Human Touch in Musculoskeletal Care: Neurophysiology, Computational Modelling, and Clinical Outcomes

Human Touch in Musculoskeletal Care: Neurophysiology, Computational Modelling, and Clinical Outcomes

Is human touch an essential, evidence-based tool in musculoskeletal (MSK) pain management?  This article examines how touch‑related sensory input is conceptualised within contemporary neurophysiological and computational models of pain, and whether these models offer a biologically and neuro‑scientifically plausible rationale for touch‑inclusive pain management strategies.

Gate Control Theory: Neurophysiological Foundation

The gate control theory of pain, first proposed by Melzack and Wall in 1965, proposes that pain transmission in the spinal cord is subject to modulation by sensory fibres (Ropero Peláez & Taniguchi, 2016). Large-diameter, myelinated Aβ fibres, which mediate low‑threshold touch and pressure, can, under certain conditions, inhibit transmission from small-diameter unmyelinated C fibres and thinly myelinated Aδ fibres, which convey slow burning and fast acute pain respectively (Ropero Peláez & Taniguchi, 2016).

Ropero Peláez and Taniguchi revisited this gate circuit using a parsimonious neurocomputational model in which mechanoreceptive and nociceptive inputs converge on dorsal horn neurons and undergo both synaptic and intrinsic plasticity. Computational models demonstrate that when tactile input co-occurs with nociceptive stimulation-such as when rubbing an injured site-gate circuits suppress pain signals through synaptic and intrinsic neuronal plasticity, in a way that is qualitatively consistent with experimentally observed gating phenomena (Ropero Peláez & Taniguchi, 2016). This mechanism provides a rationale for the observation that tactile, hands‑on interventions, such as manual therapy, soft tissue techniques or joint mobilisation, can modulate nociceptive signalling “at source” in the spinal cord before signals reach higher brain centres,  in certain conditions and time frames, and may therefore support a rationale for hands‑on therapy, exercise‑evoked sensory input and graded movement exposure in musculoskeletal pain management (Ropero Peláez & Taniguchi, 2016).

Mathematical and Computational Models: Touch Validated by Simulation

Beyond specific gate control circuits, contemporary mathematical and computational models describe pain as an emergent property of interacting sensory, spinal, and supra-spinal processes (Lang et al., 2021). These models demonstrate that pain responses depend on the dynamic interaction between sensory and nociceptive input (Lang et al., 2021). In gate‑style spinal models, when low‑threshold tactile input exceeds nociceptive drive, the circuit can converge to a pain‑inhibiting regime, whereas altered or reduced sensory input-as in demyelinating syndromes-can yield aberrant pain signalling (Ropero Peláez & Taniguchi, 2016). Collectively, such mathematical approaches reproduce key qualitative features of gate control and related mechanisms, reinforcing the neurophysiological plausibility of touch‑mediated modulation of pain while also underscoring that such simulations, on their own, do not establish clinical effectiveness or long‑term outcomes of touch‑based therapies. (Lang et al., 2021; Ropero Peláez & Taniguchi, 2016).

Towards multidimensional models of pain

However, these modelling studies also highlight important limitations. Gate control inspired models remain highly simplified representations that depend on numbers of parameters, particular choices of synaptic weights and learning rules, and controlled input patterns; and they do not by themselves capture the multidimensional experience of pain, which includes sensory‑discriminative, affective, cognitive, and social dimensions (Lang et al., 2021; Chen et al., 2023). Recent computational work therefore proposes that spinal gating mechanisms are better viewed as subcomponents within broader hierarchical frameworks-such as Bayesian, predictive‑coding, and reinforcement‑learning models-in which expectations, prior beliefs, and learned values shape how incoming nociceptive and other sensory signals are interpreted as painful or non‑painful (Chen et al., 2023; Lang et al., 2021).

Bayesian and predictive models of pain learning indicate that prior experiences, contextual cues and psychological processes-including expectations, cognitions and emotions-can, through changes in uncertainty and prior beliefs relative to current sensory input, help to explain persistent or apparently spontaneous pain that is not fully explained by ongoing tissue damage, thereby illustrating limitations of explanations based solely on spinal mechanisms (Ramne et al., 2024; Lang et al., 2021). Fuzzy cognitive map approaches extend this work by including psychological, social and individual variables alongside neural and tissue‑damage, considering a biopsychosocial network in which psychological and neural factors mutually reinforce one another and in which social factors may exert net inhibitory influences on reported pain within the simulations (Farahani et al., 2025). Nevertheless, these models remain relatively high‑level and rely heavily on expert‑defined structures and assumptions, underscoring that current computational frameworks still provide only partial approximations of the lived, multidimensional experience of pain (Farahani et al., 2025; Lang et al., 2021).

Summary and limitations

In summary, the studies discussed here indicate that touch‑related sensory input is a biologically plausible contributor to the modulation of nociceptive processing, and that spinal gating and broader computational frameworks can help to explain some aspects of pain perception. At the same time, these models are simplified, rely on specific assumptions and parameters, and do not directly test or confirm the clinical effectiveness, optimal dosing, or long‑term outcomes of touch‑based therapies. The ideas about therapeutic touch supporting trust, perceived safety and engagement with active care are therefore presented as clinically grounded reflections that are consistent with biopsychosocial perspectives, rather than as direct conclusions from the computational work itself. This article is intended to be informative and does not constitute treatment claims, and any hands‑on approach should be considered as one possible component within a person‑centred, multi‑modal plan developed in line with current professional guidance (Lang et al., 2021; Chen et al., 2023; Ramne et al., 2024; Farahani et al., 2025).

Integrating touch within a holistic, biopsychosocial movement model

Osteopathic practice shows me how skilfully applied human touch can ease pain and help people feel safer and more at home in their bodies. That experience naturally raises an important clinical question: does this early relief and sense of being cared for enable patients to engage with, and stick to, their aftercare plans? Could therapeutic touch act as an enabler for exercise adherence by building trust, reducing perceived threat, and creating conditions in which people feel ready to move again? I care deeply about these possibilities, both as a clinician at the treatment table and as someone committed to supporting patients through the longer, often challenging work of active self‑management.

That leads to a broader challenge: how can clinicians integrate therapeutic touch within person‑centred, exercise‑based care so that it genuinely strengthens and sustains patients’ active self‑management? I will be exploring some practical, evidence‑informed possibilities for this integration next week.

References

Chen, Z. S., & Wang, J. (2023). Pain, from perception to action: A computational perspective. iScience, 26(1), 105707.

Farahani, H., Kovač, N., Fardi, H., & Watson, P. C. (2025). Modelling pain perception using fuzzy cognitive maps. Journal of Pain Research, 18,

Lang VA, Lundh T, Ortiz-Catalan M. Mathematical and Computational Models for Pain: A Systematic Review. Pain Medicine. 2021;22(12):2806-2817. doi:10.1093/pm/pnab177

Ramne, M., & Sensinger, J. (2024). A computational framework for understanding the impact of prior experiences on pain perception and neuropathic pain. PLOS Computational Biology, 20(10), e1012097.

Ropero Pelez FJ, Taniguchi S. The Gate Theory of Pain Revisited: Modeling Different Pain Conditions with a Parsimonious Neurocomputational Model. Neural Plasticity. 2016;2016:4131395. doi:10.1155/2016/4131395

Tags:

Leave a Reply

Your email address will not be published. Required fields are marked *