1Department of Rehabilitation Medicine, Seoul National University Hospital, Seoul, Korea
2Department of Rehabilitation Medicine, Seoul National University College of Medicine, Seoul, Korea
3Department of Human Systems Medicine, Seoul National University College of Medicine, Seoul, Korea
4Interdisciplinary Program in Bioengineering, The Graduate School, Seoul National University, Seoul, Korea
5Institute of Convergence Medicine with Innovative Technology, Seoul National University Hospital, Seoul, Korea
6Institute on Aging, Seoul National University, Seoul, Korea
Correspondence: Han Gil Seo Department of Rehabilitation Medicine, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul 03080, Korea. Tel: +82-2-2072-1659 Fax: +82-2-6072-5244 E-mail: hangilseo@snu.ac.kr
• Received: July 21, 2025 • Revised: December 3, 2025 • Accepted: January 6, 2026
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
To investigate and compare proprioception characteristics in healthy adults using different measurement methods.
Methods
Participants engaged in three tasks using a device developed to assess elbow joint proprioception: the ipsilateral matching task (IMT), contralateral matching task (CMT), and pointing task (PT). Each task was performed three times at different angles (30°, 50°, and 70°) in a randomised order for nine trials and tested for both the right and left arms. Activity level was measured using the International Physical Activity Questionnaire-Short Form.
Results
Twenty healthy adults (10 males, mean age: 39.80±12.19 years) were enrolled. The absolute error of the IMT was significantly lower than that of the CMT and PT (Bonferroni [Bonf.] p=0.029 and 0.016, respectively). PT showed a higher variable error than that of IMT and CMT (Bonf. p<0.001, and 0.040, respectively). There were no significant differences in errors across tasks based on sex or age. The total International Physical Activity Questionnaire-Short Form score showed statistically significant correlations with the absolute error (r=-.460, p=0.041) and constant error (r=-.469, p=0.037) of the CMT and variable error of the PT (r=-.478, p=0.033).
Conclusion
This study demonstrated that different proprioceptive tasks can assess the unique characteristics of proprioceptive function. The IMT produced lower error values than those of the CMT and PT, with the PT exhibiting higher variability. These differences may stem from distinct mechanisms that depend on the nature of each task and warrant further investigation.
Proprioception refers to the sense of joint and body movement, as well as the spatial position of the body or its parts [1]. As a somatosensation, proprioception involves the passive reception of sensory signals from peripheral mechanoreceptors, including muscle spindles, Golgi tendon organs, and cutaneous and joint receptors. Information on the force and effort generated by muscle activity, as well as the weight of external objects, provides essential input for proprioception [2]. Additionally, proprioceptive processing relies on central mechanisms through corollary discharge, such as attention and memory, which integrate sensory feedback with efferent motor commands [3]. Proprioception plays a crucial role in motor control, motor learning, and functional recovery. Loss of proprioception can result in sensory ataxia, balance impairment, and loss of coordination [4]. In cases of proprioceptive deficits, individuals depend on visual compensation, which further exacerbates the functional impairments associated with proprioceptive loss.
Approximately 50% of patients with stroke experience a decline in proprioception [5] and struggle with accurate motor control because of both muscle weakness and a diminished ability to utilise proprioceptive feedback as a result of the stroke [6]. Furthermore, it has been demonstrated that proprioceptive function in patients with stroke is closely linked to functional recovery following constraint-induced movement therapy or robotic rehabilitation [7-9]. This suggests that proprioception is indispensable for achieving optimal functional recovery in poststroke therapy, particularly in high-intensity repetitive training that induces Hebbian plasticity [8]. Moreover, proprioception is impaired not only in stroke but also in other conditions, such as traumatic brain injury and degenerative neurological diseases, including Parkinson’s disease [10,11]. Balance impairment is common in patients with traumatic brain injury, prompting research on the sensory systems involved, including proprioception. Proprioceptive deficits are particularly prevalent among patients with concussion, with studies reporting that 38% of such patients with balance impairments primarily exhibit proprioceptive impairments rather than visual and vestibular dysfunction [10]. Upper-limb sensory training has been reported to enhance both proprioception and motor function in these patients [12,13]. Therefore, it is imperative to develop reliable assessment tools with established validity to objectively evaluate and enhance proprioceptive function in patients with neurological disorders.
However, methods for assessing proprioception remain ambiguous [14]. Given the involvement of such complex neurophysiological mechanisms, there is no standardised method for quantitatively assessing proprioceptive accuracy. Among the various assessment tools, joint position reproduction (JPR) and threshold to detection of passive motion (TTDPM) are widely used by clinicians and researchers [1,15]. JPR is a method for evaluating proprioceptive accuracy by assessing an individual’s ability to replicate a target joint position, often requiring motor control, cognitive function, and interhemispheric communication. JPR can be administered in active or passive modes, using either ipsilateral or contralateral matching, and may also involve memory retention to reproduce the joint position. TTDPM assesses proprioceptive sensitivity without engaging the motor system, as participants detect subtle passive movements. Each proprioceptive assessment technique evaluates distinct neurophysiological aspects of proprioception, highlighting the multiple facets of sensory and motor integration necessary for accurate proprioceptive function [1,15]. However, the existing equipment for the quantitative assessment of proprioception is costly and requires complex setups, posing considerable limitations in both clinical and research environments.
Recent studies have further suggested that human position sense encompasses two distinct modalities: one that relies on signals from muscle spindles to perceive the relative position of one body part to another, and another that uses exteroceptive signals, such as visual, tactile, and auditory cues, to perceive the position of the body within external space [16,17]. These distinctions highlight that different proprioceptive tasks can involve unique central processing pathways and may rely on various sensory inputs, such as muscle spindle activity for bilateral matching and additional visual or external cues for pointing. However, the extent to which these differences manifest and the precise mechanisms underlying them remain unclear.
In this study, we developed a portable and user-friendly tool to assess distinct proprioceptive functions and evaluated its reliability. The study aim was to investigate and compare the characteristics of proprioception in healthy adults across different proprioceptive tasks.
METHODS
Study design and participants
This was an exploratory cross-sectional pilot study. The study protocol was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. 2306-136-1440, approved on 30 June 2023). This study was conducted in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki.
The inclusion criteria for healthy adults were as follows: (1) aged 19 years or older, and (2) those who voluntarily consented in writing and agreed to participate in the study. The exclusion criteria were as follows: (1) individuals with neurological or orthopaedic conditions that may affect motor or sensory function in the upper extremities, and (2) individuals with comorbidities that may hinder study participation based on the researcher’s judgment. The participants were recruited between August 2023 and September 2023, with the last follow-up date being 18 September 2023. Written informed consent was obtained from all participants.
Device for proprioception assessment
The system consists of two main bodies, each designed to measure the elbow joint angle of one arm (Fig. 1). Each main body includes a silicone pad to support the elbow, a 3D-printed handle, and two adjustable aluminium frames. The frames were designed to be adjusted according to the length of the forearm. A rotary encoder (SME360AP-05DP-XY; SERA) was used to measure the angles at the elbow joint area of each body, and a crown gear was used for angle fixation. The rotary encoder is an absolute-type encoder with magnetic detection and a resolution of 0.08 degrees. The crown gear, which was designed with 30 teeth, allowed joint fixation angle adjustments in 12-degree increments. The base to which the two main bodies were fixed featured a width-adjustment rail, enabling it to be adjusted and secured according to the participant’s shoulder width. Signals from the two rotary encoders were collected using a microcontroller (Arduino Mega; Arduino) and processed on a connected laptop through a graphical user interface developed using Processing 4. This interface enabled real-time monitoring of the angles of both elbow joints, as well as the difference between the two angles, and allowed data storage.
Experimental design
The participants were seated in an upright position with their hip and knee joints at 90°. The chair height was adjusted to ensure that this position was maintained while the elbow was resting on the table. The apparatus was calibrated to accommodate the arm and shoulder widths of each participant.
The participants engaged in three JPR tasks using the device that we developed to assess elbow joint proprioception: the ipsilateral matching task (IMT), contralateral matching task (CMT), and pointing task (PT) (Fig. 2). In the IMT, with the participant’s vision occluded, the examiner positioned the reference elbow joint at a specific angle, held it in place for 5 seconds, and subsequently reset it to 0°. In this study, 0° represented the mechanical zero defined by the device, indicating a standardized starting posture. In this posture, the elbow and forearm were aligned with the apparatus frame and the forearm was positioned horizontally (parallel to the table surface), and this position did not correspond to anatomical full extension. The participant was then instructed to actively reposition the limb to match the angle of the same limb. The CMT followed a similar setup in which the examiner, with the participant wearing an eye mask, adjusted the reference elbow of one limb to a set angle. The participants attempted to match this angle using the opposite limb. For the PT, the screen obscured the participant’s view of the reference upper arm. The examiner positioned the joint at a predetermined angle, while the participant used a lever on an external device with the opposite arm to reproduce the target angle as accurately as possible.
The target angles for each task were set at 30°, 50°, and 70° to allow physiologically meaningful testing conditions. Mid-range elbow flexion angles minimize the contribution of capsuloligamentous receptors, which predominantly respond at the extremes of joint range, while enhancing afferent input from muscular receptors that are most active during functional movements [15,17,18]. Angles greater than 90° were avoided because contact between the upper arm and forearm may introduce cutaneous cues that artificially facilitate position sense and confound proprioceptive assessment. These angles correspond to flexion ranges frequently used in daily activities such as reaching and object manipulation, thereby increasing the functional relevance of the assessment [19].
Target angles were practiced once before assessment. Three trials were conducted for each angle, totalling nine trials per set, with the angles arranged in random order within each set. Additionally, the IMT, CMT, and PT tasks were administered in a randomised sequence across 20 participants to ensure balanced order effects. All tasks were performed on both the right and left sides. In all three tasks, once the participant indicated they had achieved the target angle, the examiner recorded the measurement by pressing the “REC” button on a laptop program linked to the device.
Outcome measures
The JPR position error was assessed using three outcome measures [14]. The absolute error (AE) represents the magnitude of the difference between the reference and matched angles, disregarding the direction of the error, such as under- or overshooting. The constant error (CE) represents the signed difference between the matched and target angles (matched-target), with positive values indicating overshooting (i.e., reproducing a greater elbow flexion angle than the target) and negative values indicating undershooting (i.e., reproducing a smaller elbow flexion angle than the target). The variable error (VE) is defined as the standard deviation of the CE, reflecting the stability of repositioning errors and indicating whether participants consistently maintained their error range irrespective of the error magnitude.
The Edinburgh Handedness Inventory is a screening tool for assessing handedness that consists of 10 items [20]. The score on the Edinburgh Handedness Inventory was calculated by taking the difference between the number of items responded to with ‘right hand’ and ‘left hand,’ dividing by the total number of items, and then multiplying by 100. A score of -40 or below was interpreted as indicating left-handedness, +40 or above as right-handedness, and scores in between as ambidextrous [21]. The activity levels of the participants were measured using the International Physical Activity Questionnaire-Short Form (IPAQ-SF) to examine whether physical activity level was associated with proprioceptive performance across tasks [22]. The IPAQ is a self-reported questionnaire that evaluates vigorous-intensity activity, moderate-intensity activity, walking, and sitting over the past 7 days [23]. The total IPAQ-SF score (MET-min/week) was calculated as follows: (vigorous activity×8)+(moderate activity×4)+(walking×3.3).
Statistical analysis
Baseline characteristics and assessment data were expressed as the mean and standard deviation for continuous variables. The relative reliability of the three repeated measurements was assessed by calculating the intraclass correlation coefficient (ICC). Repeated measures analysis of variance with Bonferroni [Bonf.] post hoc tests was used to evaluate the characteristics and variability of position errors across the three tasks. A paired t-test was used to examine differences between the right and left hands of a subgroup of right-handed participants. Pearson’s correlation analysis was conducted to examine the relation between the IPAQ-SF scores and position error values. A p-value of <0.05 was considered statistically significant. SPSS Statistics 26.0 for Windows (IBM Corp.) was used for all analyses.
RESULTS
The participants’ characteristics are presented in Table 1. Twenty healthy adults (10 males, mean age: 39.80±12.19 years) were enrolled and completed the study process. Only one participant scored -40 on the Edinburgh Handedness Inventory (left-handedness), and 19 participants were right-handed. Relative reliability was the highest for CMT (ICC=.716, p<0.001), followed by IMT (ICC=.566, p<0.001) and PT (ICC=.495, p<0.001) (Table 2).
Table 3 shows position errors in the IMT, CMT, and PT. Across all position errors, the AE and VE were lower in the IMT (4.10±1.22° and 4.16±0.89°) compared with that in CMT (5.29±1.80° and 4.96±0.93°) and PT (5.29±1.55° and 6.02±1.46°). The AE of the IMT was statistically significantly lower than those of the CMT and PT (Bonf. p=0.029 and .016, respectively) (Table 4). Overall, error magnitude was lower at 70° than at 30° and 50°. PT (5.53±1.88°) showed higher VE compared with that of IMT (4.16±0.89°) and CMT (4.96±0.93°) (Bonf. p=0.000 and 0.040, respectively). Conversely, the CE of the PT (0.79±5.29°) was the lowest among the three measurements. However, caution is warranted in interpreting CE, as it represents the mean of values with directional bias. There were no significant differences in errors across tasks based on sex or age groups. The total IPAQ score was significantly correlated with the AE of the CMT (r=-.460, p=0.041) and the VE of the PT (r=-.478, p=0.033) (Table 5).
In a subgroup of 19 right-handed participants, the differences in position errors between the right and left sides were analysed (Supplementary Table S1). For the IMT, VE was significantly greater on the left side (4.40±1.27°) compared to the right side (3.73±0.85°) (p=0.036). In the CMT, the right side exhibited significantly larger position errors than the left side across the AE, VE, and CE (p=0.002, 0.021, and 0.003, respectively).
DISCUSSION
This study validated the reliability of proprioception measurements using a newly developed device and examined proprioceptive characteristics across various proprioceptive tasks in healthy adults. The measurement tool demonstrates fair to good relative reliability [24]. In the IMT, both AE and VE were lower than those in the CMT and PT. The PT exhibited the highest VE, indicating substantial variability in this task, which may be attributed to the distinct mechanisms underlying the PT compared with the matching tasks. Additionally, physical activity level demonstrated a significant negative correlation with the AE of the CMT and the VE of the PT. In the subgroup analysis of right-handed participants, the dominant hand (right hand) showed lower error values than those of the non-dominant hand during the IMT. In the CMT, high error values for AE, VE, and CE were observed when the dominant hand served as the reference, suggesting differences in proprioceptive capacity between the dominant and non-dominant hands.
Elbow proprioception is essential for performing various activities of daily living by enabling precise joint positioning and adaptive movement control, which support functional tasks, such as reaching, lifting, and handling objects safely and effectively [25]. The elbow has been widely used in proprioception research, as its primarily single-axis hinge motion allows joint angles to be controlled and reproduced with high precision while minimizing the influence of compensatory movements that are more prominent in multi-axial joints such as the shoulder [26,27]. Compared with the wrist and fingers, where cutaneous inputs contribute substantially to position sense, the elbow enables clearer evaluation of muscle-spindle–based signals, particularly at mid-range angles where capsuloligamentous input is limited [26,28]. Furthermore, impairments in elbow proprioception have been reported to produce greater disturbances in endpoint positioning during upper-limb tasks than deficits at the shoulder, underscoring the elbow’s sensitivity as an indicator of sensorimotor performance [29]. These features collectively support the suitability of the elbow joint as a valid and reliable model for investigating joint position sense.
Our developed device was designed to enhance clinical portability and cost-effectiveness, offering a practical alternative to complex laboratory setups or expensive isokinetic dynamometers often restricted to research facilities. The system features independent measurement units for each arm equipped with high-resolution digital encoders, enabling precise, real-time quantification of proprioceptive error in various environments without requiring extensive calibration or space. Importantly, despite this focus on clinical utility, our newly developed tool demonstrated commendable reliability and yielded error values comparable to those reported in similar studies. In studies measuring AE for elbow proprioception using the IMT, values were reported as 3.30° to 7.00° [19], 6.10° [30], 5.60° [31], and 4.00° [32], which are comparable to the 4.10±1.22° observed in our study. The AE values for the CMT and PT in our study, 5.29±1.80° and 5.29±1.55°, respectively, were similar to those reported in other studies, which observed AE values of 6.70° and 4.90° [16].
Current methods for assessing proprioception lack standardisation and, depending on the approach used, evaluate distinct characteristics of proprioceptive functions [16,17,33]. Our results demonstrated that the IMT showed superior performance compared to the CMT and PT, suggesting that the IMT may be less susceptible to the effects of thixotropy as it relies primarily on memory [33]. Thixotropy, the property by which muscle spindles adjust discharge rates based on previous muscle states, has a stronger impact on tasks, such as CMT and PT, which utilise real-time sensory feedback [33]. Unlike matching tasks that process relative limb positions, PT is based on perceiving the limb position within external space [17]. In our study, the VE in the PT was significantly higher than that in the matching tasks, potentially because of the distinct mechanisms underlying the PT compared with the matching tasks. Previous studies have reported that, in PT tasks, errors tend to increase when participants take longer to determine their response, whereas CMT errors appear relatively unaffected by decision time [17,33]. In our study, the participants were not restricted by a time limit for the PT, which may have contributed to the relatively high VE observed. Additionally, PT requires more complex and integrated sensory processing, including visual information, than matching tasks, likely resulting in a higher VE for the PT [17,33].
While physical activity level was not significantly correlated with IMT error values, it showed a significant negative correlation with the AE of the CMT and VE of the PT. IMT relies on proprioceptive information generated through identical mechanoreceptors on a single limb and requires additional cognitive effort, such as memory-based central processing [1]. By contrast, both CMT and PT involve bilateral proprioceptive inputs and motor function, relying on interhemispheric processing to complete tasks with minimal reliance on memory or other cognitive demands. Individuals with higher physical activity levels may have increased interhemispheric connectivity [34] or relatively enhanced motor control function, which could explain the observed correlations [35].
Our results showed a tendency for error values to increase as the angle decreased, with higher error values observed at 30° and 50° than at 70° across all three tasks. This finding aligns with previous research indicating that angle sensing is less accurate at lower angles near full elbow extension [19]. Although our study and previous study primarily tested proprioception during flexion, similar trends have been observed in experiments measuring proprioception during elbow extension [26]. These results suggest that proprioceptive accuracy may decrease owing to reduced input data, such as diminished muscle spindle signalling and decreased cutaneous sensory input, as the elbow approaches full extension. However, a study conducted at test angles of 5°, 35°, 65°, 95°, and 125° reported that position errors were lower at extreme angles, such as 5° and 125°, compared with those at mid-range angles [33], likely because of the increased involvement of joint receptors, which are activated at end ranges to provide signals about joint position and movement [28].
This study had several limitations. First, the sample consisted of young and middle-aged adults (five participants in each age group between their 20s and 50s); a broader age range, including older individuals, would provide more comprehensive insights into proprioceptive characteristics across the lifespan. Second, this study focused on evaluating active JPR to assess sensorimotor integration capabilities. Given that proprioception is a complex construct that also encompasses passive motion sense and detection thresholds, the exclusion of passive JPR or TTDPM comparators limits the construct validity regarding the purely afferent sensory aspects of proprioception. Consequently, our results should be interpreted as reflecting the active, functional aspects of limb positioning involving motor efference copies, rather than the full spectrum of proprioceptive function. Finally, the tasks used in this study predominantly involved motor functions, potentially confounding the results and making it challenging to assess pure proprioceptive function independently of motor influence.
In conclusion, this study validated the reliability of a newly developed proprioception assessment tool and demonstrated that different proprioceptive tasks capture distinct characteristics of proprioceptive function. Our findings highlighted that the IMT yielded lower error values than those of the CMT and PT, which may be attributed to its reduced susceptibility to thixotropy and greater reliance on memory. The higher variability observed in the PT may stem from its requirement for complex, integrated sensory processing, including visual input. These results underscore the importance of using multiple assessment methods to reflect diverse aspects of proprioception and contribute to a more nuanced understanding of proprioceptive functions in healthy adults.
CONFLICTS OF INTEREST
Byung-Mo Oh is the Editor-in-Chief and Han Gil Seo is an Associate Editor of Annals of Rehabilitation Medicine. The authors did not engage in any part of the review and decision-making process for this manuscript. Otherwise, no potential conflict of interest relevant to this article was reported.
FUNDING INFORMATION
This study supported by grant no. NTRH RF-2023004 from the MOLIT Research Fund.
AUTHOR CONTRIBUTION
Conceptualization: Yun SJ, Oh BM, Seo HG. Data curation: Yun SJ, Seo HG. Formal analysis: Yun SJ, Seo HG. Investigation: Yun SJ, Seo HG. Methodology: Yun SJ, Park S, Seo HG. Software: Park S. Project administration: Yun SJ, Seo HG. Resources: Seo HG. Funding acquisition: Seo HG. Supervision: Oh BM, Seo HG. Writing – original draft: Yun SJ. Writing – review & editing: Park S, Oh BM, Seo HG. Approval of final manuscript: all authors.
DATA AVAILABILITY STATEMENT
The datasets used and/or analysed in the current study are available from the corresponding author upon reasonable request.
3. Proske U, Gandevia SC. The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiol Rev 2012;92:1651-97.
5. Meyer S, De Bruyn N, Lafosse C, Van Dijk M, Michielsen M, Thijs L, et al. Somatosensory impairments in the upper limb poststroke: distribution and association with motor function and visuospatial neglect. Neurorehabil Neural Repair 2016;30:731-42.
7. Park SW, Wolf SL, Blanton S, Winstein C, Nichols-Larsen DS. The EXCITE Trial: predicting a clinically meaningful motor activity log outcome. Neurorehabil Neural Repair 2008;22:486-93.
8. Rowe JB, Chan V, Ingemanson ML, Cramer SC, Wolbrecht ET, Reinkensmeyer DJ. Robotic assistance for training finger movement using a Hebbian model: a randomized controlled trial. Neurorehabil Neural Repair 2017;31:769-80.
10. Fino PC, Dibble LE, Wilde EA, Fino NF, Johnson P, Cortez MM, et al. Sensory phenotypes for balance dysfunction after mild traumatic brain injury. Neurology 2022;99:e521-35.
11. Konczak J, Corcos DM, Horak F, Poizner H, Shapiro M, Tuite P, et al. Proprioception and motor control in Parkinson's disease. J Mot Behav 2009;41:543-52.
12. Abbruzzese G, Trompetto C, Mori L, Pelosin E. Proprioceptive rehabilitation of upper limb dysfunction in movement disorders: a clinical perspective. Front Hum Neurosci 2014;8:961.
13. Elangovan N, Tuite PJ, Konczak J. Somatosensory training improves proprioception and untrained motor function in Parkinson's disease. Front Neurol 2018;9:1053.
14. Horváth Á, Ferentzi E, Schwartz K, Jacobs N, Meyns P, Köteles F. The measurement of proprioceptive accuracy: a systematic literature review. J Sport Health Sci 2023;12:219-25.
16. Tsay AJ, Giummarra MJ. Position sense in chronic pain: separating peripheral and central mechanisms in proprioception in unilateral limb pain. J Pain 2016;17:815-23.
19. Tsai YJ, Tsai TC, Chung TC, Hsiao CK, Tu YK. Assessment of joint position sense in active and passive modes with various elbow flexion angles and movement speeds using an isokinetic dynamometer. Musculoskelet Sci Pract 2024;73:103153.
21. Oh JY, Hwang NR, Hwang TY. Short-term response pattern of serum brain-derived neurotrophic factor to repetitive transcranial magnetic stimulation and brain neuroplasticity in patients with chronic schizophrenia. J Korean Assoc Soc Psychiatry 2017;22:47-56.
22. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International Physical Activity Questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003;35:1381-95.
23. Lee PH, Macfarlane DJ, Lam TH, Stewart SM. Validity of the International Physical Activity Questionnaire Short Form (IPAQ-SF): a systematic review. Int J Behav Nutr Phys Act 2011;8:115.
27. Elangovan N, Herrmann A, Konczak J. Assessing proprioceptive function: evaluating joint position matching methods against psychophysical thresholds. Phys Ther 2014;94:553-61.
30. Juul-Kristensen B, Lund H, Hansen K, Christensen H, Danneskiold-Samsøe B, Bliddal H. Test-retest reliability of joint position and kinesthetic sense in the elbow of healthy subjects. Physiother Theory Pract 2008;24:65-72.
32. Hattori Y, Doi K, Sakamoto S, Yukata K, Shafi M. Elbow joint position sense following brachial plexus palsy treated with double free muscle transfer. J Hand Surg Am 2009;34:1667-73.
33. Roach C, Love C, Allen T, Proske U. The contribution of muscle spindles to position sense measured with three different methods. Exp Brain Res 2023;241:2433-50.
34. Riedel D, Lorke N, Mierau A, Strüder HK, Wolf D, Fischer F, et al. The relationship between interhemispheric transfer time and physical activity as well as cardiorespiratory fitness in healthy older adults. Exp Gerontol 2023;176:112167.
35. Hübner L, Voelcker-Rehage C. Does physical activity benefit motor performance and learning of upper extremity tasks in older adults?–a systematic review. Eur Rev Aging Phys Act 2017;14:15.
