Proprioception

Abstract

Proprioception is the sense of the position and movement of one's own body — the awareness of where the limbs are and how they move without the aid of vision. Cognitive psychology and the movement sciences study it because nearly every skilled action, from reaching to standing, depends on an internal estimate of body state that proprioception supplies. This article describes the receptors that generate proprioceptive signals — muscle spindles, tendon organs, and joint and skin receptors — the discovery that the muscle spindle is the principal organ of limb position sense, the movement illusions produced by vibrating a tendon, the molecular basis of the sense in the PIEZO2 channel, and how proprioceptive acuity is measured and trained. Three interactive demonstrations let the reader drive a vibration illusion, probe a muscle spindle, and run a position-matching test.

Keywords: proprioception, kinaesthesia, muscle spindle, postural balance, touch perception

Proprioception is the sense through which the body knows its own configuration: the position of the limbs, the angles of the joints, the length and tension of the muscles, and the direction and speed of their movement (Proske & Gandevia, 2012). It works largely without awareness — a person can touch finger to nose with the eyes closed, or climb stairs without watching the feet — yet it is indispensable, and its loss is devastating. The clearest evidence comes from the rare patients who lose their large sensory afferents, and with them proprioception, to a sensory neuronopathy — a state of chronic deafferentation: deprived of body sense they cannot control movement without watching each limb, and even under visual guidance their actions are slow and clumsy, showing how completely ordinary action depends on the sense (Rothwell et al., 1982). Charles Sherrington, who named it, grouped the receptors that report the state of the body's own tissues as the proprioceptors and set proprioception alongside the exteroceptive senses that report the outside world (Sherrington, 1906). What makes proprioception a subject for cognitive psychology rather than only physiology is that the raw receptor signals do not, by themselves, specify limb position: the nervous system must interpret ambiguous, movement-dependent afferent activity through an internal model of the body to arrive at a percept of where the body is (Tuthill & Azim, 2018).

What Proprioception Is

Proprioception is one of the body senses, or somatosenses, and in MeSH it is classified as a form of sensation. It is usefully separated into several submodalities that share the same receptors but answer different questions (Proske & Gandevia, 2012). The sense of limb position reports the static angles of the joints; the sense of movement, or kinaesthesia, reports that a segment is moving and in which direction and how fast; the sense of force reports the effort a muscle is exerting; and related to these is a sense of the timing and heaviness of actions (Proske & Gandevia, 2018). These distinctions matter because the submodalities can dissociate — a signal that specifies movement need not specify absolute position — and because different experimental manipulations target them selectively (Proske & Gandevia, 2009).

Two terms are often used interchangeably, but it is worth keeping them apart. Proprioception is the broader category, covering position, movement, and force senses together; kinaesthesia refers more narrowly to the sense of movement and, by extension, position (McCloskey, 1978). The percept is constructed rather than read off directly: the same afferent discharge can arise from different combinations of muscle length and contraction, so the brain must combine the peripheral signal with knowledge of its own motor commands to disambiguate what the body is doing (Proske & Gandevia, 2012). This is why proprioception is best understood as perception-for-action: an internal estimate of body state, continuously updated, that motor control depends on (Tuthill & Azim, 2018).

Types of Proprioception

In the MeSH classification, Proprioception (tree F02.830.816.541) sits beneath the broader heading Sensation and has two narrower descriptors, shown in Table 1. As with any MeSH placement, this is an indexing scheme for organising the literature rather than a theory of how the sense divides into natural kinds; the two subtypes below name a submodality and a whole faculty that depends on proprioception rather than partitioning it cleanly, and they are orthogonal to the position-versus-movement distinction drawn above. Kinaesthesis does not yet have its own article on this site.

The Receptors of Body Sense

Proprioceptive signals arise from several kinds of mechanoreceptor, each transducing a different mechanical event. The most important is the muscle spindle, a small encapsulated organ of specialised intrafusal fibres lying in parallel with the working muscle. Because it is stretched whenever the muscle lengthens, its primary (Ia) afferent discharges at a rate that rises with both muscle length and the velocity of lengthening (Matthews, 1964). Spindles are not passive: their sensitivity is set by a separate fusimotor (gamma) innervation that contracts the intrafusal fibres, so the nervous system can adjust how responsive its own position sensors are, keeping them informative across the whole range of a movement (Matthews, 1964; Macefield & Knellwolf, 2018). Recording directly from single afferents in awake humans — the technique of microneurography pioneered by Vallbo and Hagbarth — confirmed that human spindles behave as the animal work predicted, firing in relation to muscle length and movement during natural tasks (Vallbo et al., 1979).

The spindle is not the only source. The Golgi tendon organ, in series with the muscle at the tendon, signals muscle force rather than length. Joint receptors in the capsule and ligaments were once thought to be the main position sensors, but they fire mostly near the extremes of joint range and contribute little in the mid-range, so they cannot be the primary signal (Ferrell et al., 1987). Finally, cutaneous mechanoreceptors in the skin around a joint are stretched as the joint moves, and this skin stretch turns out to be a genuine proprioceptive channel: stretching the skin of the hand alone can create a sensation of finger movement, and ensemble cutaneous input can evoke movement illusions (Edin & Abbs, 1991; Collins & Prochazka, 1996). Position sense at a single joint is therefore built from a weighted combination of muscle, joint, and skin signals rather than from any one receptor (Gandevia & McCloskey, 1976).

The Vibration Illusion: Evidence That Spindles Signal Movement

The decisive demonstration that muscle spindles underlie kinaesthesia came from a strikingly simple manipulation. Goodwin, McCloskey, and Matthews applied a vibrator to the tendon of a muscle — the biceps, say — while the arm was hidden from view. Vibration is a powerful stimulus for spindle primary endings, driving them to fire as if the muscle were being stretched. Subjects promptly experienced a compelling illusion that the forearm was extending, moving in the direction that would have lengthened the vibrated muscle, even though the arm was in fact still (Goodwin et al., 1972). If the arm was free to move, it slowly drifted to follow the illusion.

The experiment made three points at once. First, exciting spindle afferents in isolation is sufficient to generate a sense of movement, so those afferents must normally carry a movement signal. Second, the brain reads the extra discharge as lengthening of the vibrated muscle, showing how the percept is inferred from the pattern of afferent activity rather than sensed directly. Third, because the illusion has a direction and a velocity that track the vibration, the spindle signal is quantitatively, not just qualitatively, tied to kinaesthesia (Goodwin et al., 1972). Human microneurography later recorded the very spindle responses to vibration that the illusion implied (Burke et al., 1976), and the vibration illusion remains the standard tool for probing the muscle contribution to position and movement sense (Proske & Gandevia, 2012).

From Firing Rate to Percept

A single spindle afferent is ambiguous: a given firing rate can mean a long, relaxed muscle or a shorter, contracting one, because fusimotor drive shifts the whole relationship. The nervous system resolves this by combining the afferent signal with a copy of its own motor command — an efference copy or corollary discharge — so that the expected sensory consequences of a movement can be subtracted from the actual signal (Proske & Gandevia, 2009). This is why a limb moved passively feels different from the same limb moved voluntarily, and why the sense of effort — how hard a muscle is working — is partly central in origin, derived from the outgoing command rather than from any receptor (McCloskey, 1978).

Position sense is also a population computation. No one receptor specifies a joint angle; instead the angle is estimated from the combined activity of many muscle, skin, and joint afferents, weighted by how informative each is at that part of the range (Gandevia & McCloskey, 1976). Skin stretch contributes more at the extremes and around the small joints of the hand, muscle spindles dominate through the mid-range, and joint receptors add information near the limits (Edin & Abbs, 1991; Ferrell et al., 1987). The result is a robust estimate that degrades gracefully when any one channel is removed — but that can be fooled, as the vibration illusion shows, when a channel is driven artificially (Proske & Gandevia, 2012).

The Molecular Basis: PIEZO2

For most of its history proprioception was studied at the level of receptors and afferent nerves, with the transduction step — how a mechanical force becomes an electrical signal — left as a black box. That box was opened with the identification of PIEZO2, a mechanically gated ion channel that opens when the membrane is deformed. Work in mice showed that deleting Piezo2 from sensory neurons abolishes normal proprioception: the animals' limb coordination is severely disrupted, establishing Piezo2 as the principal mechanotransduction channel for the sense (Woo et al., 2015). The finding tied the whole edifice of proprioceptive physiology to a single, identifiable molecule.

The human case followed quickly and confirmed the picture. Rare individuals carrying loss-of-function mutations in PIEZO2 have profound deficits in proprioception and touch: they cannot reliably tell the position of their limbs with the eyes closed and rely on vision to guide movements that others make automatically, yet their sense of pain and temperature is spared (Chesler et al., 2016). These natural experiments show that PIEZO2 is necessary for human body sense, and that proprioception is dissociable from the other somatosenses at the molecular level. The muscle spindle itself depends on the channel, linking the classical organ of position sense to its molecular transducer (Kröger & Watkins, 2021).

Worked Example

How much does proprioceptive acuity matter for action? Acuity is commonly measured by joint position matching: the experimenter moves a limb to a target angle out of sight, returns it, and the person reproduces the angle with the same or the opposite limb; the average error is the acuity score (Goble, 2010; Han et al., 2016). An angular error at a joint translates into a positioning error at the end of the limb by simple geometry: for a segment of length r, an angular error of θ radians produces an endpoint error of arc length s = r × θ.

Take a matching error of 3° at the elbow with a forearm-plus-hand length of r = 0.25 m. Converting the angle to radians, θ = 3 × (π / 180) = 0.05236 rad, so the endpoint error is s = 0.25 × 0.05236 = 0.0131 m, about 13 mm. A 3° uncertainty at the elbow thus places the hand roughly 13 mm from where the person believes it to be — enough to miss a small target reached for without vision.

Now suppose targeted proprioceptive training halves the matching error to 1.5°, a realistic gain given the evidence that such training improves proprioceptive function (Aman et al., 2015). The endpoint error scales linearly with the angle, so it also halves: θ = 1.5 × (π / 180) = 0.02618 rad, giving s = 0.25 × 0.02618 = 0.0065 m, about 6.5 mm. The lesson is quantitative: because endpoint error grows with limb length, the same few degrees of joint uncertainty matter more for a long limb than a short one, and a modest improvement in acuity yields a proportionate improvement in the precision of unsighted movement — which is why proprioceptive training has measurable functional value in rehabilitation (Röijezon et al., 2015). The third demonstration lets the reader run this matching task and see the error accumulate.

Discussion

Proprioception earns its place in cognitive psychology because it is not a simple readout of receptors but a constructed estimate of body state. The peripheral signals are ambiguous: a spindle afferent's rate confounds length with fusimotor drive, joint receptors are informative only near the extremes, and no single receptor specifies a joint angle. The percept of where the body is emerges only when these signals are combined with one another and with a copy of the motor command, through an internal model of the body's mechanics (Proske & Gandevia, 2012; Tuthill & Azim, 2018). That inferential character is exactly what the vibration illusion exposes, and it is what makes proprioception a case study in perception as inference rather than transcription.

The account has open seams. How the brain weights and combines the muscle, joint, and skin channels, and how it retunes those weights as fusimotor drive and task demands change, is only partly understood (Macefield & Knellwolf, 2018). The central contribution to the senses of effort and heaviness remains debated, as does how far position sense depends on efference copy versus afference (Proske & Gandevia, 2009). And individual differences are large: proprioceptive acuity varies between people, declines with age, and is degraded by injury and neurological disease, with consequences for balance and skilled movement that a single normative model does not capture (Goble, 2010; Han et al., 2016).

Current Directions

Contemporary work is advancing on three fronts. The first is molecular and cellular: since the identification of PIEZO2 as the proprioceptive transducer, research has turned to how the channel is deployed in muscle spindles and tendon organs, how spindle function is altered in disease and ageing, and whether the pathway can be a therapeutic target (Woo et al., 2015; Chesler et al., 2016; Kröger & Watkins, 2021). The second is circuit-level: modern neuroscience is mapping how proprioceptive afferents connect to spinal and brain circuits and how those circuits build and use an estimate of limb state to guide movement, drawing the classical psychophysics together with genetics and systems neuroscience (Tuthill & Azim, 2018; Macefield & Knellwolf, 2018). The third is translational: standardising how proprioceptive acuity is measured, and establishing when and how proprioceptive training improves function, has become a priority in sports science and rehabilitation as the evidence for trainability has firmed up (Han et al., 2016; Aman et al., 2015; Röijezon et al., 2015).

Common Misconceptions

Proprioception is a sixth sense separate from the ordinary body senses.

It is one of the body senses, sharing mechanoreceptors and pathways with touch; Sherrington defined it by the source of the stimulus — the body's own tissues — not by any exotic organ (Sherrington, 1906; Proske & Gandevia, 2012).

Joint receptors are the main sensors of limb position.

Joint receptors fire mostly near the extremes of range; the muscle spindle is the principal organ of position and movement sense through the mid-range, supplemented by skin stretch receptors (Ferrell et al., 1987; Edin & Abbs, 1991).

Position sense is a direct readout of receptor activity.

The signal is ambiguous and must be interpreted: the brain combines afferent activity with a copy of the motor command, so the percept is inferred, as the vibration illusion makes vivid (Goodwin et al., 1972; Proske & Gandevia, 2009).

Glossary

Corollary discharge.

An internal copy of a motor command used to predict the sensory consequences of a movement; also called efference copy, it lets the brain distinguish self-generated from externally imposed motion.

Cutaneous mechanoreceptor.

A skin receptor sensitive to deformation; stretch of the skin around a joint provides a genuine proprioceptive signal about joint movement.

Deafferentation.

The loss of sensory (afferent) input from a body region, as in a large-fibre sensory neuronopathy; deafferented patients lose proprioception and can move only under continuous visual guidance, and even then clumsily.

Efference copy.

A duplicate of an outgoing motor command retained centrally; its comparison with incoming afferent signals supports the senses of effort and self-movement.

Fusimotor system.

The gamma motor neurons that contract the intrafusal fibres of the muscle spindle, adjusting its sensitivity so it stays informative across the full range of a movement.

Golgi tendon organ.

A receptor in series with the muscle at the tendon that signals muscle force rather than length, contributing to the sense of effort.

Ia afferent.

The large, fast primary sensory axon of the muscle spindle, whose firing rate rises with both muscle length and the velocity of lengthening.

Joint position matching.

A standard test of proprioceptive acuity in which a person reproduces a target joint angle without vision; the average reproduction error indexes acuity.

Joint receptor.

A mechanoreceptor in the joint capsule and ligaments that fires chiefly near the extremes of joint range, contributing to position sense at the limits of movement.

Kinaesthesia.

The sense of movement of the body's segments — that a limb is moving, and in which direction and how fast; the dynamic component of proprioception.

Microneurography.

A technique for recording impulses from single afferent axons in the peripheral nerves of awake humans, used to characterise spindle and cutaneous receptor responses during natural movement.

Muscle spindle.

An encapsulated sensory organ of intrafusal fibres lying in parallel with the working muscle; the principal receptor for limb position and movement sense.

PIEZO2.

A mechanically gated ion channel that transduces force into electrical signals; the principal mechanotransduction channel for proprioception and light touch.

Proprioception.

The sense of the position, movement, and force of the body's own segments, derived from receptors in muscles, tendons, joints, and skin; largely unconscious and essential for movement.

Sense of effort.

The perception of how hard a muscle is working, thought to derive partly from a central copy of the motor command rather than solely from peripheral force receptors.

Vibration illusion.

The illusory sense of limb movement produced by vibrating a muscle tendon, which artificially excites spindle afferents; direct evidence that spindles signal kinaesthesia.

Key Researchers

Benoni B. Edin. Neurophysiologist at Umeå University; he showed that cutaneous mechanoreceptors in the skin around joints signal limb movement, establishing skin stretch as a genuine proprioceptive channel. ORCID - Faculty Page

Simon C. Gandevia. Neuroscientist at Neuroscience Research Australia and the University of New South Wales; he mapped the peripheral and central signals underlying kinaesthesia, including the sense of effort and central contributions to movement perception. ORCID - Wikipedia

Jürgen Konczak. Movement scientist at the University of Minnesota; he studies proprioceptive contributions to motor learning and their rehabilitation, including systematic evidence that proprioceptive training improves motor function. ORCID - Lab Page

Peter B. C. Matthews. Physiologist at the University of Oxford; he established the physiology of the muscle spindle and its fusimotor control and showed that spindle afferents contribute to the conscious sense of limb position and movement. Wikipedia - Wikidata

Ardem Patapoutian. Molecular neurobiologist at Scripps Research and the Howard Hughes Medical Institute and 2021 Nobel laureate; he identified the PIEZO2 mechanotransduction channel and showed it to be the principal molecular sensor for proprioception. ORCID - Wikipedia

Arthur Prochazka. Neuroscientist at the University of Alberta; he analysed proprioceptive feedback during natural movement and its use in motor control and neuroprosthetics, including movement illusions from ensemble cutaneous input. Faculty Page - Google Scholar

Uwe Proske. Physiologist at Monash University; he co-authored the modern synthesis of the proprioceptive senses, establishing the muscle spindle as the principal organ of position and movement sense. ORCID - Faculty Page

Charles Scott Sherrington. Physiologist at the University of Oxford and 1932 Nobel laureate; he coined the term proprioception and described the proprioceptive system, establishing the reflex and integrative basis of the body's sense of its own position. Wikipedia - Wikidata

John C. Tuthill. Neuroscientist at the University of Washington; he leads circuit-level work on proprioception, dissecting how proprioceptive receptors and their central circuits encode limb state to guide movement. ORCID - Lab Page

Åke B. Vallbo. Neurophysiologist at the University of Gothenburg; he pioneered microneurography, recording from single afferents in awake humans to characterise muscle spindle and cutaneous receptor responses during natural movement. Wikipedia

Frequently Asked Questions

What is proprioception?
Proprioception is the sense of the position, movement, and force of one's own body segments, derived from receptors in the muscles, tendons, joints, and skin; it operates largely without awareness and is essential for movement (Proske & Gandevia, 2012).

How is proprioception different from kinaesthesia?
Proprioception is the broader category covering position, movement, and force senses; kinaesthesia refers more narrowly to the sense of movement and, by extension, limb position (McCloskey, 1978).

Which receptor is most important for knowing where a limb is?
The muscle spindle is the principal organ of limb position and movement sense through the mid-range, supplemented by skin stretch receptors and, near the extremes of range, by joint receptors (Matthews, 1964; Ferrell et al., 1987).

What does vibrating a tendon do?
Vibration excites the muscle spindle's afferents as if the muscle were lengthening, producing a compelling illusion that the limb is moving in the direction that would stretch the vibrated muscle, which is direct evidence that spindles signal movement (Goodwin et al., 1972).

What is PIEZO2 and why does it matter?
PIEZO2 is a mechanically gated ion channel that converts force into a nerve signal; it is the principal transducer for proprioception, and people who lack it have profound deficits in knowing their limb positions without vision (Woo et al., 2015; Chesler et al., 2016).

How is proprioception measured?
A common method is joint position matching, in which a person reproduces a target joint angle without vision and the reproduction error indexes acuity; several complementary methods exist (Goble, 2010; Han et al., 2016).

Can proprioception be improved with training?
Yes. Systematic reviews find that targeted proprioceptive training can improve proprioceptive acuity and motor function, which is why it is used in sports and rehabilitation (Aman et al., 2015; Röijezon et al., 2015).

Why does the sense of effort not come only from the muscles?
The perception of how hard a muscle is working depends partly on a central copy of the motor command, so effort can be felt even when peripheral force signals are ambiguous or reduced (McCloskey, 1978; Proske & Gandevia, 2009).

References