A novel sensor-based real-time tracking game for assessing ankle proprioceptive effects on sensorimotor performance Authors: Monica Hruzd, Rylie Watson, Mohammad Hosseinalizadeh, Kubra Akbas, Nima Toosizadeh Published: 2025 ABSTRACT AbstractProprioceptive deficits have been linked to balance and risk of falling in older adults. In this study, we assessed the effects of proprioceptive deficits on sensorimotor performance using a sensor-based real-time tracking game (SRT) that utilizes a moving target on a screen. The current methods to measure proprioceptive deficits can be difficult to use and may not be sensitive to minor deficits. We recruited 19 young participants (mean age = 21.8 ± 1.32, 55% female) and 10 older adults (age = 73.6 ± 7.73, 60% female). We placed a gyroscope on the top of the intermediate cuneiform bone to track movements. Both legs were tested with three separate SRT tracks with different difficulties. To disturb visuo-proprioceptive performance, we applied vibration to the tibialis anterior, peroneus longus, soleus, and gastrocnemius. We hypothesized that younger participants would score better than older participants, the dominant ankle would perform better than the non-dominant ankle, and participants would perform better without vibratory stimulation. We measured the amplitude and directional accuracy of the performance. Amplitude accuracy was measured by calculating the straight-line distance from the tracker to predetermined track. Directional accuracy was defined as the percentage of time the participant remained in the free zone (defined by average performance among the young sample) at every instant of time. Results showed significant effects for amplitude and directional accuracy by age (p < 0.001), vibration condition (p < 0.012), and testing side (p < 0.026). Our results showed that SRT is sensitive enough to measure effects of visuo-proprioception changes caused by aging, dominant vs. non-dominant sides, and outside disturbance. Through accurate detection of visuo-proprioceptive deficits, we may potentially identify risk of falling associated with such deficits. Introduction One out of every four older adults fall every year in the US, leading to devastating consequences, such as fractures and brain injuries [24]. In addition to lower limb muscle strength assessment and training, which are the primary research focus areas for addressing fall risk in older adults, proprioceptive assessment and improvement are just as crucial, since they can be major contributing factors to postural balance and poor balance recovery after tripping [3]. Proprioception is defined as the ability to sense position and movement of limbs, as well as the senses of force, effort, and heaviness [37]. In the current study, the focus was on the kinesthetic senses of proprioception, such as the ability to sense position and movement of body segments using information from muscle spindles, Golgi tendon units, and joint receptors [4, 18, 38]. With aging, the efficacy of these sensory components diminishes, due to the natural decline in neural fibers and demyelination, which can lead to a gradual reduction in proprioceptive performance in older adults [10, 18, 33, 42]. This deterioration in proprioceptive performance may lead to an elevated risk of fall in some older adults [49]. Therefore, it is critical to accurately measure proprioceptive performance as a core component of balance to identify those who can potentially benefit from rehabilitation programs targeting proprioceptive deficits. Current conventional methods of evaluating ankle proprioception out of a variety of different testing protocols, are joint position sense (JPS), kinesthesia, and force sense, which are aimed to analyze proprioceptive senses. JPS testing involves assessing proprioception by evaluating the ability to perceive and replicate joint positioning without visual input [28]. In kinesthesia testing, the ankle joint is moved very slowly, often using a robotic device, and the individual must indicate the exact instant they perceive movement [28]. JPS also assesses kinesthesia, as kinesthesia refers to a person’s recognition of the position and movement of a body segment [38]. Similarly, force sense testing incorporates replicating a previously perceived moment in a joint, such as the ankle joint [26]. Force sensing, which is not limited to the ankle joint, includes the ability to perceive a force applied to a joint with visual feedback and replicate this particular force after removing the visual feedback [15]. There are some limitations for the above approaches; it has been argued in several reviews that such measurements are not sensitive enough to capture minor deficits in proprioception, are not strongly associated with balance performance, can be partially subjective, and require customized platforms for assessments (e.g., Biodex) [27, 45, 56]. Therefore, there is a need for more accurate and scalable approaches of assessing proprioceptive performance feasible for older adults, especially those who are at high risk of fall. In the current study, we developed and evaluated a sensor-based real-time tracking game (SRT) for assessing the effects of visuo-proprioceptive performance of the ankle joint on sensorimotor function. Proprioception in this approach assesses the way the body senses its position and movement through the use of mechanoreceptors located in the muscles and joints of the lower limbs [18]. We believe that SRT can be highly influenced by proprioceptive acuity. While motor and sensory performance is interconnected in most physical functions, we used SRT as a method to capture the alterations in sensorimotor performance that are caused by differences in proprioception. SRT incorporates tracking a flying target by moving the ankle joint in the plantarflexion/dorsiflexion (up/down) and abduction/adduction (right/left) directions within a visuo-proprioceptive sensor-based feedback mechanism. By measuring the amplitude and directional accuracy outcomes during the tracking, effects of ankle proprioceptive performance on sensorimotor function is assessed based on a correction mechanism of tracking error through the short-latency reflexive responses that activate parent muscle to restore joint positions and long-latency adjustment within the central nervous system to direct orientation of limbs [20, 22]. Of note, our test combines visual feedback, without the need of complex equipment. Accordingly, the SRT approach is designed to be feasible for both in-clinic and in-home applications, incorporating one inexpensive motion sensor on the foot (~ $10) and a tablet or monitor and no requirement for additional platforms. Our hypotheses were: a) SRT would provide a high test–retest reliability; b) SRT would be sensitive enough to capture differences in sensorimotor performance caused by proprioceptive differences in the dominant vs. non-dominant foot; c) SRT performance would decline when muscles/tendons are disturbed by outside vibratory disturbance; and d) due to aging effects, SRT performance would be worse among older adults compared to younger participants. Results Participants The demographic information for the participants is presented in Table 1. Nineteen young control participants (8 males and 11 females, age = 21.8 ± 1.32 years), and 10 older adults (4 males and 6 females, age = 73.6 ± 7.73 years) were recruited. None of the older adult participants had any severe clinical condition that would exclude them from the study, such as dizziness, risk of falling or pain in lower extremities. Of note, all participants were right foot dominant. Table 1Demographics and clinical measuresDemographicsYoung Controls (n = 19)Older Adults (n = 10)p valueEffect SizeAge, years (SD)21.8 (1.32)73.6 (7.73) < 0.000111.282Male, n (% of the group)8 (40%)4 (45%)0.9970.020Weight, kg (SD)72.67 (14.03)74.8 (19.09)0.7600.133Height, m (SD)170.89 (9.21)171.6 (11.87)0.8710.070BMI (SD)24.75 (4.20)25.20 (4.78)0.8510.120Clinical measuresYoung controls (n = 19)Older adults (n = 10)p valueEffect sizeSTEADI score, 0–14 (SD)–1.8 (1.48)––FES-I, 16–64 (SD)–18.8 (2.9)––PHQ-9, 0–27 (SD)–0.6 (0.89)––Pain VAS, 0–10 (SD)–1.4 (2.07)––DHI score, 0–100 (SD)–1.8 (3.03)––CAIT, 0–30 (SD)–4 (4.69)––SD: Standard Deviation, STEADI: Stopping Elderly Accidents, Deaths and Injuries, FES-I: Fall Efficacy Scale-International, PHQ: Patient Health Questionnaire, VAS: Visual Analog Scale, DHI: Dizziness Handicap Inventory, and CAIT: Cumberland Ankle Instability Tool Demographics and clinical measures SD: Standard Deviation, STEADI: Stopping Elderly Accidents, Deaths and Injuries, FES-I: Fall Efficacy Scale-International, PHQ: Patient Health Questionnaire, VAS: Visual Analog Scale, DHI: Dizziness Handicap Inventory, and CAIT: Cumberland Ankle Instability Tool Effect of testing conditions on SRT Figure 1 displays the graphical representation of the results of the SRT measurements across conditions and groups. The results of repeated measures analysis showed that amplitude accuracy was significantly affected by age (p < 0.0001, ES = 0.905), vibration condition (p = 0.012, ES = 0.240), and ankle testing side (p = 0.015, ES = 0.197). The young control participants performed better than the older participants for both the right ankle (40% better on average across conditions) and the left ankle (41% better on average across conditions). Vibration condition contributed to worse performance for both the young control participants and the older participants; on average, between both groups, there was a 10% better performance for no vibration compared to 40 Hz. On average, the older participants performed 9% better and the young control participants performed 7% better with the right foot, which was the dominant side for all participants. Graphical representation of results of SRT measurements for both young and older adults for left foot and right foot and across no vibration and 40 Hz vibration. A smaller value of amplitude accuracy and a larger value of directional accuracy correspond to a better performance. Amplitude accuracy is defined as the average straight-line distance from the participant’s tracking circle to the predetermined track. Directional accuracy is defined as the average amount of time the participant is within the free zone. For amplitude accuracy, significant effects of age (p < 0.0001), vibration condition (p = 0.012), and ankle testing side (p = 0.015) were observed. For directional accuracy, significant effects of age (p < 0.001), vibration condition (p = 0.009), and ankle testing side (p = 0.026) were observed For directional accuracy, we observed significant effects of age (p < 0.001, ES = 0.658), vibration condition (p = 0.009, ES = 0.228), and ankle side (p = 0.026, ES = 0.175). The young control participants performed better than the older participants for the right ankle (59% better across conditions) and the left ankle (49% better across conditions). Similar to amplitude accuracy, on average and across both groups, better performance was observed for the 0 Hz vibration condition compared to 40 Hz vibration (12% better across conditions). There was better performance with the right side for amplitude accuracy; on average, the older participants performed 1% better and the young participants performed 20% better with the right ankle. The results of the participants’ average performance across different tracks are reported in Table 2. The young participants performed better than the older participants across all three tracks for both amplitude accuracy and directional accuracy. No interaction effects of age with testing track were observed (p > 0.05). Table 2Results of the three tracks of testing for older and young participantsOutcomeTrack 1Track 2Track 3Older Adults (n = 10) Amplitude Accuracy, pixels (SD)22.14 (10.41)15.92 (5.31)17.83 (7.22) Directional Accuracy, % (SD)65.28 (19.14)70.23 (16.85)63.52 (20.93)Young Controls (n = 19) Amplitude Accuracy, pixels (SD)11.73 (2.85)10.96 (2.34)11.28 (2.44) Directional Accuracy, % (SD)85.86 (9.94)85.77 (9.38)82.30 (9.86)A smaller value of amplitude accuracy and a larger value of directional accuracy correspond to a better performance. Amplitude accuracy is defined as the average straight-line distance from the participant’s tracking circle to the predetermined track. Directional accuracy is defined as the average amount of time the participant is within the free zone Results of the three tracks of testing for older and young participants A smaller value of amplitude accuracy and a larger value of directional accuracy correspond to a better performance. Amplitude accuracy is defined as the average straight-line distance from the participant’s tracking circle to the predetermined track. Directional accuracy is defined as the average amount of time the participant is within the free zone Test–retest reliability Testing was done one week apart among eight young participants (age = 22 ± 1 years). Test–retest reliability was confirmed for both amplitude accuracy and directional accuracy. On average, the ICC values for two tests were 0.86 ± 0.01. Discussion Effect of age on visuo-proprioception In the current study, we addressed the capability of SRT to assess the effect of proprioceptive decline on sensorimotor function caused by aging, based on a cross-sectional comparison of SRT performance among young and older adults. As hypothesized, significant age effects were found in SRT performance for both amplitude accuracy and directional accuracy. Age had the largest effect size among all factors (compared to testing side and vibratory stimulation effects). Deterioration in proprioceptive performance with age is linked to degeneration of sensory receptors, decline in peripheral nerve function, changes in the central nervous system (CNS) processing, and changes in muscle and joint performance. Proprioceptive feedback from muscle spindles, Golgi tendon units, and other joint receptors provide information about limb orientations [20]. With aging, the efficacy of these receptors diminishes due to morphological changes in receptors (e.g., increased spindle capsule thickness and reduction in spindle diameter) and the natural loss of myelinated sensory receptors, leading to a gradual decline in proprioception [10, 25, 31, 33, 42]. Furthermore, decline in peripheral nerve function is observed as there are changes in the spinal network and afferent fibers, leading to less efficient synaptic transmission of signals from muscles, joints, and tendons [54]. In addition to changes in the peripheral nervous system, CNS regions involved in proprioception, including the somatosensory cortex, cerebellum, and parietal cortex, show atrophy or functional decline due to aging [5, 36]. Deterioration in proprioceptive performance leads to an elevated risk of fall in older adults. Therefore, it is critical to accurately measure proprioceptive performance and effects of proprioceptive decline on sensorimotor performance for identifying high fall risk older adults and to identify those who can potentially benefit from rehabilitation programs targeting proprioceptive deficits. All these factors may lead to inefficient reflexive responses and delayed or inaccurate feedback to the CNS about limb position, especially ankle joints, which can cause poorer balance [3, 5, 49]. This is because two factors associated with balance recovery deterioration among older adults are delayed placement of the recovery leg and delayed moment generation in the stance leg [33]. In both factors, proprioception from the ankle joint muscles has the critical role for feedback (reactive) and feedforward (preparatory) control, and the regulation of muscle stiffness, to achieve specific movement acuity, coordination, and balance, and eventually a quick and efficient recovery response to balance perturbations [19, 23, 30]. However, consequences of aging-related deficits in ankle proprioceptive information during balance recovery are not well-understood. Previous work suggested that there are age-related alterations in implications of proprioceptive information from the ankle joint, for maintaining balance during upright standing [12, 32, 49]. In our recent work, we showed that alterations in proprioceptive information from the ankle is also altered by age during the balance recovery from perturbation [3]. Overall, previous evidence suggests a reduction in reliance on ankle proprioceptive information for maintaining balance in older adults compared to young adults, which may occur due to age-related impairments in ankle proprioceptive performance. This highlights the importance of assessing ankle proprioceptive performance. Current results showed promise for using SRT in detecting proprioceptive alteration in the ankle joint due to aging. Potentially using SRT as a visuo-proprioception measure to detect age-related changes in sensorimotor function due to ankle proprioception decline may provide a tool to separate the effect of ankle strength and vestibular deficits from visuo-proprioception. SRT may also be used to assess disease-related changes in proprioception caused by, for example, spinal cord injury, diabetes, and stroke. Due to its objectivity and capability of developing different tracking scenarios, learning effects can be minimized, making SRT a tool for tracking longitudinal changes in sensorimotor function caused by changes in visuo-proprioceptive performance. Although current results were promising in showing differences in SRT performance between young and older adults, there may be other factors other than proprioceptive performance that could cause poorer SRT performance for older adults. By combining a visual task with a motor exercise, visual and lower limb motor deficits could also play roles in the observed worse SRT performance due to aging. While all participants were screened for a history of vision and ankle problems, as well as for any cognitive declines, older adults are privier to such defects [27, 36]. Therefore, findings from the current study should be interpreted cautiously, as SRT may not represent pure deficits in proprioceptive performance. Effect of testing right vs. left side on visuo-proprioception The SRT performance was significantly different between the left vs. right testing side; better performance was observed for the right ankle, which was the dominant side for every participant in the current study. Interestingly, the observed better performance in the dominant side was true for both the older adult and the young control participants and for both amplitude accuracy and directional accuracy. Other studies suggest that postural balance, within monopedal postural experiments, may be influenced by testing side being the dominant or non-dominant [35]; nevertheless, findings from these studies are contradictory, as postural balance is highly associated with other factors, such as muscle strength. Focusing on the knee joint, testing the effect of dominancy in previous work showed significant effect when testing the proprioceptive performance using JPS, during both weight-bearing and non-weight-bearing scenarios [17]. Interestingly, focusing on the ankle joint, previous work suggested that JPS or kinesthesia tests were not sensitive enough to detect differences in proprioceptive performance between dominant vs. non-dominant sides [41]. This might be due to the fact that both JPS and kinesthesia rely on conscious perception of afferent signals and are highly influenced by the attention level for memorizing the perceived position/force [27, 45, 56]. During these measurements, the participant needs to pay close attention and remember a particular movement and replicate it [38]. By incorporating engaging visuo-proprioceptive sensor-based feedback within SRT, the reliance on memory may be minimized, leading to less bias due to these factors in the testing procedure. Accordingly, we were able to detect minor differences in sensorimotor performance caused by visuo-proprioceptive differences between the dominant vs. non-dominant sides in both young control and older adult participants. Effect of vibratory stimulation on visuo-proprioception To assess that SRT outcomes are influenced by disturbance of proprioceptive performance, we implemented vibratory stimulation on muscles to disturb proprioception acuity and recorded the alteration in proprioceptive performance. Results from this study showed a significantly worse SRT performance when visuo-proprioception is disturbed through the addition of vibration devices, suggesting that SRT performance is sensitive to alterations in proprioception deficits. The application of vibratory stimulation for disturbing postural sway during upright standing, timed up and go, and balance recovery performance has been established in previous work [2, 7, 48, 49, 51, 55]. In these studies, it has been indicated that such mechanical vibration disrupts balance through the activation of the afferent nerves of muscle spindles and other sensory receptors by changing the firing rate. Vibratory stimulation could also cause kinematic illusions in the brain that results in a decreased ability to sense joint position. Although the observed influence of vibration on SRT may not confirm that SRT directly measures proprioceptive performance, it provides evidence that SRT is highly influenced by changes in proprioceptive performance. Nevertheless, while with the experimental setup we tried to limit vestibular and strength effects that exist in balance and timed up and go tests, other factors may still play role in SRT performance, including vision. In the current study, the application of four vibration devices at 40 Hz, positioned on the tibialis anterior, peroneus longus, soleus, and gastrocnemius, during the SRT was associated with a worse performance in both participant samples and for both objective outcomes (amplitude accuracy and directional accuracy). The selection of 40 Hz frequency here was based on previous work that suggested it as the required frequency to perturb upright standing and dynamic balance [2, 48, 49, 51]. Such vibration creates an increased excitement of type Ia afferents in muscle spindles, leading to a limited adjustment of the muscles [40, 49, 51]. The targeted muscles were selected, because during static and dynamic balance, proprioceptive information from these muscles provides the most pertinent source of afferent information for ankle strategy balance performance and have been the targeted muscles for vibratory stimulation in previous research [7, 8, 13, 53]. As seen in the Results section, there was no noticeable performance difference with and without the addition of 40 Hz for the right foot for older adult participants. We speculate that these results may be due to the fact that on average, older adults already have impaired proprioception. Thus, adding perturbation may, on some occasions, not cause a decreased performance. We observed similar effects of vibration on balance performance in older adults in our previous work [49]. In summary, current findings suggest that SRT is sensitive enough to capture differences in sensorimotor performance caused by visuo-proprioceptive alterations in the ankle joint due to outside stimulation imposed by vibration of ankle muscles. Test–retest reliability of visuo-proprioception assessment In a sub-sample of healthy young participants, we assessed test–retest reliability of SRT for measurements one week apart. Results showed excellent test–retest reliability on average across different task difficulties (ICC = 0.86 ± 0.01). By combining visual feedback and proprioceptive testing, the observed high test–retest reliability was achieved, which is higher on average compared to most reported values for JPS, kinesthesia, and force sense tests in the literature [6, 44]. In these methods, including JPS, kinesthesia, and force sense tests, the level of attention for memorizing a certain level of movement/force may play a more important role for accurate assessment of proprioceptive performance. SRT may provide a more engaging setup through the incorporation of visual feedback. However, JPS and force sense rely on memory of a particular movement that the participant has to replicate. [38]. In contrast, while in performing SRT, there is some reliance on the level of attention, the participant’s ability to memorize a sensation may not have an important influence on the measurement. Limitations and future work To interpret the results of the study, few limitations need to be considered. One limitation is the lack of participants that are classified as high fall risk. Among our older adult participants, only one had reported an incident of falling in the past year. The inclusion of older adult participants with higher risk of fall is important to reinforce SRT for assessing proprioceptive decline across a wider spectrum of proprioceptive performance. Nevertheless, our findings showed promise in detecting age-related decline, which is expected to be less pronounced compared to those at high risk of falling or those who develop disease-related deficits in proprioceptive performance (e.g., stroke or diabetes). Accordingly, we expect that the application of SRT would go beyond age-related decline in proprioceptive performance, for assessing more severe decline in proprioception, which will be addressed in the future work. Furthermore, we aim to recruit a larger number of participants, especially older adults, in the future to better justify and support the conclusions presented here. Another limitation of the current study is the lack of cross-sectional comparison between SRT and common balance testing, such as upright standing body sway or balance recovery, which are highly rely on acute proprioceptive information. Similarly, there is a lack of comparison with JPS and kinesthesia tests for the ankle joint. In related future studies, cross-sectional assessment of SRT association with balance and proprioceptive tests are required. We also did not account for participant reaction time in our calculation of amplitude and directional accuracy. In the future, we wish to create a composite score for SRT to take into account both of our outcome measurements, as well as reaction time. Similarly, another limitation was the lack of measurement to assess CNS control and reflexive responses. Due to the nature of measurements in the current experiment, the differentiation between short-latency muscular reflexive and long-latency feedback control through the central nervous system was not investigated. We will continue our research in the future to explore how SRT performance is associated with reflexive responses and brain function using EMG and EEG assessments. Furthermore, our proposed method includes dynamic testing; however, it lacks assessment under weight-bearing circumstances, such as standing or walking. Weight-bearing condition could introduce vestibular feedback and muscle strength effects. Although in the current approach visual feedback plays role, the effect of other components, such as vestibular and strength is limited. Conclusion and clinical implications This study demonstrated the potential of the SRT test as a sensorimotor task that may reflect alterations in visuo-proprioceptive performance due to intrinsic and/or extrinsic factors. Across trials of the game on both the left and right ankles, with and without exposures to proprioceptive disturbance imposed by vibratory stimulation, we measured directional accuracy and amplitude accuracy among young control and older adult participants. Significant effects of age, vibration, and testing side on SRT performance were observed, with age showing the largest effect size among these independent variables. These results suggest the possibility to use SRT as a sensorimotor task to assess sensory integration that is essential for motor control, which can be related to proprioceptive deficits along with other aging-related factors that can alter visuo-proprioceptive performance (e.g., visual feedback). Specifically, it is critical to accurately measure proprioceptive performance as one important component of balance, to identify those who can potentially benefit from rehabilitation programs targeting proprioceptive deficits. Nevertheless, isolating proprioceptive performance from other balance components, such as muscle strength and vestibular acuity is challenging. To achieve this, the method for the current work was a non-weight-bearing setup. Finally, transferring this sensory-based assessment to a smartwatch is practical based on previous work [21] and will be investigated in the future for improved scalability of the technology. Providing SRT within a smartwatch platform would make it suitable for in-home and in-clinic use. Methods Participants Two groups of participants were recruited: young adult participants (18–30 years) and older adult participants (65 + years). Participants were recruited from Tucson, Arizona at the University of Arizona and from Newark, New Jersey at Rutgers University. All participants provided written informed consent in accordance with the Declaration of Helsinki [43], with approval from the Review Boards of the University of Arizona and Rutgers University. Inclusion criterion for both groups of participants was the ability to understand study instructions. The exclusion criteria for the young participants were cognitive impairments [montreal cognitive assessment (MoCA) < 20], mobility disorders, history of ankle injuries, and history of dizziness, vertigo, sedating medication, or alcohol consumption within 24 h of testing. For older adults, exclusion criteria included disorders causing severe motor and balance deficits (e.g., stroke, Parkinson’s disease, severe arthritis, lower-extremity amputation, spinal cord pathologies, and diabetes), history of ankle injuries or severe lower leg orthopedic conditions, history of severe vestibular disorders, central nervous system diseases, cognitive impairment (MoCA score < 20), vision problems affecting balance, and history of dizziness, vertigo, sedating medication, or alcohol consumption within 24 h of testing. Clinical measures Questionnaires were collected from all older participants to account for confounding variables that can potentially influence balance. Such clinical questionnaires were reordered for only the older participants, as these measures were more relevant to this age range. Questionnaires included Stopping Elderly Accidents, Deaths and Injuries (STEADI) [47] for fall risk assessment, fear of falling using Fall Efficacy Scale-International (FES-I) [57], Charlson comorbidity score [11], depression using Patient Health Questionnaire (PHQ-9) [29], pain in lower-extremity based on Visual Analog Scale (VAS) [30], vestibular deficits using Dizziness Handicap Inventory (DHI) [23], and Cumberland Ankle Instability Tool (CAIT) [19]. DHI evaluates dizziness-related impairment and vestibular symptoms, which are relevant for balance. We also recorded the dominant side for each participant; the dominant leg was defined as the leg the person would use to kick a soccer ball [14]. Study design and vibration stimulation Participants performed 12 trials of SRT including a full combination of three tracks (Fig. 2) on both lower limbs (left and right ankle), with and without proprioceptive disturbance. Vibratory stimulation was used in the current study for proprioceptive disturbance. All trials were randomized, as per our study protocol, to minimize the participants learning and fatigue effects as well as potential residual effect of vibration exposures. Vibration was turned on 5 s before the task and turned off 5 s after each trial was completed. In addition, there were 1-min breaks between trials both to minimize the after-effect of the disturbance caused by the vibration, as well as to minimize fatigue. Of note, previous work suggests that depending on the duration of exposure, there would be long-term effects of vibration on muscle spindles that fades away, usually in an amount of time equal to exposure duration [39]. Stochastic vibratory stimulation was used as a method for disturbing the proprioceptive performance, which has been validated in previous upright standing and dynamic balance testing research [2, 3, 49, 51]. By applying such stochastic vibration, there is a rise in the excitement of type Ia afferents in muscle spindles [49, 51]. Signals from muscle spindles are directed to motor neurons, which activate the parent muscles to restore joint position. Stochastic vibration can affect this short-latency reflexive mechanism [34]. Proprioceptive feedback from muscle spindles provides information regarding limb orientation, which is processed in the central nervous system, and stochastic vibration can also influence such long-latency responses [16, 52]. a Participant set up for the SRT testing; the red circles from top to bottom indicate the set of straps to keep the participant seated, the set of straps to restrict movement to the ankle, and the sensor placement for measuring ankle rotation; b vertical and horizontal movements based on rotation; c Tracking session showing the blue dot that represents the participant and the red dot that represents the predefined track. Of note, the tracking path is highlighted in the figure; however, during the actual testing, participants were only able to see the blue and red dots; d predefined tracks, from left to right: track 1, track 2, and track 3; e actual tracking session for a young control participant In this approach, we use stochastic vibration to minimize the anticipation effects [46, 51]. 40 Hz was selected based on our previous work that showed vibration frequencies around 40 Hz may provide adequate noise for altering both young and control older adults’ muscle spindle performance and consequently postural balance [50]. For this, vibratory stimulation was applied bilaterally to muscles surrounding the ankle area, including the tibialis anterior, peroneus longus, soleus, and gastrocnemius. Vibratory devices were placed on the belly of each muscle. These muscles were selected as they provide most critical proprioceptive information for balance [1, 9, 13]. Magnetic actuator systems (C-2HDLF Tactor, Engineering Acoustics, FL, USA) were used for stimulating muscles, and a Universal Controller (TDK, Engineering Acoustics, FL, USA) was used for providing 40 Hz Gaussian noise for stimulation. These vibration devices were secured with adjustable and stretchable bands that were fastened with Velcro straps. Each device was placed into a small pocket within the band to assure they remain without movement. For assessing test–retest reliability, a sub-sample of healthy young participants performed six trials of SRT, one week apart from the original measurements, including a full combination of three tracks for two sides without any perturbation. SRT testing SRT incorporates 30-s trials of tracking a flying target (red circle) by moving a blue circle using ankle joint movements in the plantarflexion/dorsiflexion and abduction/adduction directions, as shown in Fig. 2c. Each participant sat on a tall adjustable chair to avoid contact between feet and the floor (Fig. 2a). The thigh and shin of the testing side were fixed to the chair using Velcro, to limit the tracking to the ankle and not the knee and hip movements. Participants wore a custom-made motion sensor device (core sensor: L3GD20H triaxial gyroscope, sampling frequency of 500 Hz) on the testing foot (Fig. 2a), placed on the top of the intermediate cuneiform. First, to assure that the range of motion for the ankle joint was within the appropriate threshold for the testing, each participant rotated the ankle to the maximum angle of plantarflexion/dorsiflexion and abduction/adduction (Fig. 2b). This task was done using a rectangular plane centered at (0,0) with the axes extended out to 300 pixels in all directions. The threshold of 300 pixels was selected based on 85% of the range of motion among participants we recruited in our pilot studies. Participants were required to reach the threshold in each of the direction and were excluded if they could not reach the range of motion threshold due to post-traumatic stiffness, arthritis, bone spurs, tendon tightness, or any other reason. Of note, only one young control participant was excluded due to the inability to reach such threshold. For SRT testing, participants saw two circles, one blue and one red, on a monitor 100 cm away (Fig. 2c). The monitor had a refresh rate of 60 Hz, and there was no delay between ankle movement and cursor movement on the monitor. One degree of ankle movement in each direction represented an average of 8.86 pixels per cursor movement on the screen. The monitor had a 92.5 pixels per inch value, and with this set up, one centimeter was equivalent to 36.4 pixels. The participants were instructed to follow the red circle as closely as possible by rotating their ankle, and they were informed to get ready for the task prior to starting each trial. There was no practice trial; however, the order of trials was randomized to limit learning effects. The red circle was derived from the SRT software to represent the predefined track, and the blue circle was derived from the motion sensor on the foot. Only the circles were visible on the screen, not the trajectories, to avoid prediction of movement. We established three movement pattern difficulties based on the number and type of path curvatures and angles, as well as speed of movement of the blue circle. For track 1, the range of motion in the x axis was − 148 to 95 to pixels, in the y axis was − 112 to 89 to pixels, the average speed of movement was 28.3 pixels/sec, and the average rate of angular path change was 6.5 degrees/sec. Corresponding values for track 2 were − 49 to 84 pixels, − 40 to 95 pixels, 20.2 pixels/sec, and 3.7 deg/sec, and for track 3 were − 73 to 93 pixels, − 93 to 93 pixels, 31.7 pixels/sec, and 3.9 deg/sec. SRT outcomes Using a custom Python software, we recorded the distance between the two circles; we implemented the coordinates data (X: abduction/adduction, Y: plantarflexion/dorsiflexion) from the red (the predefined track) and blue (the participant ankle movement) circles, to extract outcomes, including amplitude accuracy and directional accuracy. After the participant completed a trial, a set of (X, Y) coordinates was generated, that matched the exact number of (X, Y) coordinates for the track, allowing us to compare between the participant and the track. Amplitude accuracy was obtained by calculating the absolute straight-line distance between the two circles during the test sequence; a larger amplitude accuracy corresponds to worse performance. Directional accuracy was calculated as the percentage of the time that the participant kept the blue circle within a free zone of the red circle. For each track, we defined the shape and size of an elliptical free zone, at each instant of time, to cover one standard deviation of distances between the blue and red circles from the healthy young participants' data within our cohort. We recruited a relatively larger number of young participants in comparison with older participants to provide a robust control database for calculation of the free zone within the directional accuracy outcome. We removed outliers for the calculation of the free zone. A lower directional accuracy corresponds to a worse performance. Statistical analysis Demographic information and clinical measures were reported and compared, when possible, between young and older participants using analysis of variance (ANOVA) models for continuous variables and using chi-square analysis for sex distributions. We chose to use ANOVA, since the data were normally distributed. We checked outliers (using Box Plots) and normality (using the Shapiro–Wilk test). The association between SRT outcomes and ankle side (dominant vs. non-dominant), vibration conditions (0 Hz or no vibration and 40 Hz vibration), and groups (young vs. older) was determined using repeated measures mixed effects models. In these models, independent variables of vibration conditions and ankle side, were considered as within-subject factors; group was considered as the between-subject variable. Sex and BMI were considered as covariates. Effect sizes (ES) were calculated using Cohen's d calculations, using G*Power (version 3.1.9.7, copyright 2020, University of Dusseldorf, Germany). To evaluate the test–retest reliability of SRT between two assessments, intraclass correlation coefficient (ICC) two-way mixed effects F test models were implemented. All analyses were done using JMP (Version 14, SAS Institute Inc., Cary, NC, USA), and statistical significance was concluded when p < 0.05. Acknowledgements This project was supported by an award from the Rutgers Health Advance—Advancing Biomedical Technologies. We want to thank Mustafa Erfan Allababidi and Alexis Sullivan for their contributions to data collection and analysis. Author contributions Conceptualization, N.T. Methodology, M.H., M.HA, K.A., R.W., and N.T. Software, M.H. and R.W. Validation, M.H. Formal analysis, M.H. Investigation, M.H., M. HA., K.A., and R.W. Data curation, M.H., M.HA., K.A., R.W. Writing—original draft preparation, M.H. Writing—review and editing, M.HA., K.A., R.W., and N.T. Visualization, M.H., M. HA., K.A., R.W., and N.T. Supervision, N.T.; Project administration, N.T.; Funding acquisition, N.T. All authors have read and agreed to the published version of the manuscript. Funding This research was funded by Advancing Biomedical Technologies: sensor-based real-time tracking game (SRT), obtained by Dr. Nima Toosizadeh. Data availability The data are not publicly available at this time. The data were analyzed using a custom python software. The materials used for this study belong to Rutgers University. Data and materials can be requested from Dr. Nima Toosizadeh. Declarations Ethics approval and consent to participate All participants provided informed written and verbal consent in accordance, with the Declaration of Helsinki, to participate in this study, and with approval from the Institutional Review Boards of the University of Arizona and Rutgers University (protocol ID: Pro2023002282). Competing interest Nima Toosizadeh has patent pending to Rutgers University. All other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data are not publicly available at this time. The data were analyzed using a custom python software. The materials used for this study belong to Rutgers University. Data and materials can be requested from Dr. Nima Toosizadeh.