Proprioception: A New Look at an Old Concept

By Martin E. Héroux, Annie A. Butler, Lucy S. Robertson, Georgia Fisher, and Simon C. Gandevia

Proprioception, which can be defined as the awareness of the mechanical and spatial state of the body and its musculoskeletal parts, is critical to motor actions and contributes to our sense of body ownership. To date, clinical proprioceptive tests have focused on a person’s ability to detect, discriminate, or match limb positions or movements, and reveal that the strength of the relationship between deficits in proprioception and physical function varies widely. Unfortunately, these tests fail to assess higher-level proprioceptive abilities. In this Perspective, we propose that to understand fully the link between proprioception and function, we need to look beyond traditional clinical tests of proprioception. Specifically, we present a novel framework for human proprioception assessment that is divided into 2 categories: low-level and high-level proprioceptive judgments. Low-level judgments are those made in a single frame of reference and are the types of judgments made in traditional proprioceptive tests (i.e., detect, discriminate, or match). High-level proprioceptive abilities involve proprioceptive judgments made in a different frame of reference. For example, when a person indicates where their hand is located in space. This framework acknowledges that proprioception is complex and multifaceted and that tests of proprioception should not be viewed as interchangeable, but rather as complimentary. Crucially, it provides structure to the way researchers and clinicians can approach proprioception and its assessment. We hope this Perspective serves as the catalyst for discussion and new lines of investigation.

The term “proprioception” was coined by Charles Sherrington in 1907 when he stated, “In muscular receptivity, we see the body itself acting as a stimulus to its own receptors—the proprioceptors.” Early controversy revolved around whether proprioceptive signals were solely of peripheral origin, or whether central signals also contributed. After spirited debates, it is now accepted that proprioceptive signals include peripheral inputs from muscle spindles, Golgi tendon organs, cutaneous, and joint receptors, along with central inputs from efferent motor commands (i.e., corollary discharges). Together, these proprioceptive signals allow us to perceive the position and movement of our body, the force and effort generated by our muscles, and the weight of objects we lift. Although perceptions of body position and movement have been grouped and referred to as kinesthesia, proprioception encompasses all of these perceptions. Thus proprioception can be defined as the awareness of the mechanical and spatial state of the body and its musculoskeletal parts.

Proprioception contributes to our sense of ownership of our body and its parts, as well as our sense of self. Proprioception is also critical to learn, plan, execute and correct motor actions: consider the case of Ian Waterman—his acute loss of activity in all large-diameter afferents initially rendered him paralyzed. However, in other clinical conditions in which proprioceptive deficits are less extreme, the strength of the relationship between deficits in proprioception and physical function varies widely. Importantly, this relationship is associational, not causal, with other deficits that impact function and performance on proprioceptive tests often present.

The choice of proprioceptive measure is another factor that is important to understand the link between proprioception and function. Proprioception, like other senses, requires sensory signals to first be processed by dedicated brain areas before being integrated into higher-level brain functions. For other senses, clinical tests have evolved to assess both these aspects. For example, to assess hearing, an audiologist might administer a pure tone detection test or assess the ability to interpret sound and recognize speech. Similarly, to assess somatosensation, a neurologist might test 2-point tactile discrimination or assess the ability to identify objects by touch alone. But, when it comes to proprioception, clinicians and clinical researchers have generally focused on the first category of tests, those that require people to detect, discriminate, or match limb positions or movements. To understand fully the link between proprioception and function, it will be necessary to look beyond traditional tests, to those that assess higher-level proprioceptive abilities.

When healthy people are asked to detect, discriminate, or match proprioceptive stimuli, errors tend to be relatively small, with little person-to-person variability. For example, people can discriminate between 2 arm positions where the location of their index finger differs by only 1 to 1.5 cm. However, when their arm is in a similar configuration and they are instead asked to select a line from a visible ruler to indicate the location of their index finger, people make errors of several centimeters, with differences between people as large as 15 cm. These observations are in line with the view that there are 2 distinct position senses: one that codes limb position relative to the body, and one that codes limb location relative to the external world.

But, what makes a proprioceptive ability higher-level? The answer was alluded to above with the 2 types of position sense. High-level proprioceptive abilities involve proprioceptive judgments made in a different frame of reference. For example, the brain possesses dozens of spatial maps (e.g., retinotopic, somatotopic, egocentric, face-centered, object-centered, and world-centered), each with their own frame of reference (i.e., coordinate system). To ask a person to indicate the location of their index finger when their upper limb is hidden from view is a higher-level proprioceptive judgment because, to indicate the location of the index finger, for example by selecting a line on a visible ruler, coordinates of where the index finger is located relative to the body must be transformed into coordinates of where the index finger is located in the visible external world. This is different from when a person is asked to match the configuration of 1 of their arms because, in this situation, the brain simply matches the central representation of the 2 arms, it matches like with like. Jastrow came to a similar conclusion more than a century ago. With the help of several clever contraptions, Jastrow made thousands of observations on judgments of perceived width made with the eyes, the hands, and the arms. In some trials, the “receiving sense” and the “expressing sense” were the same; for example, when a person had to report the width of a grasped block with their opposite hand. In other trials, the “receiving sense” and the “expressing sense” were different; for example, when a person had to report the width of a grasped block by selecting from a series of visible lines. Based on his observations, Jastrow eloquently concluded:

“The processes involved in the above-described experiments can be represented thus: A length presented to the receiving sense makes a certain impression on my brain-center; the problem then is to reproduce the objective stimulation which shall give me an equivalent sensation. The two operations being simultaneous, the sensations can be compared and the judgment corrected until they agree. When the receiving and expressing senses are the same, the comparison is between homogeneous sensations, involving 1 brain-center; the operation is easy and the error small. When the expressing sense differs from the receiving sense, heterogeneous sensations must be compared, involving 2 brain-centers; a difficult operation with a large error. The large error seems to be due to a looseness of association between heterogeneous space-centers; it is a path of high resistance.” (Jastrow J. The perception of space by disparate senses. Mind os-XI. 1886;549. doi:10.1093/mind/os-XI.44.539.)

Traditional tests of proprioception involve a single frame of reference; thus they cannot capture deficits in higher-level proprioceptive abilities. As a first step to address this shortcoming, we propose a novel framework for human proprioception assessment (Figure). The framework is divided into 2 categories: low-level proprioceptive judgments and high-level proprioceptive judgments.

Low-level proprioceptive judgments are those made in a single frame of reference and are shown along a continuum of increasing neural complexity. They are the types of judgments made in traditional proprioceptive tests, those that require a person to detect, discriminate, or match a proprioceptive stimulus.

High-level proprioceptive judgments are those made in a different frame of reference. Versions of these tests are common in laboratory-based studies, including in psychology where they are referred to as cross-modal judgments. For example, with their upper limb resting on a table and hidden from view, a person might choose from a series of visible lines of different widths to indicate the perceived width of a grasped object, or they might report the number of milliliters in a milk carton equivalent to the perceived weight of a lifted object. The challenge will be to devise high-level proprioceptive tests that are clinically viable and functionally relevant.

This framework acknowledges that proprioception is complex and multifaceted. As such, tests of proprioception should not be viewed as interchangeable, but rather as complimentary. This makes intuitive sense as different peripheral and central neural processes and brain areas contribute to different proprioceptive senses, and more complex proprioceptive tests, for example, those that require a person to recall a previous limb position or point to the location of a hidden hand, require more complex neural computations and thus recruit additional brain areas. Moreover, impaired proprioception on a bilateral matching task is not necessarily accompanied by impaired proprioception on a unilateral matching task, nor is impaired proprioception in the upper limbs necessarily accompanied by impaired proprioception in the lower limbs. Thus, a single test of proprioception is unlikely to capture a person’s overall proprioceptive ability, a crucial point for future investigations on the link between proprioception and function.

Testing human proprioception comes with challenges. First, performance on proprioceptive and functional tests can be confounded by other deficits. For example, stroke survivors may present with cognitive, visual, motor, attention, and memory deficits, all of which can influence performance on proprioceptive and functional tests. Thus, to investigate the link between proprioception and function, these deficits should be included in causal models. Second, proprioceptive tests are tied to perceptual (i.e., conscious) judgments. Yet, these judgments do not necessarily reflect how proprioceptive signals are processed or interpreted centrally to plan, execute, or correct motor commands. Accordingly, because some proprioceptive signals are processed subconsciously to plan motor outputs, there may be a limit to what proprioceptive tests can tell us about proprioception and its causal link to function.

The proposed framework comprised of both low-level and high-level proprioception provides structure to the way researchers and clinicians could approach proprioception and its assessment. However, as with any new framework, it is necessarily tentative; it will evolve as new insights are gained, and clinicians and researchers put its logic to the test. Thus, we hope this Perspective serves as the catalyst for discussion and new lines of investigation.

This article has been excerpted from “Proprioception: a new look at an old concept,” by the same authors, which was published February 10, 2022, in the Journal of Applied Physiology. https://doi.org/10.1152/japplphysiol.00809.2021. Editing has occurred and references have been removed due to space limitations. Reprinted with permission from the American Physiological Society; all rights reserved.