Close your eyes and raise one hand above your head.
Unless something in the sensory or motor system is disrupted, you do not need a mirror to know where the hand went.
You can touch your nose in darkness. You can adjust a foot on a stair while looking forward. You can reach behind yourself for a chair, maintain a joint angle and know whether a limb is moving even when vision is unavailable.
This largely unnoticed capacity depends on proprioception.
Definition
Proprioception is the sensory processing that provides information about the position and movement of the body and its parts, especially through signals arising from muscles, tendons and joints.
It is closely tied to kinesthesia, the sense of movement. Terminology varies, and some authors use proprioception broadly enough to include both position and movement sense.
The critical point is that proprioception is not simply a conscious “feeling of the body.” It is a physiological sensory system that contributes continuously to posture, movement control and body representation.
Much of its work happens without becoming a vivid conscious sensation.
The principal receptors
Muscle spindles are among the principal receptors for proprioception.
They are specialized sensory structures within skeletal muscles that respond to changes in muscle length and to aspects of stretch dynamics. Their afferent signals provide the central nervous system with information relevant to limb position and movement.
Golgi tendon organs provide information related to muscle tension and force. Joint and cutaneous receptors can also contribute, especially in particular ranges of movement or tasks.
The system is therefore distributed.
There is no single “proprioception organ.”
Why muscle spindles matter
Muscle spindles are unusually important because the nervous system needs continuous information about the mechanical state of muscles in order to control them.
Santuz and Zampieri's 2024 review summarizes modern genetic and systems work showing how proprioceptive sensory neurons support body-position sensing and motor control.
Recent research from Frida Torell and Michael Dimitriou at Umeå University adds a particularly useful human detail. Their 2024 study showed that local pressure applied to muscle can stimulate principal proprioceptive receptors, including muscle spindles, producing measurable perceptual effects.
The result matters because it connects receptor physiology to manipulable human sensation without turning proprioception into a vague metaphor for “embodiment.”
Proprioception is not interoception
Interoception concerns sensing and integrating signals from the body's internal physiological condition, such as cardiovascular, respiratory, gastric and other visceral states.
Proprioception primarily concerns the mechanical configuration and movement of the body.
The systems interact.
A person running uphill receives proprioceptive information about joint position and muscle movement, vestibular information about head motion and balance, exteroceptive information about the slope and surface, and interoceptive information about breathing effort, heartbeat, temperature and fatigue.
Embodied experience is assembled from several sensory domains.
Calling all of them “interoception” makes the science less precise.
Proprioception is not the vestibular sense
The vestibular system, located in the inner ear, is specialized for head motion, acceleration, orientation and balance-related signals.
Proprioception supplies information from the body itself, including limbs and muscles.
The two systems cooperate extensively.
When you stand on a moving train, vestibular input indicates acceleration and head movement while proprioception tracks joint angles and muscle stretch. Vision adds optic flow and external reference points.
Balance emerges from integration, not from one sense acting alone.
Position sense and movement sense
Researchers often distinguish aspects of proprioceptive function.
Position sense concerns the perceived position or angle of a body part.
Movement sense, or kinesthesia, concerns detection of movement and its direction or velocity.
Force or effort-related information can contribute to perceived muscular state, though effort should not be treated as identical to afferent proprioceptive input because motor command and other signals also contribute.
These distinctions matter for measurement.
Someone can perform differently on a joint-position task, a threshold-to-detect-passive-motion task and a force-matching task.
One score should not be promoted into a global trait without evidence.
How proprioception is measured
Researchers use several task families.
Joint-position reproduction asks a participant to reproduce a target joint angle. Position matching asks one limb to match the position of another. Threshold-to-detect-passive-motion tasks test detection of small externally imposed movements. Some protocols assess movement direction, vibration-induced illusions or force-related judgments.
A systematic review by Horváth and colleagues found substantial heterogeneity in how proprioceptive accuracy is measured.
That is a major limitation.
Differences in equipment, joint tested, active versus passive movement, visual information, movement speed and scoring can all affect results.
“Proprioception improved” is therefore incomplete unless the measure is specified.
Vision can hide proprioceptive difficulty
Human movement is multisensory.
If one source becomes less reliable, another can compensate.
Vision is especially powerful. A person may perform a movement accurately while watching the limb yet struggle when visual feedback is removed. That does not mean vision is cheating. It means the nervous system uses available information.
Experimental proprioception tasks often remove or control vision precisely to isolate the contribution of body-position signals.
Everyday life rarely isolates them.
What loss of proprioception reveals
Rare neurological cases show how fundamental proprioception is.
When large-fiber sensory input is profoundly disrupted, people can lose reliable information about limb position and movement. Voluntary action then becomes effortful and may depend heavily on visual monitoring.
Such cases reveal something ordinary movement hides: motor control requires sensory information about the body being controlled.
Movement is not simply a command leaving the brain.
It is an ongoing loop of command, mechanical change and sensory feedback.
Proprioception and body schema
Proprioceptive information contributes to body schema, the action-oriented representation of the body's size, configuration and possibilities for movement.
Body schema is not reducible to proprioception. It also incorporates tactile, vestibular, visual and motor information.
But proprioception gives the system a continuously updated source of information about where body parts are relative to one another.
That makes it central to embodied action.
Proprioception and pain
Pain can alter movement, attention and motor strategy, and musculoskeletal conditions can coexist with changes in proprioceptive performance.
But pain and proprioception should not be collapsed.
Nociception concerns neural encoding of noxious stimuli. Pain is a subjective experience. Proprioception concerns body position and movement.
They can influence each other while remaining distinct systems and constructs.
This matters in rehabilitation, where a deficit in position sense should not be inferred merely from pain, and a pain complaint should not be dismissed as a “proprioception problem.”
Can proprioception be trained?
Specific motor and rehabilitation training can improve performance on specific sensorimotor tasks.
Practice may improve joint-position matching, balance or coordination under particular conditions. But transfer is not automatic.
A person can become better at one task because of motor learning, attention, strength, strategy or familiarity rather than a general increase in a unitary proprioceptive faculty.
A strong study therefore separates the trained task from the proposed mechanism.
It asks whether improvement persists, generalizes and survives changes in visual or contextual cues.
Why “poor proprioception” is too vague
The phrase is common in sport, rehabilitation and wellness settings.
Scientifically, it needs unpacking.
Poor on which measure? At which joint? During active or passive movement? With vision available or removed? Is the difficulty sensory, motor, attentional or strategic? Is there pain, fatigue, neurological injury or medication that changes performance?
A broad label can conceal the mechanism.
Precision does not make embodied practice less human.
It makes the claim more useful.
Proprioception across the lifespan
Proprioceptive function can change with development, aging, injury and disease.
But age effects vary by task and body region, and performance is influenced by motor strength, cognition, sensory integration and experience.
Older adults may rely differently on vision or vestibular information. Athletes and dancers may show task-specific refinements. Neurological and musculoskeletal conditions can alter sensory pathways or the way signals are used.
The field therefore benefits from avoiding a single normative story.
Embodied competence is adaptive and context-dependent.
The sensory body is not one thing
Proprioception exposes a broader conceptual problem in public discussions of “body awareness.”
When someone says they are “good at feeling their body,” they may mean they notice heartbeat, muscle tension, joint position, pain, hunger, balance or emotional arousal.
Those are not one sensory channel.
Research improves when the channel is named.
The body is not a single sensor.
The nervous system reweights signals when they disagree
Everyday movement rarely depends on proprioception alone.
Vision, touch, vestibular information, motor commands and proprioceptive input usually agree well enough that their separate contributions disappear into one coherent sense of where the body is. Experimental conflict makes the integration visible.
Close your eyes and a movement can still be guided by joint and muscle information. Add vision and position estimates often improve. Distort visual information and the nervous system may rely more heavily on proprioceptive or other cues, depending on the task and the reliability of each source.
This process is sometimes described as sensory reweighting.
The important point is not that one sense wins. The nervous system combines partially reliable information. The weighting can change with lighting, fatigue, injury, surface stability, training and the structure of the task.
That makes “proprioceptive ability” less like a single internal meter and more like one contribution to a changing multisensory estimate.
Skilled action makes proprioception easier to overlook
A violinist shifts a finger by millimeters without looking. A climber adjusts force through a foot that is partly hidden from view. A person reaches for a glass while continuing a conversation.
The smoothness of these actions can create the impression that movement is commanded from above and then executed by the body.
In fact, skilled movement depends on continuous sensory updating. Proprioceptive information about muscle length, change in length, force and joint configuration contributes to the rapid corrections that keep an intended action viable as the body and environment change.
Expertise does not eliminate sensory dependence. It changes how information is predicted, weighted and used.
This is one reason a training effect must be described precisely. Improved performance after practice may reflect better proprioceptive discrimination, better motor prediction, greater strength, improved task strategy, increased visual compensation or some combination of these. Calling every improvement “better proprioception” hides the mechanism that changed.
Measurement is local before it is global
Horváth and colleagues' systematic review of proprioceptive accuracy measures makes a methodological problem plain: the literature contains many tasks, joints, movement directions and scoring conventions.
Joint-position reproduction, threshold-to-detection of passive movement and force-matching tasks do not ask the same question.
A person can perform differently across joints and across methods. A knee result cannot simply be promoted into a whole-person proprioception score.
Measurement also changes the sensory situation. Active reproduction includes motor commands and muscular effort. Passive movement can reduce some of those contributions but introduces other constraints. Visual occlusion removes one source of compensation while also making the task less like ordinary movement.
For rehabilitation, sport and research, the implication is practical: name the task, joint, movement condition and outcome. “Proprioception improved” is too broad when the evidence actually shows that error decreased in one particular joint-position task.
Swedish research and the receptor level
Work by Filip Torell and Martin Dimitriou at Umeå University offers a useful Nordic example of why physiological detail matters. Their 2024 human experiment showed that local muscle pressure can stimulate muscle-spindle afferents, the principal receptors associated with proprioception.
The result is not a new general theory of body awareness. It is a tightly scoped physiological finding about how mechanically sensitive receptors respond.
That specificity is exactly its value.
It reminds us that a concept often discussed in broad embodied language ultimately depends on physical receptors with measurable response properties. The route from spindle discharge to coordinated action is complex, but it should not be replaced with metaphor.
Proprioception is not a personality type
Popular embodiment language sometimes turns a sensory system into a trait description: a person is said to be “highly proprioceptive,” “out of proprioception” or generally “bad at knowing where their body is.”
The scientific construct is narrower.
Proprioceptive performance depends on the receptor population, the joint or muscle examined, the movement, the task, the presence of vision and the scoring method. It can also be affected by neurological injury, peripheral sensory loss, pain, fatigue and aging. None of this maps neatly onto a single personality-like dimension.
That boundary protects both science and people. A clumsy moment is not proof of a proprioceptive deficit. A strong yoga or dance practice does not establish superior proprioceptive accuracy across the body. Feeling embodied is psychologically meaningful, but it is not a substitute for sensory measurement.
This is precisely where a rigorous embodied science becomes more humane. It resists turning technical language into a new way of judging the person.
What the evidence does not show
Current evidence does not show that:
- proprioception is the same as interoception;
- proprioception is the same as balance;
- one joint-position task measures a universal proprioceptive trait;
- high athletic skill proves superior proprioception across all measures;
- pain automatically means impaired proprioception;
- more conscious attention to movement always improves proprioceptive function;
- a person can accurately diagnose a proprioceptive deficit from subjective feeling alone;
- proprioception resides in one receptor, one nerve or one brain region;
- training gains necessarily generalize across joints and contexts;
- “embodiment” can be used as a scientific synonym for proprioception.
Institute synthesis: invisible information for action
Institute synthesis: proprioception is a useful example of a human capacity that is essential precisely because it is often quiet.
We do not need every joint angle to become conscious.
The system works by making enough bodily state information available to support action.
For Inner Technology, the relevant principle is not maximal awareness. It is reliable access to the right signal at the right level.
Sometimes action needs conscious correction. Sometimes automatic sensorimotor control is better left alone.
This is an Institute framing, not an established proprioception scale.
From individual bodies to designed environments
Buildings, tools and interfaces assume bodies.
Stair dimensions, chair heights, vehicle controls, prosthetics, haptic systems and virtual reality environments all interact with proprioceptive expectations.
When digital systems alter visual feedback without matching body movement, the mismatch can change performance and comfort. When robotic or prosthetic systems restore movement without adequate sensory feedback, control may remain cognitively demanding.
Embodied intelligence therefore belongs in design as much as in neuroscience.
A system that understands human action has to understand that movement depends on sensing the moving body.
Editorial limitation note
Proprioceptive difficulty can arise from neurological, musculoskeletal or sensory causes and should not be self-diagnosed from general clumsiness or discomfort. Persistent or sudden changes in body-position sense, coordination or balance can warrant professional assessment.
Research references
- Santuz A, Zampieri N. Making sense of proprioception. Trends in Genetics. 2024;40(1):20-23. https://doi.org/10.1016/j.tig.2023.10.006
- Proske U, Gandevia SC. The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiological Reviews. 2012;92(4):1651-1697. https://doi.org/10.1152/physrev.00048.2011
- Horváth Á, Ferentzi E, Schwartz K, Jacobs N, Meyns P, Köteles F. The measurement of proprioceptive accuracy: A systematic literature review. Journal of Sport and Health Science. 2023;12(2):219-225. https://doi.org/10.1016/j.jshs.2022.04.001
- Torell F, Dimitriou M. Local muscle pressure stimulates the principal receptors for proprioception. Cell Reports. 2024;43(9):114699. https://doi.org/10.1016/j.celrep.2024.114699
- Donato R, et al. The mechanosensitive ion channel ASIC2 mediates both proprioceptive sensing and spinal alignment. Experimental Physiology. 2023. PMID: 36951012. https://pubmed.ncbi.nlm.nih.gov/36951012/
- Mao D, Gu Y. Multisensory coding of self-motion and its contribution to navigation. Nature Reviews Neuroscience. 2025;26:715-732. https://doi.org/10.1038/s41583-025-00970-x