟 𝐞𝐯𝐢𝐝𝐞𝐧𝐜𝐞

Current evidence for the autonomic and clinical effects of massage is of low to moderate certainty. Many primary trials are limited by small sample sizes, brief or inconsistent follow‑up, incomplete blinding, heterogeneous protocols (pressure, duration, control conditions), and variable outcome measurement (HRV metrics, blood‑pressure timing), which together reduce confidence in long‑term effect estimates and complicate meta‑analytic synthesis (Roura et al., 2021). There is relatively less high‑quality evidence that massage alone produces durable reductions in chronic pain or trauma‑related symptoms; where clinical benefits are reported they are often modest and context‑dependent (Moseley & Butler, 2015). Mechanistic inferences are further constrained by the translational gap between robust in‑vitro or chronic‑loading cellular findings and the low‑magnitude, short‑duration mechanical stimuli used in routine therapeutic massage (Discher et al., 2009; Paszek et al., 2005). Readers should interpret reported benefits as short‑term physiological modulation that supports multimodal care, not as evidence of durable cures.

𝐋𝐢𝐦𝐢𝐭𝐚𝐭𝐢𝐨𝐧𝐬

• 𝐒𝐞𝐥𝐞𝐜𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐝𝐞𝐬𝐢𝐠𝐧 𝐛𝐢𝐚𝐬: This is a narrative, non‑systematic review; selection bias is possible because sources were chosen to illustrate mechanistic constraints and representative clinical findings rather than to exhaustively identify all studies.

• 𝐋𝐚𝐧𝐠𝐮𝐚𝐠𝐞 𝐚𝐧𝐝 𝐩𝐮𝐛𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧 𝐛𝐢𝐚𝐬: Non‑English sources and non‑peer‑reviewed reports were excluded, which may omit relevant data and introduce language/publication bias.

• 𝐀𝐛𝐬𝐞𝐧𝐜𝐞 𝐨𝐟 𝐩𝐨𝐨𝐥𝐞𝐝 𝐞𝐬𝐭𝐢𝐦𝐚𝐭𝐞𝐬 𝐟𝐨𝐫 𝐬𝐨𝐦𝐞 𝐨𝐮𝐭𝐜𝐨𝐦𝐞𝐬: High heterogeneity and inconsistent reporting precluded robust meta‑analytic pooling for several outcomes; where pooled estimates exist they are often unstable.

• 𝐌𝐞𝐚𝐬𝐮𝐫𝐞𝐦𝐞𝐧𝐭 𝐡𝐞𝐭𝐞𝐫𝐨𝐠𝐞𝐧𝐞𝐢𝐭𝐲 𝐚𝐧𝐝 𝐬𝐡𝐨𝐫𝐭 𝐟𝐨𝐥𝐥𝐨𝐰‑𝐮𝐩: Variable HRV metrics, short recording durations, and brief clinical follow‑up limit interpretability and generalizability.

• 𝐓𝐫𝐚𝐧𝐬𝐥𝐚𝐭𝐢𝐨𝐧𝐚𝐥 𝐥𝐢𝐦𝐢𝐭𝐬: Cellular mechanobiology findings derive largely from in‑vitro or chronic‑loading models and do not directly map onto routine therapeutic massage parameters (Discher et al., 2009; Paszek et al., 2005).

𝐏𝐞𝐫𝐢𝐩𝐡𝐞𝐫𝐚𝐥 𝐢𝐧𝐩𝐮𝐭, 𝐧𝐨𝐜𝐢𝐜𝐞𝐩𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐰𝐡𝐚𝐭 𝐭𝐢𝐬𝐬𝐮𝐞𝐬 𝐜𝐚𝐧 𝐚𝐧𝐝 𝐜𝐚𝐧𝐧𝐨𝐭 𝐝𝐨

Mechanical forces from massage activate low‑threshold mechanoreceptors, alter interstitial fluid dynamics, and engage mechanotransduction pathways in cells and ECM (Discher et al., 2009; Iskratsch et al., 2014; Paszek et al., 2005). These adaptations can change fibroblast contractility, collagen stiffness, and gene expression (“mechanical memory”), but do not encode autobiographical experience. Clinically, such changes may contribute to short‑term comfort but are unlikely to produce large‑scale structural remodeling typical of chronic loading or injury.

Nociception begins with peripheral receptors, but persistent pain typically reflects central processing that integrates sensory input with affect, attention, and context; central sensitization and altered threat appraisal are more explanatory for chronic pain than any hypothetical “memory” stored in collagen matrices (Moseley & Butler, 2015).

𝐏𝐞𝐫𝐢𝐩𝐡𝐞𝐫𝐚𝐥–𝐬𝐩𝐢𝐧𝐚𝐥–𝐛𝐫𝐚𝐢𝐧 𝐩𝐚𝐭𝐡𝐰𝐚𝐲𝐬: 𝐟𝐫𝐨𝐦 𝐦𝐞𝐜𝐡𝐚𝐧𝐨𝐫𝐞𝐜𝐞𝐩𝐭𝐢𝐨𝐧 𝐭𝐨 𝐚𝐮𝐭𝐨𝐧𝐨𝐦𝐢𝐜 𝐚𝐧𝐝 𝐢𝐧𝐭𝐞𝐫𝐨𝐜𝐞𝐩𝐭𝐢𝐯𝐞 𝐦𝐨𝐝𝐮𝐥𝐚𝐭𝐢𝐨𝐧

𝐌𝐞𝐜𝐡𝐚𝐧𝐨𝐫𝐞𝐜𝐞𝐩𝐭𝐨𝐫 𝐚𝐜𝐭𝐢𝐯𝐚𝐭𝐢𝐨𝐧

Mechanical stimulation from massage activates low‑threshold mechanoreceptors (Aβ fibers) and slowly adapting receptors in skin and muscle, increasing non‑nociceptive afferent traffic to the dorsal horn and brainstem; this input transiently inhibits nociceptive transmission via spinal inhibitory interneurons (gate‑control) and reduces peripheral nociceptor sensitization (Moseley & Butler, 2015).

𝐋𝐨𝐜𝐚𝐥 𝐛𝐢𝐨𝐜𝐡𝐞𝐦𝐢𝐜𝐚𝐥 𝐬𝐡𝐢𝐟𝐭𝐬

Massage alters interstitial fluid dynamics and can transiently change metabolites, cytokines, and neuropeptides; these reversible shifts may reduce neurogenic inflammation and local nociceptor excitability for short periods (Discher et al., 2009; Paszek et al., 2005).

𝐒𝐩𝐢𝐧𝐚𝐥 𝐚𝐧𝐝 𝐛𝐫𝐚𝐢𝐧𝐬𝐭𝐞𝐦 𝐦𝐨𝐝𝐮𝐥𝐚𝐭𝐢𝐨𝐧

Increased non‑nociceptive input recruits descending inhibitory systems (periaqueductal gray, rostroventral medulla) and brainstem autonomic centers (nucleus tractus solitarius), producing modest autonomic shifts measurable as HF‑HRV increases and small reductions in heart rate and blood pressure immediately after massage (Diego & Field, 2009; Roura et al., 2021).

𝐀𝐮𝐭𝐨𝐧𝐨𝐦𝐢𝐜 𝐫𝐞𝐠𝐮𝐥𝐚𝐭𝐢𝐨𝐧, 𝐢𝐧𝐭𝐞𝐫𝐨𝐜𝐞𝐩𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐡𝐢𝐠𝐡𝐞𝐫‑𝐨𝐫𝐝𝐞𝐫 𝐧𝐞𝐭𝐰𝐨𝐫𝐤 𝐢𝐧𝐭𝐞𝐠𝐫𝐚𝐭𝐢𝐨𝐧

𝐀𝐮𝐭𝐨𝐧𝐨𝐦𝐢𝐜 𝐦𝐚𝐫𝐤𝐞𝐫𝐬

Moderate‑pressure massage reliably produces small, short‑lived increases in HF‑HRV (transient vagal predominance) and modest reductions in blood pressure; effects are reproducible but typically acute and return toward baseline without repeated or adjunctive interventions (Diego & Field, 2009; Roura et al., 2021).

𝐇𝐅‑𝐇𝐑𝐕 𝐦𝐞𝐭𝐫𝐢𝐜𝐬 𝐚𝐧𝐝 𝐦𝐞𝐚𝐬𝐮𝐫𝐞𝐦𝐞𝐧𝐭 𝐜𝐚𝐯𝐞𝐚𝐭𝐬

When reporting HF‑HRV, specify the metric used (absolute HF power, normalized HF units, or natural‑log transformed HF power) because effect sizes and interpretability differ; note that HRV derived from very short recordings (<~5 minutes) can be unreliable and sensitive to respiration and posture. Where possible, report recording length, sampling rate, and whether HF power was normalized or log‑transformed.

𝐈𝐧𝐭𝐞𝐫𝐨𝐜𝐞𝐩𝐭𝐢𝐯𝐞 𝐬𝐢𝐠𝐧𝐚𝐥𝐢𝐧𝐠 𝐚𝐧𝐝 𝐩𝐫𝐞𝐜𝐢𝐬𝐢𝐨𝐧

Interoceptive afferents ascend via lamina I to the insula, where bodily states are integrated with limbic and prefrontal networks (Craig, 2009). Massage can alter interoceptive precision, biasing predictive processing toward safer interpretations of bodily cues and reducing threat‑biased attention — without implying peripheral mnemonic storage (Craig, 2009; Lanius et al., 2015).

𝐓𝐨𝐩‑𝐝𝐨𝐰𝐧 𝐫𝐞𝐚𝐩𝐩𝐫𝐚𝐢𝐬𝐚𝐥 𝐚𝐧𝐝 𝐩𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐯𝐞 𝐩𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠

Altered interoceptive input interacts with prefrontal and limbic circuits to permit cognitive reappraisal and emotional regulation. Subjective reports of “release” during massage are best explained by interoceptive recalibration and context‑dependent cortical processing rather than retrieval of information stored in connective tissue (Moseley & Butler, 2015; Fenster et al., 2018).

𝐏𝐫𝐞𝐝𝐢𝐜𝐭𝐢𝐯𝐞‑𝐩𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠 𝐚𝐜𝐜𝐨𝐮𝐧𝐭 𝐨𝐟 𝐩𝐞𝐫𝐜𝐞𝐢𝐯𝐞𝐝 “𝐫𝐞𝐥𝐞𝐚𝐬𝐞.” Sensations during massage are interpreted through a predictive‑processing hierarchy: prior beliefs (threat‑weighted priors), interoceptive precision, and incoming sensory evidence jointly determine subjective meaning. Under chronic stress or trauma, priors are biased toward threat and precision weighting favors interoceptive signals that confirm danger. Massage transiently alters interoceptive input and autonomic tone (reduced sympathetic drive, increased vagal influence), lowering precision on threat‑biased priors and permitting prediction‑error resolution. The resulting affective “release” reflects relaxation of maladaptive priors and re‑weighting of interoceptive evidence, not retrieval of symbolic episodic content from peripheral tissue (Craig, 2009; Lanius et al., 2015).

𝐂𝐨𝐦𝐦𝐨𝐧 𝐌𝐢𝐬𝐢𝐧𝐭𝐞𝐫𝐩𝐫𝐞𝐭𝐚𝐭𝐢𝐨𝐧𝐬 𝐚𝐧𝐝 𝐖𝐡𝐲 𝐓𝐡𝐞𝐲 𝐅𝐚𝐢𝐥 𝐌𝐞𝐜𝐡𝐚𝐧𝐢𝐬𝐭𝐢𝐜𝐚𝐥𝐥𝐲

• “𝑪𝒍𝒊𝒆𝒏𝒕𝒔 𝒄𝒓𝒚 𝒐𝒏 𝒕𝒉𝒆 𝒕𝒂𝒃𝒍𝒆, 𝒔𝒐 𝒕𝒊𝒔𝒔𝒖𝒆 𝒓𝒆𝒍𝒆𝒂𝒔𝒆𝒔 𝒎𝒆𝒎𝒐𝒓𝒚.” Emotional expression during massage is explained by limbic–autonomic coactivation and interoceptive signaling: affective arousal can be triggered by bodily sensations and contextual safety cues without any peripheral mnemonic substrate.

• “𝑭𝒂𝒔𝒄𝒊𝒂 𝒉𝒐𝒍𝒅𝒔 𝒕𝒆𝒏𝒔𝒊𝒐𝒏 𝒑𝒂𝒕𝒕𝒆𝒓𝒏𝒔 𝒕𝒉𝒂𝒕 𝒆𝒒𝒖𝒂𝒍 𝒕𝒓𝒂𝒖𝒎𝒂.” Tension patterns reflect motor output, altered neural drive, and local tissue mechanics; they do not contain episodic representations. Motor patterns can be shaped by central sensitization and predictive coding rather than peripheral storage.

• "𝑪𝒆𝒍𝒍𝒔 𝒓𝒆𝒎𝒆𝒎𝒃𝒆𝒓 𝒍𝒐𝒂𝒅, 𝒔𝒐 𝒕𝒉𝒆𝒚 𝒄𝒂𝒏 𝒓𝒆𝒎𝒆𝒎𝒃𝒆𝒓 𝒕𝒓𝒂𝒖𝒎𝒂." Cellular mechanotransduction produces phenotype and stiffness changes but these are non‑symbolic, low‑resolution adaptations (mechanical history), not representational memory of events or affect.

• “𝑺𝒐𝒎𝒂𝒕𝒊𝒄 𝒎𝒆𝒎𝒐𝒓𝒚 𝒊𝒔 𝒔𝒕𝒐𝒓𝒆𝒅 𝒆𝒗𝒆𝒓𝒚𝒘𝒉𝒆𝒓𝒆.” Distributed cortical engrams and hippocampal indexing explain how episodic content is stored and retrieved; peripheral tissues lack the necessary circuit and synaptic architecture.

𝐀𝐝𝐝𝐫𝐞𝐬𝐬𝐢𝐧𝐠 𝐭𝐡𝐞 𝐬𝐭𝐫𝐨𝐧𝐠𝐞𝐬𝐭 𝐜𝐨𝐮𝐧𝐭𝐞𝐫𝐚𝐫𝐠𝐮𝐦𝐞𝐧𝐭𝐬 - The most persuasive objections conflate different biological scales (molecular/cellular vs. network/neural) and different information types (mechanical history vs. episodic content). Each counterargument fails because it mistakes state changes (autonomic, interoceptive) for representational storage and overlooks the computational and anatomical requirements for episodic encoding.

𝐌𝐞𝐜𝐡𝐚𝐧𝐨𝐭𝐫𝐚𝐧𝐬𝐝𝐮𝐜𝐭𝐢𝐨𝐧, 𝐞𝐱𝐭𝐫𝐚𝐜𝐞𝐥𝐥𝐮𝐥𝐚𝐫 𝐦𝐚𝐭𝐫𝐢𝐱, 𝐚𝐧𝐝 𝐭𝐡𝐞 “𝐦𝐞𝐜𝐡𝐚𝐧𝐢𝐜𝐚𝐥 𝐦𝐞𝐦𝐨𝐫𝐲” 𝐦𝐢𝐬𝐜𝐨𝐧𝐜𝐞𝐩𝐭𝐢𝐨𝐧

Mechanotransduction pathways have been well described (integrins → focal adhesions → cytoskeleton → nuclear signaling; YAP/TAZ) (Discher et al., 2009; Iskratsch et al., 2014; Paszek et al., 2005). Massage loads are substantially lower and far shorter in duration than the mechanical environments used to induce ECM stiffening and YAP/TAZ‑mediated transcriptional changes in vitro, so routine therapeutic sessions do not reproduce the sustained, high‑magnitude stimuli required for those cellular programs (Discher et al., 2009; Paszek et al., 2005). These processes produce persistent changes in fibroblast contractility and ECM stiffness in experimental and chronic‑loading models. While these adaptations affect tissue mechanics, they do not encode autobiographical experience.

𝐖𝐡𝐚𝐭 𝐅𝐚𝐬𝐜𝐢𝐚 𝐂𝐚𝐧 𝐒𝐭𝐨𝐫𝐞 𝐯𝐬. 𝐖𝐡𝐚𝐭 𝐈𝐭 𝐂𝐚𝐧𝐧𝐨𝐭 𝐒𝐭𝐨𝐫𝐞

𝐖𝐡𝐚𝐭 𝐟𝐚𝐬𝐜𝐢𝐚 𝐜𝐚𝐧 𝐬𝐭𝐨𝐫𝐞: Connective tissue reliably records mechanical history: changes in ECM stiffness, fibroblast phenotype shifts, and low‑resolution structural adaptations mediated by mechanotransduction pathways (integrins → focal adhesions → cytoskeleton → nuclear signaling; YAP/TAZ). These adaptations alter tissue mechanics and can influence local nociceptive sensitivity, contributing to transient changes in comfort, tone, and peripheral sensitization. This storage is non‑symbolic and encodes load‑response patterns (how tissue responds to force), not semantic or episodic content.

𝐖𝐡𝐚𝐭 𝐟𝐚𝐬𝐜𝐢𝐚 𝐜𝐚𝐧𝐧𝐨𝐭 𝐬𝐭𝐨𝐫𝐞: Fascia and muscle lack the synaptic networks, neurotransmission, recurrent circuitry, and consolidation mechanisms required for episodic or autobiographical memory. They cannot encode symbolic content, narrative meaning, affective valence, or trauma scripts; those functions require distributed neural ensembles and hippocampal‑cortical consolidation (Craig, 2009; Paszek et al., 2005). Framing ECM adaptation as “memory” is therefore a category error: durable, low‑level mechanical adaptation is not equivalent to representational memory.

𝐂𝐚𝐭𝐞𝐠𝐨𝐫𝐲 𝐞𝐫𝐫𝐨𝐫 𝐜𝐥𝐚𝐫𝐢𝐟𝐢𝐞𝐝 — 𝐦𝐞𝐜𝐡𝐚𝐧𝐢𝐬𝐭𝐢𝐜 𝐜𝐡𝐚𝐢𝐧 (𝐰𝐡𝐲 𝐟𝐚𝐬𝐜𝐢𝐚 𝐜𝐚𝐧𝐧𝐨𝐭 𝐬𝐭𝐨𝐫𝐞 𝐭𝐫𝐚𝐮𝐦𝐚)

Autobiographical and episodic memory require neural circuit architectures that support representational coding and synaptic plasticity (hippocampal–neocortical ensembles, recurrent networks, LTP/LTD, neurotransmission and oscillatory coordination). Fascia and extracellular matrix (ECM) lack synapses, neurotransmission, recurrent neural circuitry, and the molecular machinery for representational consolidation; they therefore cannot instantiate hippocampal‑style engrams. Mechanotransduction in fibroblasts and ECM produces biophysical and transcriptional adaptations (stiffness, phenotype shifts) that are non‑representational: they encode mechanical history (load, strain) at low spatial and informational resolution, not episodic content or affective meaning. Fear and trauma encoding depend on distributed limbic‑cortical systems (amygdala, hippocampus, prefrontal cortex) that tag salience, assign affective valence, and update predictive priors; these processes require synaptic ensembles and network‑level plasticity absent from connective tissue. In short: no synapses → no representational code → no episodic engram (Paszek et al., 2005; Discher et al., 2009).

𝐂𝐞𝐧𝐭𝐫𝐚𝐥 𝐬𝐞𝐧𝐬𝐢𝐭𝐢𝐳𝐚𝐭𝐢𝐨𝐧, 𝐭𝐫𝐚𝐮𝐦𝐚‑𝐫𝐞𝐥𝐚𝐭𝐞𝐝 𝐧𝐞𝐭𝐰𝐨𝐫𝐤 𝐝𝐲𝐬𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐭𝐫𝐞𝐚𝐭𝐦𝐞𝐧𝐭 𝐢𝐦𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬

𝐂𝐞𝐧𝐭𝐫𝐚𝐥 𝐬𝐞𝐧𝐬𝐢𝐭𝐢𝐳𝐚𝐭𝐢𝐨𝐧

Persistent pain often reflects central nervous system plasticity — enhanced dorsal horn excitability, altered descending modulation, and cortical reorganization — rather than ongoing peripheral tissue damage. Central sensitization amplifies afferent signals and lowers pain thresholds, producing widespread hyperalgesia and allodynia (Moseley & Butler, 2015). Transient increases in vagal tone reduce sympathetic drive and interoceptive threat bias, creating a temporary window of reduced hypervigilance that supports top‑down reappraisal.

𝐋𝐚𝐫𝐠𝐞‑𝐬𝐜𝐚𝐥𝐞 𝐧𝐞𝐭𝐰𝐨𝐫𝐤 𝐝𝐲𝐬𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧 𝐢𝐧 𝐭𝐫𝐚𝐮𝐦𝐚

Long‑term changes in pain and threat perception are best explained by altered resting‑state connectivity among networks (insula, ACC, amygdala, hippocampus) that modify interoceptive precision and threat appraisal (Craig, 2009; Fenster et al., 2018; Lanius et al., 2015; Nicholson et al., 2016). Chronic stress and trauma are commonly associated with shifts in precision weighting toward threat and over‑prediction; durable clinical improvements therefore reflect changes in brain–body regulation — learning to interpret bodily sensations as safe — rather than peripheral “corrections.”

𝐏𝐬𝐞𝐮𝐝𝐨𝐬𝐜𝐢𝐞𝐧𝐭𝐢𝐟𝐢𝐜 𝐜𝐥𝐚𝐢𝐦𝐬 𝐜𝐨𝐧𝐭𝐫𝐚𝐬𝐭𝐞𝐝 𝐰𝐢𝐭𝐡 𝐞𝐯𝐢𝐝𝐞𝐧𝐜𝐞

𝐂𝐥𝐚𝐢𝐦: Muscles or fascia store autobiographical memories. 𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: No plausible neurobiological substrate supports episodic memory encoding in ECM or muscle tissue; episodic memory requires synaptic ensembles and hippocampal‑cortical consolidation (Craig, 2009; Moseley & Butler, 2015).

𝐑𝐞𝐚𝐥𝐢𝐭𝐲: Tissue changes reflect mechanical history and cellular adaptation (Discher et al., 2009; Paszek et al., 2005).

𝐂𝐥𝐚𝐢𝐦: Massage “releases toxins” or emotional energy.

𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: Transient metabolic or inflammatory marker changes after massage do not equate to liberation of stored emotional content; systemic clearance mechanisms rapidly metabolize local metabolites (Diego & Field, 2009).

𝐑𝐞𝐚𝐥𝐢𝐭𝐲: Subjective feelings of release are mediated by interoceptive recalibration and cognitive‑emotional processing.

𝐂𝐥𝐚𝐢𝐦: Palpable “knots” are repositories of trauma.

𝐄𝐯𝐢𝐝𝐞𝐧𝐜𝐞: Palpable tightness or trigger points reflect localized muscle tone, altered motor control, or fascial stiffness; these are physiological states influenced by neural drive and tissue mechanics (Moseley & Butler, 2015).

𝐑𝐞𝐚𝐥𝐢𝐭𝐲: Addressing motor patterns, autonomic state, and cognitive appraisal explains symptom change more parsimoniously than tissue‑memory narratives.

Because mechanistic findings, autonomic changes, and clinical outcomes occur at different biological scales, it is important to rate evidence separately using a structured framework. The following GRADE summary highlights where the evidence is strongest (cellular mechanotransduction) and weakest (durable clinical outcomes).

𝐆𝐑𝐀𝐃𝐄 𝐞𝐯𝐢𝐝𝐞𝐧𝐜𝐞 𝐬𝐮𝐦𝐦𝐚𝐫𝐲

𝐎𝐯𝐞𝐫𝐚𝐥𝐥 𝐬𝐭𝐚𝐭𝐞𝐦𝐞𝐧𝐭: The evidence base is heterogeneous and imprecise: physiological effects are generally small and short‑lived; mechanistic cellular findings are robust but largely preclinical; and clinical outcomes for chronic pain are low‑certainty. Heterogeneity, inconsistency, and imprecision across trials make pooled estimates unstable, so interpret effect sizes and confidence intervals cautiously. Report 95% confidence intervals and formal risk‑of‑bias assessments alongside pooled estimates to make imprecision and inconsistency explicit.

• 𝐀𝐜𝐮𝐭𝐞 𝐇𝐅‑𝐇𝐑𝐕 𝐜𝐡𝐚𝐧𝐠𝐞 𝐚𝐟𝐭𝐞𝐫 𝐚 𝐬𝐢𝐧𝐠𝐥𝐞 𝐬𝐞𝐬𝐬𝐢𝐨𝐧 — 𝐒𝐦𝐚𝐥𝐥 𝐢𝐧𝐜𝐫𝐞𝐚𝐬𝐞; 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲: 𝐦𝐨𝐝𝐞𝐫𝐚𝐭𝐞. 𝑅𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑒: Multiple trials report consistent small HF‑HRV increases, but heterogeneity in metrics and short follow‑up limit certainty (Diego & Field, 2009; Roura et al., 2021).

• 𝐑𝐞𝐬𝐭𝐢𝐧𝐠 𝐛𝐥𝐨𝐨𝐝 𝐩𝐫𝐞𝐬𝐬𝐮𝐫𝐞 𝐚𝐟𝐭𝐞𝐫 𝐚 𝐬𝐢𝐧𝐠𝐥𝐞 𝐬𝐞𝐬𝐬𝐢𝐨𝐧 — 𝐒𝐦𝐚𝐥𝐥 𝐫𝐞𝐝𝐮𝐜𝐭𝐢𝐨𝐧 (𝐭𝐲𝐩𝐢𝐜𝐚𝐥𝐥𝐲 𝐨𝐧𝐥𝐲 𝐚 𝐟𝐞𝐰 𝐦𝐦𝐇𝐠 𝐢𝐧 𝐢𝐧𝐝𝐢𝐯𝐢𝐝𝐮𝐚𝐥 𝐭𝐫𝐢𝐚𝐥𝐬); 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲: 𝐥𝐨𝐰. 𝑅𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑒: Individual trials show small, transient reductions, but pooled estimates are unstable and measurement timing varies (Roura et al., 2021).

• 𝐂𝐡𝐫𝐨𝐧𝐢𝐜 𝐩𝐚𝐢𝐧 𝐫𝐞𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐟𝐫𝐨𝐦 𝐦𝐚𝐬𝐬𝐚𝐠𝐞 𝐚𝐥𝐨𝐧𝐞 — 𝐒𝐦𝐚𝐥𝐥 𝐨𝐫 𝐧𝐨 𝐝𝐮𝐫𝐚𝐛𝐥𝐞 𝐞𝐟𝐟𝐞𝐜𝐭; 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲: 𝐥𝐨𝐰. 𝑅𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑒: Trials are heterogeneous, often underpowered, and benefits frequently attenuate without adjunctive therapies (Moseley & Butler, 2015).

• 𝐅𝐮𝐧𝐜𝐭𝐢𝐨𝐧𝐚𝐥 𝐨𝐮𝐭𝐜𝐨𝐦𝐞𝐬 (𝐞.𝐠., 𝐏𝐑𝐎𝐌𝐈𝐒‑𝟐𝟗) — 𝐈𝐧𝐜𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐯𝐞; 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲: 𝐥𝐨𝐰. 𝑅𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑒: Limited and inconsistent data with short follow‑up and variable outcome selection prevent confident conclusions (Monteiro et al., 2025 (in press)).

• 𝐂𝐞𝐥𝐥𝐮𝐥𝐚𝐫 𝐦𝐞𝐜𝐡𝐚𝐧𝐨𝐭𝐫𝐚𝐧𝐬𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐞𝐟𝐟𝐞𝐜𝐭𝐬 — 𝐃𝐞𝐦𝐨𝐧𝐬𝐭𝐫𝐚𝐭𝐞𝐝 𝐢𝐧 𝐯𝐢𝐭𝐫𝐨 𝐚𝐧𝐝 𝐜𝐡𝐫𝐨𝐧𝐢𝐜‑𝐥𝐨𝐚𝐝𝐢𝐧𝐠 𝐦𝐨𝐝𝐞𝐥𝐬; 𝐜𝐞𝐫𝐭𝐚𝐢𝐧𝐭𝐲: 𝐡𝐢𝐠𝐡 𝐟𝐨𝐫 𝐜𝐞𝐥𝐥𝐮𝐥𝐚𝐫 𝐩𝐡𝐞𝐧𝐨𝐦𝐞𝐧𝐚 𝐛𝐮𝐭 𝐧𝐨 𝐞𝐯𝐢𝐝𝐞𝐧𝐜𝐞 𝐟𝐨𝐫 𝐩𝐞𝐫𝐢𝐩𝐡𝐞𝐫𝐚𝐥 𝐞𝐧𝐜𝐨𝐝𝐢𝐧𝐠 𝐨𝐟 𝐚𝐮𝐭𝐨𝐛𝐢𝐨𝐠𝐫𝐚𝐩𝐡𝐢𝐜𝐚𝐥 𝐦𝐞𝐦𝐨𝐫𝐲. 𝑅𝑎𝑡𝑖𝑜𝑛𝑎𝑙𝑒: Mechanobiology robustly shows ECM and fibroblast adaptations under sustained loading, yet these processes are non‑representational and do not provide a substrate for episodic memory (Discher et al., 2009; Paszek et al., 2005).

𝐑𝐞𝐬𝐞𝐚𝐫𝐜𝐡 𝐩𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐞𝐬 𝐚𝐧𝐝 𝐭𝐫𝐢𝐚𝐥 𝐠𝐮𝐢𝐝𝐚𝐧𝐜𝐞

• Larger, preregistered RCTs with standardized massage protocols, longer follow‑up, pre‑specified mechanistic and clinical outcomes, and open data.

• Expect small effects; plan trials with hundreds per arm; label smaller studies as pilot/feasibility.

• Use objective force measures (pressure sensors), report stroke rate/duration, and document therapist training and fidelity.

• Include active/attention controls; blind outcome assessors and statisticians; use participant blinding where feasible.

• Primary outcome at clinically meaningful follow‑up (~3 months); include 24‑hour HRV, validated functional measures, pain scales, inflammatory biomarkers, and optional brain imaging.

• When pooling is impossible, report reasons, use effect‑direction plots, harmonized estimates where feasible, and structured narrative synthesis with formal bias assessment (RoB 2 / ROBINS‑I) and GRADE.

𝐂𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐠𝐮𝐢𝐝𝐚𝐧𝐜𝐞 𝐚𝐧𝐝 𝐦𝐞𝐬𝐬𝐚𝐠𝐢𝐧𝐠

𝐂𝐥𝐢𝐧𝐢𝐜𝐢𝐚𝐧 𝐬𝐮𝐦𝐦𝐚𝐫𝐲: Massage is an evidence‑based supportive intervention that reliably produces short‑term autonomic modulation and symptom relief for some patients; it is not a mechanism for releasing stored memories from tissues. Use massage within multimodal care and set measurable goals.

𝐒𝐮𝐠𝐠𝐞𝐬𝐭𝐞𝐝 𝐩𝐚𝐭𝐢𝐞𝐧𝐭 𝐥𝐚𝐧𝐠𝐮𝐚𝐠𝐞: “Massage can calm your nervous system and help your brain interpret body signals as safer, which often reduces discomfort. It does not pull-out memories from your muscles.”

𝐑𝐚𝐭𝐢𝐨𝐧𝐚𝐥𝐞 𝐟𝐨𝐫 𝐜𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐠𝐮𝐢𝐝𝐚𝐧𝐜𝐞: Massage reliably modulates autonomic state and interoceptive precision but lacks mechanisms to alter maladaptive predictive priors without concurrent cognitive reappraisal or psychotherapeutic processing. Therefore, massage functions primarily as a state regulator (reducing arousal, improving safety signaling) rather than a mechanism for durable trauma processing.

𝐂𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐫𝐨𝐥𝐞𝐬: Massage primarily acts as a state regulator: it transiently downshifts autonomic arousal and recalibrates interoceptive signals, creating a physiological context of increased safety. Psychotherapy primarily acts as a meaning‑maker: it targets maladaptive priors, supports cognitive reappraisal, and updates long‑term predictive models. When combined, massage can create a physiological window in which psychotherapeutic interventions are more likely to update priors and produce durable change.

𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐨𝐧 𝐜𝐡𝐞𝐜𝐤𝐥𝐢𝐬𝐭: Explain mechanistic rationale (interoceptive recalibration, autonomic modulation); avoid tissue‑memory or toxin‑release language; offer measurable goals (e.g., improved PROMIS‑29, reduced analgesic use); document outcomes and share with the multidisciplinary team.

𝐂𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐈𝐦𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬 𝐟𝐨𝐫 𝐓𝐫𝐚𝐮𝐦𝐚‑𝐈𝐧𝐟𝐨𝐫𝐦𝐞𝐝 𝐌𝐚𝐧𝐮𝐚𝐥 𝐓𝐡𝐞𝐫𝐚𝐩𝐲

Massage is effective as a state regulator: it transiently downshifts sympathetic arousal, modestly increases vagal markers, and recalibrates interoceptive signals in ways that often reduce perceived threat and muscle tone. Massage is not a mechanism for extracting or erasing autobiographical memories; clinicians should avoid language that implies tissue‑level retrieval or toxin‑release. When patients report emotional “release,” frame this ethically as a change in bodily state and predictive priors — a window of lowered threat bias that can make cognitive or exposure‑based interventions more effective. Integrate massage into multimodal, trauma‑informed care by pairing sessions with psychoeducation, graded activity, and psychotherapy rather than presenting massage as a standalone trauma cure. Explicitly correct common misconceptions (e.g., “breaking adhesions,” “releasing stored trauma”) and set measurable, functional goals (pain scores, PROMIS‑29, activity tolerance). Document outcomes and coordinate with mental‑health colleagues so that state regulation from hands‑on care is leveraged safely and transparently to support durable, centrally mediated recovery.

𝐂𝐨𝐧𝐜𝐥𝐮𝐬𝐢𝐨𝐧

Massage helps the nervous system feel safer — it does not extract or unlock stored trauma. Benefits include transient reductions in threat perception, muscle tone, and nociceptive signaling, with modest autonomic improvements and improved interoceptive regulation. Connective tissue lacks the synaptic architecture and network consolidation required for episodic memory, and mechanotransduction produces durable but non‑representational mechanical adaptations; fascia therefore cannot plausibly store autobiographical trauma.

Massage produces short‑lived autonomic and interoceptive shifts that can lower threat‑biased precision and create a physiological window for cognitive or behavioral interventions, but these shifts are not equivalent to retrieval or erasure of episodic content.

𝐶𝑙𝑖𝑛𝑖𝑐𝑎𝑙 𝑡𝑎𝑘𝑒𝑎𝑤𝑎𝑦: present massage as a state‑regulating adjunct that facilitates psychotherapy and rehabilitation, avoid tissue‑memory or toxin‑release language, and prioritize interventions that directly target central plasticity for durable recovery.

𝐑𝐞𝐟𝐞𝐫𝐞𝐧𝐜𝐞𝐬

Craig, A. D. (2009). How do you feel—now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10(1), 59–70. https://doi.org/10.1038/nrn2555

Diego, M. A., & Field, T. (2009). Moderate pressure massage elicits a parasympathetic nervous system response. International Journal of Neuroscience, 119(5), 630–638. https://doi.org/10.1080/00207450802329605

Discher, D. E., Mooney, D. J., & Zandstra, P. W. (2009). Growth factors, matrices, and forces combine and control stem cells. Science, 324(5935), 1673–1677. https://doi.org/10.1126/science.1171643

Dudai, Y. (2004). The neurobiology of consolidations, or, how stable is the engram. Annual Review of Psychology, 55, 51–86. https://doi.org/10.1146/annurev.psych.55.090902.142050

Fenster, R. J., Lebois, L. A. M., Ressler, K. J., & Suh, J. (2018). Brain circuit dysfunction in post‑traumatic stress disorder: From mouse to man. Nature Reviews Neuroscience, 19(9), 535–551. https://doi.org/10.1038/s41583-018-0039-7

Iskratsch, T., Wolfenson, H., & Sheetz, M. (2014). Appreciating force and shape—the rise of mechanotransduction in cell biology. Nature Reviews Molecular Cell Biology, 15(12), 825–833. https://doi.org/10.1038/nrm3903

Lanius, R. A., Frewen, P. A., Tursich, M., Jetly, R., & McKinnon, M. C. (2015). Restoring large‑scale brain networks in PTSD and related disorders: A proposal for neuroscientifically‑informed treatment interventions. European Journal of Psychotraumatology, 6, 27313. https://doi.org/10.3402/ejpt.v6.27313

LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–184. https://doi.org/10.1146/annurev.neuro.23.1.155

Monteiro, E. R., Aguilera, L. M., Ruá‑Alonso, M., et al. (2025). Effect of manual massage, foam rolling, and strength training on hemodynamic and autonomic responses in adults: A scoping review (in press). Healthcare, 13(12), 1371. https://doi.org/10.3390/healthcare13121371

Moseley, G. L., & Butler, D. S. (2015). Fifteen years of explaining pain: The past, present, and future. The Journal of Pain, 16(9), 807–813. https://doi.org/10.1016/j.jpain.2015.05.005

Nicholson, A. A., Sapru, I., Densmore, M., et al. (2016). Unique insula subregion resting‑state functional connectivity with amygdala complexes in posttraumatic stress disorder and its dissociative subtype. Psychiatry Research: Neuroimaging, 250, 61–72. https://doi.org/10.1016/j.pscychresns.2016.02.002

Paszek, M. J., Zahir, N., Johnson, K. R., et al. (2005). Tensional homeostasis and the malignant phenotype. Cancer Cell, 8(3), 241–255. https://doi.org/10.1016/j.ccr.2005.08.010

Roura, S., Álvarez, G., Solà, I., & Cerritelli, F. (2021). Do manual therapies have a specific autonomic effect? An overview of systematic reviews. PLOS ONE, 16(12), e0260642. https://doi.org/10.1371/journal.pone.0260642

Van Dijk, M. J., Ravesloot, J. H., van Wingerden, J., & van Mechelen, W. (2020). The effect of mechanical massage and mental training on heart rate variability: A randomized trial. Frontiers in Public Health, 8, 82. https://doi.org/10.3389/fpubh.2020.00082