Getting older adults back on their bicycle; a pretest-posttest case-control study on the improvement of bicycle balance control

Getting older adults back on their bicycle; a pretest-posttest case-control study on the improvement of bicycle balance control | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Getting older adults back on their bicycle; a pretest-posttest case-control study on the improvement of bicycle balance control Eric Maris This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7765021/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper investigates the effectiveness of an intervention that aims at restoring bicycle balance control skills in older adults that have quit cycling. The intervention lasted 11 weeks and involved three components: (1) training on an exercise bicycle, (2) a balance control training on a bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator. In the bicycle simulator training, the difficulty of the balance control was increased gradually, slowly approaching the difficulty of bicycle balance control on the public roads. The experimental group consisted of 23 community-dwelling older adults that had quit cycling (N = 19) or that were on the verge of doing so (N = 4). The effectiveness of the intervention was evaluated by comparing the experimental with a matched control group on a posttest-minus-pretest difference in balance control skill and confidence. The intervention produced a very large improvement (Cohen’s d = 1.5, t(16) = 6.0, p < .001) in balance control skills and confidence on the public roads. This very large improvement does not rule out the importance of the slower process of acquiring a sufficient lower body strength. This study was retrospectively registered with identifier NCT07195526 on ClinicalTrials.gov at 19-09-2025. Balance control bicycle balance motor skill training bicycle simulator older adults elderly mobility intervention study case-control Figures Figure 1 Figure 2 Figure 3 Introduction With age, there is a sharp decrease in the percentage of citizens that still ride their bicycle. This is well documented in the Netherlands [ 1 ], where everyone learns to cycle at a young age, and cycling is the default means of transportation for short distances. In the Netherlands, the percentage of inhabitants that cycles at least several times per month fluctuates around 77% between age 30 and 70 [ 1 ]. However, over an average period of 15 years, this percentage decreases sharply to 39% [ 1 ]. Using life expectancy statistics [ 2 ], one can calculate that the average non-cycling period is 16 years. The physical health benefits of cycling are well documented [ 3 , 4 ]. For instance, there is a strong inverse relationship between commuter cycling and all-cause mortality [ 4 ]. Because pedaling force can be easily controlled, cycling is an excellent training stimulus for the improvement of lower body strength and power. In addition, it has a very low impact on the joints, and it offers increased mobility options that are necessary for a meaningful social life. In this paper, I describe, motivate, and evaluate a rehabilitation intervention focused on balance control (BC) skills that aims at getting older adults back on their bicycle. BC can be understood from the perspective of optimal feedback control (OFC) [ 5 – 12 ], which is the dominant theory in the neural control of movement (see Fig. 1 ). OFC distinguishes between four determinants of BC skill: (1) a computational system that computes control signals and is implemented in the neural networks of our central nervous system (CNS), (2) a motor output system (part of the somatic nervous system; SNS) that sends the control signals to the muscles via the efferent axons of the spinal motor neurons, (3) a mechanical system, characterized by a.o. joint strength and range-of-motion, as implemented by the muscles, tendons, and cartilage, and (4) a sensory system (part of the SNS) that receives feedback from the eyes, the inner ear (labyrinth, vestibular organ) and the proprioceptive sensory organs (muscle spindles and Golgi tendon organs) and transmits this to the CNS. OFC is consistent with the fact that the effects of different types of balance training are specific to the balance task that was trained [ 13 , 14 ]. Specifically, OFC claims that, separately for every balance task, the CNS optimizes BC under the constraints/limitations of the mechanical and the somatic nervous system (torque and range limits of the mechanical system, precision of the sensory feedback, motor noise …). This differs from most other views on BC that are found in the literature. These views assume a set of component BC skills that, depending on the set of balance task requirements, are recruited in different combinations. Under this assumption, training of component BC skills affects performance in all BC tasks that depend on these component skills. OFC, on the other hand, states that BC skills are task-specific, and claims that the generalization of the training effect scales with the similarity between the task used in the training and the objective of the training (here, bicycle BC on the public roads). Crucially, this similarity is defined at the level of the task and does not require a task analysis in terms of component skills. To maximize generalization of the training effect to bicycle BC on the public roads, I started from the mechanics of bicycle BC. The mechanics of bicycle BC pertains to two centers of gravity (CoGs) that both must be kept above their respective area of support (AoS): (1) the CoG of the rider-bicycle combination must be kept above the line that connects the two wheel-road contact points, and (2) the CoG of the rider’s upper body must be kept above the saddle [ 11 ]. To achieve these two objectives, steering (turning the handlebars) is the most important control action [ 11 , 15 ]. To maximize the generalization of the training effect, I used a bicycle simulator that requires steering to keep the two CoGs above their respective AoS. Bicycle BC skill is a continuum defined by perturbations that distort the balance. These perturbations will be called BC challenges. An important BC challenge are the upward reaction forces that result from pedaling, deform the upper body and perturb the steering. The impact of BC challenges is modulated by the cycling speed, and this follows from the mechanical properties of the bicycle: as the bicycle's speed increases, so do the bicycle's self-stabilizing forces [ 15 ]. In the bicycle BC training, I gradually increased the challenges of the BC task, slowly approaching the challenges of bicycle BC on the public roads. For that, I used the increase in BC challenge level with pedaling force (due to the upward reaction forces) and its decrease with speed (due to the self-stabilizing forces). This was possible thanks to the bicycle simulator, which contains a motor that drives the rear wheel. At the end of the training, when participants could maintain BC with a high pedaling force at a low speed, the bicycle simulator was used to practice a safe start-and-stop technique that had to be used when cycling outdoors. For a lost motor skill that must be restored, it is important to know to what extent it depends on (1) plasticity in the CNS (responsible for the mapping of sensory feedback onto control actions), and (2) plasticity in the organs that are responsible for registering feedback and executing actions. Plasticity in the CNS depends on the signaling properties of neurons, and these change over a much shorter timescale (hours, days) than the organs that are responsible for sensory registration and motor execution [ 16 – 18 ]. To the best of my knowledge, plasticity in the sensory organs has not yet been demonstrated, and plasticity in the muscles with functional consequences (strength/power increase) operates at a timescale of at least weeks. To dissociate central (CNS-dependent) from muscular contributions to a possible improvement in BC skill, the training included sessions on exercise (spinning) bicycles on which the functional threshold power (FTP) could be monitored. I hypothesize that the training effects generalize to bicycle BC on the public roads and that there is a separate CNS-dependent contribution to this effect. Study setup and methods Participants Participants were recruited via an article in a local newspaper (De Gelderlander) about the goal of the study (March 31, 2023). This newspaper was distributed over an area with center roughly at the city of Nijmegen, the location of the intervention. The article invited participants that stopped cycling at least six months ago and wanted to get back on their bicycle. A total of 105 candidates replied, which were all invited for an information session on the Radboud University campus in Nijmegen. Based on their responses to a questionnaire, I excluded candidates that still cycled occasionally or suffered from balance-relevant sensory disorders (peripheral neuropathy, labyrinthitis, vestibular neuritis, …). I selected participants based on how far they lived from the location of the intervention, the Radboud University campus. The sole motivation for this selection criterion was the desire to minimize travel time for participants that had to visit the campus multiple times (see Intervention ). I selected the 15 participants that lived closest to the campus. The intervention started with cycling on a stationary exercise bicycle (spinning) in the second week of June 2023. This part of the intervention was in the Radboud University Sports Center. On June 15, a second newspaper article was published in which the location of the intervention was revealed. This second article was a progress report on the study and did not ask for participation. However, after this publication, 8 additional candidates spontaneously joined the spinning sessions, and requested to participate in the study, to which I agreed. Four of these additional participants still cycled on rare occasions but considered quitting because of fear of falling. A total of 23 participants started the BC training on the bicycle simulator, and 17 participants completed it (mean age 75,3 years, age range 66–88 years). From the 6 participants that dropped out, 2 did so for a reason that was related to the BC training: one participant’s legs were so short that she could not touch the ground while sitting on the saddle, and one participant lacked a sufficient force to keep herself balanced on top of a stationary bicycle. Four participants dropped out for reasons that were unrelated to the BC training: a fracture due to a fall at home (two participants), bursitis, and an urgent eye surgery. The 17 participants that completed the BC training were matched to 17 candidates (the control group) that were not selected for the study because they were living farther away from the location of the intervention as compared to the experimental group. The matching criteria were gender and age. Bicycle simulator The bicycle simulator is shown in Fig. 2 . The core elements of the simulator are three rollers with a width of 1.75 m. This is likely to be close to the width of an average Dutch bicycle path; prior to 2022, the advised minimum width was 2 m, but 40 percent of the bicycle paths did not meet this requirement [ 19 ]. The two front rollers are covered with a conveyer belt that is supported from underneath by a low-friction platform (not visible on Fig. 2 ). The bicycle’s front wheel rests on the conveyer belt and the rear wheel rests on the rear roller. The rollers’ rotational inertia is close to the linear inertia of the average Dutch rider-bicycle combination, and this ensures that the acceleration response to a pedal stroke feels natural. A torque motor is attached to the axis of the rear roller, and this motor is essential to assist weak participants in the first stages of the training. Finally, the bicycle simulator has a virtual-reality (VR) component that projects an animation of a bicycle lane (not visible on Fig. 2 ), which is controlled by the cycling speed and the cyclist’s lateral position on the left-right axis of the simulator, which was measured using a motion capture system. In between the front and the rear rollers, there is a platform on which the rider can place his feet and the trainers can stand. The rider has one trainer on each side, supporting him by means of gentle pushes on his shoulders. As the rider’s BC skills improve, the trainers provide less support and increase their distance from the rider. The rider wears a safety harness with which he is connected to a support construction above the rider’s head (see Fig. 2 ). The safety harness does not prevent the rider from falling, but in case he would fall, his knees would not touch the floor before the suspension becomes active. Intervention The intervention was over a period of 11 weeks. In that period, the participants engaged in (1) spinning, (2) BC training on the bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator. The three components of the intervention were also administered in this order, with spinning and BC training overlapping in time for the last 8 weeks. Spinning The first three weeks of the intervention only involved a group spinning training, which was given by two professional spinning coaches and one amateur. Participants trained at an individually determined power level, controlled by the one-minute functional threshold power (1minFTP). One spinning session lasted 45 min., excluding warming-up and cooling-down, but including a 5 min. break in the middle. Every session started with a 3-minute block in which the 1minFTP was assessed. The remainder of the session had the structure of an interval training. The high intensity blocks lasted at most 1 min. and were alternated with low intensity blocks at 70 percent of 1minFTP that lasted at most 2,5 min. Increasing intensity was achieved either by increasing the resistance (and keeping the RPM fixed) or by increasing the RPM (and keeping the resistance fixed). Assessing the 1minFTP The 1minFTP was used for a double purpose: (1) as a measure of leg power, to investigate its relationship with bicycle BC skill, and (2) to train at an individually determined power level. The 1minFTP is the highest power a participant can maintain for one minute while cycling at 60 revolutions per minute (RPM). The 1minFTP was assessed by having participants monitor their breath after a one-minute effort at some power level: if the participant could say only a few words at a time (a proxy for reaching the ventilatory threshold), the 1minFTP was reached. Assessing the 1minFTP required keeping a constant high-power output, and this turned out to be very difficult for our participants. First, assessing the 1minFTP required that the participants could operate the bicycle computer that provided feedback about their power output. It took participants two weeks to learn to operate this computer. Second, it took several participants substantially more time to learn to produce a constant power output. Initially, all participants lowered their RPM when the resistance was increased above a certain level, thereby preventing a higher power output. Several participants continued doing this until the fourth week. Third and last, participants’ performance was highly variable with respect to the maximum power output they endured, and this was both within sessions (no longer following the trainer’s instructions with respect to power output) and between sessions (lowering the 1minFTP at the beginning of a session if one was feeling tired). Bicycle simulator BC training BC training started in the fourth week of the intervention and was in one of the research labs of the Radboud University (Faculty of Social Sciences). The BC training was individual and was given by two research assistants. A BC training session lasted approximately 30 min. At the start, participants put on a safety harness and were connected to the support construction. The saddle height was adjusted such that the participant’s feet touched the platform in between the front and the rear rollers. For as long as the participant could not balance independently and/or showed signs of fear, the trainers gently pushed their hands against the participant’s shoulders. They gradually released the pressure as the participant’s BC became more independent, and he gained confidence. The guiding principle in the BC training was to increase the difficulty in small steps and following the participant’s progress and confidence. Every session started with a BC challenge level at which the participant was comfortable, and that level was slowly increased if the participant agreed. In the first session, the bicycle was brought up to a speed of approximately 12 km./h. by means of a torque motor. In this condition, the participant’s only task was to turn the handlebars for balance control. The BC challenge level was increased in four ways: (1) by decreasing the speed, (2) by decreasing the assistive force at the rear roller, (3) by asking participants to accelerate from standstill and to stop at any desired moment, and (4) by asking the participant to tap the support surface with both feet, place them back on the pedals, and to continue pedaling (feet tapping exercise). When participants could balance independently while also pedaling, they practiced a safe start-and-stop technique in which the participant could always touch the floor with his feet. To achieve that, the saddle had to be set at the appropriate height, which was made easy by installing a dropper post in the seat tube. At standstill, the participant was seated on the saddle and had both feet on the ground. He started moving by applying a forceful pedal stroke with his favorite leg and continued with both legs. When stopping, the participant took his feet from the pedals and placed them on the stationary floor. When feeling adequately supported by his legs, the participant pulled the brakes, thereby stopping the rollers of the bicycle simulator. Participants were suggested to use the dropper post lever to increase the saddle height when they were up to speed and to lower it when transitioning to a stop. However, none of the participants opted for this. Cycling outdoors The participants were instructed to start cycling outdoors on their own bicycles when they could do 10 feet taps consecutively without stopping in between, and at a resistance level slightly above the resistance on the public roads. When cycling outdoors, the participants were instructed to use the safe start-and-stop technique (feet touching the road surface) that was practiced on the bicycle simulator, to start cycling in an environment without other road users (a park, a parking lot), and to gradually move to more busy environments. A few participants informed the trainers that they had started cycling outdoors before they were instructed to do so. Those participants were given the same instructions as the other participants. Effect evaluation The effect of the intervention was evaluated according to the design of a pretest-posttest case-control study. In the experimental group, the outcome variable (see Outcome variable ) was measured twice, the first time in the beginning of the intervention and the second time six weeks after the end of the intervention. To correct for a possible spontaneous recovery, a control group was included that did not receive the intervention, but in which the change in the outcome variable was assessed over the same time interval as in the experimental group (see Participants ). The effect of the intervention was evaluated statistically using a dependent samples t-test that compared the change in the outcome variable between the matched experimental and control group participants. Outcome variable The outcome variable was obtained from a questionnaire that evaluated participants’ bicycle BC skills and confidence while riding on the public roads. Because there was no existing questionnaire that measured the relevant dependent variable for this study, I developed a dedicated questionnaire, of which an English translation is added to the Supplementary Materials; the original questionnaire was in Dutch. To maximize the validity of this questionnaire, the questions were formulated in terms of easily observable criteria. The introduction to the questionnaire was the following: Three environments will be described, and you must indicate whether you rode your bicycle in that environment in the past two weeks. By “riding your bicycle”, we mean starting from a standstill and stopping at a desired location. For every environment, you must indicate what applies to you: (1) I did not ride my bicycle here, (2) I rode my bicycle here, but I try to avoid it, and (3) I rode my bicycle here without problems. The three environments were the following: (1) without traffic (e.g., an empty parking lot, a quiet park), (2) with light traffic (e.g., small roads in an outside area, a bicycle lane on a quiet moment), and (3) in busy traffic (e.g., city center during traffic hours, busy crossroads with mixed traffic). The answers to every question/environment were converted into a score by assigning 0, 1, and 2 points to the respective options. Summing these scores over the three questions/environments results in a number between 0 and 6, which is the outcome variable. In the experimental group, this questionnaire was administered once in the beginning of the intervention and a second time six weeks after the end of the intervention. The questionnaire was sent as an attachment to an email, and participants were called on their phone if they had not replied after one week. For some participants, it was necessary to clarify some aspects of the questions. In the control group, the questionnaire was administered only once and at around the same time that the second questionnaire was sent to the experimental group. None of the control group participants was riding their bicycle at the start of the intervention and they were asked whether and how this had changed in the meantime. Results I conducted a pretest-posttest case-control study to evaluate the effectiveness of an intervention aimed at restoring bicycle BC skill. The experimental group consisted of 23 participants, of which 17 completed the intervention (see Methods ). The intervention lasted 11 weeks and involved three components: (1) cycling on a stationary exercise bicycle (spinning), (2) BC training on a bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator (see Methods ). Cycling on the public roads started when participants were successful in a performance-defining bicycle simulator task: making 10 consecutive feet taps without stopping in between, and at a resistance level slightly above the resistance on the public roads. At the start of the BC training, no participant could do this task, but at end of the training all participants that completed the training were successful. The effectiveness of the intervention was evaluated using an outcome variable that did not directly reflect the performance on a task that was extensively practiced during the training (e.g., the feet tapping task). Instead, the effectiveness was evaluated using an outcome variable that is much closer to cycling outdoors: a questionnaire that assesses the participant’s skill/confidence on his own bicycle when cycling outdoors (see Methods ). In the experimental group, this outcome measure was obtained twice, first in the beginning of the intervention, and then six weeks after the intervention. To correct for a possible spontaneous recovery of cycling skill/confidence, I used a matched control group that did not receive the intervention, but in which the change in cycling skill/confidence over the same time interval was assessed (see Methods ). The improvement in cycling skill/confidence was significantly larger in the experimental group (M = 3.1, SD = 2.1) compared to a matched control group (M = 0, SD = 0), t(16) = 6.0, p < .001, 95% CI [2.0 4.1]. Note that none of the control group participants had a spontaneous recovery of their cycling skill/confidence. The raw outcome measures of the experimental group are shown in Fig. 3 . The effect size (Cohen’s d) was 1.5, which is a very large effect [ 20 ]. This effect was obtained with an average of 4.8 simulator BC training sessions of 30 min. each. It was my original plan to dissociate central (CNS-dependent) and muscular contributions to a possible improvement in BC skill. For that, it was necessary to quantify the increase in maximum power output on the spinning bicycle, for which I used the one-minute functional threshold power (1minFTP), which was assessed at the beginning of every spinning session. However, keeping a constant high-power output (a requirement for a reliable FTP assessment) turned out to be very difficult for our participants and required several weeks of training (see Methods ). I conclude that I failed in obtaining a reliable quantification of the increase in maximum leg power increase in this group of participants. It is useful to report the mean 1minFTP on which participants converged after learning to produce a constant force output. This mean 1minFTP was 93.6 Watt (sd = 17.2). FTP is usually expressed relative to bodyweight, and in this participant group the average 1 minute power-to-weight ratio was 1.2 Watt/kg (sd = 0.3). For reference, it is useful to compare this value with published 1 minute power-to-weight ratios for untrained amateur cyclists [ 21 ] (defined as the 5-th percentile of the population of amateur cyclists): 3.9 for males and 3.3 for females. Thus, the participants in this study had an average 1minute power-to-weight ratio that is approximately three times smaller than that of an untrained cyclist. Discussion and conclusions A very large improvement in older cyclists’ BC skills and confidence can be realized by means of a short intervention (11 weeks), even if that person has quit cycling for a long time. This is remarkable given the task-specificity of the effects of different types of balance training [ 13 , 14 ], and is most likely due to the high similarity between the BC training on the simulator and cycling outdoors. Limitations This study has three important limitations: (1) it used only an indirect measure of bicycle BC skills on the public roads, (2) it did not investigate whether the participants can deal successfully with external BC challenges, and (3) it could not distinguish between an effect at the level of the CNS and an effect at the level of the muscles. First, in the absence of an existing direct measure of BC skills on the public roads, I used a self-report measure. Although all questions were formulated in terms of observable criteria, a self-report is not a direct measure of the behavior of interest. Nevertheless, a self-report measure is to be preferred over the direct measures that were collected as a part of the BC training (e.g., the assistive force required to accelerate from standstill, the performance on the feet tapping task). These direct measures reflect the performance on tasks that were extensively practiced during training, and there is no information about their predictive value for BC skills on the public roads. Second, this study did not investigate whether the participants can deal successfully with some important external BC challenges such as an uneven road surface (e.g., curbstones, potholes) and sharp turns. It is very likely that several BC challenges require a sufficient lower body strength to prevent perturbations from affecting the steering movements. For example, (1) standing on the pedals while covering an obstacle, and (2) choosing an appropriate lean angle before entering a turn. In addition, cycling safely not only involves BC (staying upright) but also navigation (avoiding obstacles and/or following a track), which requires additional skills such as selective attention, anticipation, and planning. These have not been investigated. Third and last, from the perspective of OFC, it is tempting to conclude that the improvement is mainly due to the fast plasticity in the CNS, and not to the comparatively slow plasticity in the muscles. However, to make a strong case for this claim, it is essential to have a reliable measure of maximum power output, and this turned out to be very difficult. The existing literature on the relation between, on the one hand, muscle strength, and on the other hand, the incidence of falls and balance task performance, is inconsistent: one systematic review (focusing on falls) [ 22 ] argued for a strong relation and another (focusing on balance task performance) [ 23 ] argued for a very weak one. This inconsistency is probably related to the fact that, in the population of interest, falls cannot be predicted from the performance on balance tasks [ 24 ]. The complicated relation between muscle strength, balance control, and falls For several reasons, the relation between muscle strength, balance control, and falls, is a complicated one. First, many studies have been conducted in homogeneous groups (institutionalized participants, a small age range, …) and this restriction-of-range results in an underestimation of the correlation in the heterogeneous/unselected population. Crucially, restriction-of-range also operates at the individual level because most people are aware of the range of BC challenges that their BC skill allows, and they try to stay within this range. This restriction-of-range at the individual level reduces the correlation between, on the one hand, muscle strength and BC skills, and on the other hand, the incidence of falls. Experimental studies in which falls are elicited do not suffer from this restriction-of-range at the individual level, and this methodology has been applied successfully to investigate tripping-induced falls during walking [ 25 , 26 ]. One such study found a convincing correlation between leg strength and the ability to recover from tripping [ 26 ]. Second, it is very likely that (1) there are large differences between muscle groups and joint ranges-of-motion with respect to their relevance for BC, and (2) this differential relevance depends on the BC task. BC during walking, such as the ability to recover from tripping, must depend on leg strength because the leg muscles are used to bring the body CoG back over its AoS. However, in bicycle BC, different muscle groups can be used to control the combined rider-bicycle and the upper body CoG: (1) by controlling the handlebars, the arm muscles can bring the combined rider-bicycle CoG over its LoS, and (2) by controlling the upper body CoG, the core muscles in the trunk and the hip bring this CoG over its AoS, the saddle. Thus, the relevance of a particular muscle group for balance control likely depends on the BC task (walking versus cycling). Third and last, BC not only depends on the muscles for bringing the CoG above its AoS (the muscles’ motor role) but also for informing the CNS about the position of the CoG relative to the AoS, and this depends on the muscle spindles (the muscles’ sensory feedback role) [ 12 , 27 ]. To be useful, muscle spindle output must reflect the tension in the force-producing fibers, but its precise contribution to the loss of BC is unknown. Conclusion I have demonstrated that it is possible to realize a very large improvement in an older cyclist’s BC skills and confidence over a short period (11 weeks). This is most likely due to the high similarity between the BC training and cycling outdoors, but it does not rule out the importance of the slower process of building (and losing) muscular strength/power. Declarations Ethics Approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the faculty of Social Sciences (ECSS) of the Radboud University (ECSS-2020-132). All participants gave their written informed consent for inclusion before they participated in the study. This informed consent was obtained in two stages: (1) by completing an online questionnaire in which candidates requested for participation, and (2) by reaffirming this request in a second questionnaire (distributed via email) after the information session. Consent for publication I did not ask the participants to give written informed consent for their personal or clinical details along with any identifying images to be published in this study. In line with this, no identifying information about the participants is included in this paper. Figure 2 shows a photograph showing three research assistants operating the bicycle simulator. I obtained a written informed consent for publication from these research assistants. Availability of data and materials In the Supplementary Materials, I share the dataset supporting the conclusions of this article and the questionnaire used to measure the outcome variable. Competing interests The authors declare that they have no competing interests. Funding This research was funded by the Radboud Center Social Sciences. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Authors' contributions This is a single author paper. Acknowledgements The author would like to thank Mauritz Kleyn and Mirna Linden for their assistance with the spinning training and Maartje Floris and Loes van Leeuwen for their assistance with the bicycle simulator training and the data collection. References Mobycon. Research facts and numbers ‘Cycling for everyone’ [Onderzoek feiten en cijfers ‘Fietsen voor Iedereen’]. Delft: Mobycon, commissioned by the Ministery of Infrastructure and Water Management.; 2022. VZinfo.nl. Bilthoven: RIVM; 2024 [cited 2024 25 January]. Available from: https://www.vzinfo.nl/levensverwachting/leeftijd-en-geslacht#Levensverwachting. De Hartog JJ, Boogaard H, Nijland H, Hoek G. Do the health benefits of cycling outweigh the risks? Environmental health perspectives. 2010;118(8):1109-16. Oja P, Titze S, Bauman A, De Geus B, Krenn P, Reger‐Nash B, et al. Health benefits of cycling: a systematic review. Scandinavian journal of medicine & science in sports. 2011;21(4):496-509. Åström KJ. Introduction to stochastic control theory: Courier Corporation; 2012. Todorov E. Optimality principles in sensorimotor control. Nature neuroscience. 2004;7(9):907-15. Todorov E, Jordan MI. Optimal feedback control as a theory of motor coordination. Nature neuroscience. 2002;5(11):1226-35. Wolpert DM. Computational approaches to motor control. Trends in cognitive sciences. 1997;1(6):209-16. Scott SH. Optimal feedback control and the neural basis of volitional motor control. Nature Reviews Neuroscience. 2004;5(7):532-45. Wolpert DM, Ghahramani Z, Jordan MI. An internal model for sensorimotor integration. Science. 1995;269(5232):1880-2. Maris E. A bicycle can be balanced by stochastic optimal feedback control but only with accurate speed estimates. PLoS one. 2023;18(2):e0278961. Maris E. Internal sensory models allow for balance control using muscle spindle acceleration feedback. Neural Networks. 2025;189:107571. Epub 20250516. doi: 10.1016/j.neunet.2025.107571. PubMed PMID: 40412019. Giboin L-S, Gruber M, Kramer A. Task-specificity of balance training. Human movement science. 2015;44:22-31. Kümmel J, Kramer A, Giboin L-S, Gruber M. Specificity of balance training in healthy individuals: a systematic review and meta-analysis. Sports Medicine. 2016;46:1261-71. Meijaard JP, Papadopoulos JM, Ruina A, Schwab AL. Linearized dynamics equations for the balance and steer of a bicycle: a benchmark and review. Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences. 2007;463(2084):1955-82. Debanne D, Inglebert Y, Russier M. Plasticity of intrinsic neuronal excitability. Current opinion in neurobiology. 2019;54:73-82. Magee JC, Grienberger C. Synaptic plasticity forms and functions. Annual review of neuroscience. 2020;43:95-117. Mozzachiodi R, Byrne JH. More than synaptic plasticity: role of nonsynaptic plasticity in learning and memory. Trends in neurosciences. 2010;33(1):17-26. CROW-Fietsberaad. Updated recommendations for cycle path width 2022 [Geactualiseerde aanbevelingen voor de breedte van fietspaden 2022]. In: Waterstaat] MoIaWMMvIe, editor. Ede: CROW-Fietsberaad; 2022. Sawilowsky SS. New effect size rules of thumb. Journal of modern applied statistical methods. 2009;8(2):26. Johnstone D. How does your cycling power output compare? 2018 [cited 2024 29-2]. Available from: https://www.cyclinganalytics.com/blog/2018/06/how-does-your-cycling-power-output-compare. Moreland JD, Richardson JA, Goldsmith CH, Clase CM. Muscle weakness and falls in older adults: a systematic review and meta‐analysis. Journal of the American Geriatrics Society. 2004;52(7):1121-9. Muehlbauer T, Gollhofer A, Granacher U. Associations between measures of balance and lower-extremity muscle strength/power in healthy individuals across the lifespan: a systematic review and meta-analysis. Sports medicine. 2015;45:1671-92. Boulgarides LK, McGinty SM, Willett JA, Barnes CW. Use of clinical and impairment-based tests to predict falls by community-dwelling older adults. Physical therapy. 2003;83(4):328-39. Pijnappels M, Bobbert MF, van Dieën JH. Push-off reactions in recovery after tripping discriminate young subjects, older non-fallers and older fallers. Gait & posture. 2005;21(4):388-94. Pijnappels M, Van der Burg J, Reeves ND, van Dieën JH. Identification of elderly fallers by muscle strength measures. European journal of applied physiology. 2008;102:585-92. Proske U, Gandevia SC. The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiological reviews. 2012. Additional Declarations No competing interests reported. Supplementary Files BackOnTheBikeData4Sharing.csv CyclingSkillQuestionnaireBackOnTheBike2023.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7765021","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":550003434,"identity":"9581c84d-68d9-42b6-aa4c-edc46ce2aa40","order_by":0,"name":"Eric Maris","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYFACxgYwxcbAfECCgYEZLi5DhBa2BBQtPERYx2NAnBZ+6cONHxhz7PL5+M98vPFxh7Wcef8Bts88DHdwapHsS2yWYNyWbNkmkbvZcuaZdGOZGwnMs3kYnuHUYnCGsY2BcRuzAZsE7zZp3rbDiTMkGJiZeRgO49RiD9FSb8DGf+YZSEv9DP4D+LUY8IC1HDZgY8hhA2kBBlsCfi0SZxibJRK3HQc6LM3YcmZbuuEMicRmxjkGuLXw97A//PBxW7WBfP/hhzc+tlnLS/AfPszwpuKwHC4tYJCAygVFrgFeDaNgFIyCUTAKCAAAy8lGUw/ty8sAAAAASUVORK5CYII=","orcid":"","institution":"Radboud University","correspondingAuthor":true,"prefix":"","firstName":"Eric","middleName":"","lastName":"Maris","suffix":""}],"badges":[],"createdAt":"2025-10-02 08:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7765021/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7765021/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97460176,"identity":"74ba9712-5ec8-4cc8-9309-dafc4f5f9ae9","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8353347,"visible":true,"origin":"","legend":"","description":"","filename":"BackOnTheBike20250915v14.docx","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/1ae7c886e2f0ce62d8112263.docx"},{"id":97460167,"identity":"1650986d-7899-4653-9116-2fa8f06a939a","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3501,"visible":true,"origin":"","legend":"","description":"","filename":"397a8601210040e298b2cd9d4c1f690f.json","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/fcbee426e8f605f45057e7c6.json"},{"id":97460169,"identity":"bac37800-1a01-4a66-88b9-7f0ddf68c856","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"csv","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":739,"visible":true,"origin":"","legend":"","description":"","filename":"BackOnTheBikeData4Sharing.csv","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/242ae8d6656141eba16160e9.csv"},{"id":97669179,"identity":"6a12ad08-bef3-4ff8-b110-5ef4593e6a0f","added_by":"auto","created_at":"2025-12-08 09:27:31","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19335,"visible":true,"origin":"","legend":"","description":"","filename":"CyclingSkillQuestionnaireBackOnTheBike2023.docx","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/c2deee99a297cd193a5317f9.docx"},{"id":97460172,"identity":"b2c50932-9c32-435e-840f-f41fe882b662","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"xml","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":72166,"visible":true,"origin":"","legend":"","description":"","filename":"397a8601210040e298b2cd9d4c1f690f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/64a339fb3c192309f5226d1a.xml"},{"id":97460180,"identity":"0aeb5a30-934b-47b4-ad5b-85758f1c600f","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"emf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2565784,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.emf","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/fb3ecd33b28565b52d961633.emf"},{"id":97668380,"identity":"8e709539-78e5-43a2-8b05-ff14452b75ea","added_by":"auto","created_at":"2025-12-08 09:25:26","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4972755,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/4ddb336e9372960ac476f3d7.jpeg"},{"id":97460182,"identity":"4d0aef4b-769f-4d19-b280-7e30bfb50528","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2721647,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/65a656ea8ffc39fe86ba4007.jpeg"},{"id":97669064,"identity":"921abc7d-2340-47b5-ab86-5780a0a3514f","added_by":"auto","created_at":"2025-12-08 09:27:10","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22454,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/bda421e2f25014517d4f75cd.png"},{"id":97669696,"identity":"f03f6999-1cce-4c0e-b87b-c7ce8f56c360","added_by":"auto","created_at":"2025-12-08 09:28:42","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":716154,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/82959f32d1b74c79833e8e37.png"},{"id":97460174,"identity":"7d109f57-73d2-4941-94b4-20c3255100a9","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20471,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/9fc41c8ddd8175c6a21d283e.png"},{"id":97669670,"identity":"a7d41455-8aec-499f-92ed-10d11c14c565","added_by":"auto","created_at":"2025-12-08 09:28:39","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":69531,"visible":true,"origin":"","legend":"","description":"","filename":"397a8601210040e298b2cd9d4c1f690f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/4e7850f79e9db871497716d0.xml"},{"id":97460179,"identity":"23d14793-91c4-459d-a595-5dd808359337","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"html","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78352,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/dc73cd0569ad37c28bba100f.html"},{"id":97460166,"identity":"b65eff80-4ec1-4c69-a966-286c834d6cfa","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":120611,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eA closed-loop feedback control model for bicycle balance.\u003c/strong\u003e The model has four components: (1) a computational system, (2) a motor output system, (3) a mechanical system, and (4) a sensory system. See text, for a description.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/e4b6e3ce2daba19cbec6bef6.png"},{"id":97460171,"identity":"2d020b8e-4395-4210-8baf-19ed84970b98","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":958833,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eActor riding on the bicycle simulator while being supported by trainers.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/0b485ee27796d5db44ed4cce.png"},{"id":97668246,"identity":"a0b2049d-b2db-4cbe-a041-03ac5b160163","added_by":"auto","created_at":"2025-12-08 09:25:07","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":116501,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRaw outcome measures of the experimental group (a small random jitter was added to see overlapping lines).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/29949a90c0466db4f1172598.png"},{"id":97773170,"identity":"84f577d2-e70f-420a-b546-b367b4374800","added_by":"auto","created_at":"2025-12-09 08:24:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1865840,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/5a5eed5e-4637-41f1-9b6d-8f9dfc9d0067.pdf"},{"id":97460165,"identity":"b0c56342-4c3c-4aad-9f5b-ccc3f76968b0","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"csv","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":739,"visible":true,"origin":"","legend":"","description":"","filename":"BackOnTheBikeData4Sharing.csv","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/cd0321839d8c034fc2fb0fd7.csv"},{"id":97460168,"identity":"736dce3f-8346-4285-94b3-43dcd56ff171","added_by":"auto","created_at":"2025-12-04 15:20:02","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":19335,"visible":true,"origin":"","legend":"","description":"","filename":"CyclingSkillQuestionnaireBackOnTheBike2023.docx","url":"https://assets-eu.researchsquare.com/files/rs-7765021/v1/1c7b29c66e7d4947e369add9.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Getting older adults back on their bicycle; a pretest-posttest case-control study on the improvement of bicycle balance control","fulltext":[{"header":"Introduction","content":"\u003cp\u003eWith age, there is a sharp decrease in the percentage of citizens that still ride their bicycle. This is well documented in the Netherlands [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e], where everyone learns to cycle at a young age, and cycling is the default means of transportation for short distances. In the Netherlands, the percentage of inhabitants that cycles at least several times per month fluctuates around 77% between age 30 and 70 [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. However, over an average period of 15 years, this percentage decreases sharply to 39% [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Using life expectancy statistics [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], one can calculate that the average non-cycling period is 16 years.\u003c/p\u003e\u003cp\u003eThe physical health benefits of cycling are well documented [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. For instance, there is a strong inverse relationship between commuter cycling and all-cause mortality [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Because pedaling force can be easily controlled, cycling is an excellent training stimulus for the improvement of lower body strength and power. In addition, it has a very low impact on the joints, and it offers increased mobility options that are necessary for a meaningful social life.\u003c/p\u003e\u003cp\u003eIn this paper, I describe, motivate, and evaluate a rehabilitation intervention focused on balance control (BC) skills that aims at getting older adults back on their bicycle. BC can be understood from the perspective of optimal feedback control (OFC) [\u003cspan additionalcitationids=\"CR6 CR7 CR8 CR9 CR10 CR11\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], which is the dominant theory in the neural control of movement (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). OFC distinguishes between four determinants of BC skill: (1) a computational system that computes control signals and is implemented in the neural networks of our central nervous system (CNS), (2) a motor output system (part of the somatic nervous system; SNS) that sends the control signals to the muscles via the efferent axons of the spinal motor neurons, (3) a mechanical system, characterized by a.o. joint strength and range-of-motion, as implemented by the muscles, tendons, and cartilage, and (4) a sensory system (part of the SNS) that receives feedback from the eyes, the inner ear (labyrinth, vestibular organ) and the proprioceptive sensory organs (muscle spindles and Golgi tendon organs) and transmits this to the CNS.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eOFC is consistent with the fact that the effects of different types of balance training are specific to the balance task that was trained [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Specifically, OFC claims that, separately for every balance task, the CNS optimizes BC under the constraints/limitations of the mechanical and the somatic nervous system (torque and range limits of the mechanical system, precision of the sensory feedback, motor noise \u0026hellip;). This differs from most other views on BC that are found in the literature. These views assume a set of component BC skills that, depending on the set of balance task requirements, are recruited in different combinations. Under this assumption, training of component BC skills affects performance in all BC tasks that depend on these component skills. OFC, on the other hand, states that BC skills are task-specific, and claims that the generalization of the training effect scales with the similarity between the task used in the training and the objective of the training (here, bicycle BC on the public roads). Crucially, this similarity is defined at the level of the task and does not require a task analysis in terms of component skills.\u003c/p\u003e\u003cp\u003eTo maximize generalization of the training effect to bicycle BC on the public roads, I started from the mechanics of bicycle BC. The mechanics of bicycle BC pertains to two centers of gravity (CoGs) that both must be kept above their respective area of support (AoS): (1) the CoG of the rider-bicycle combination must be kept above the line that connects the two wheel-road contact points, and (2) the CoG of the rider\u0026rsquo;s upper body must be kept above the saddle [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. To achieve these two objectives, steering (turning the handlebars) is the most important control action [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. To maximize the generalization of the training effect, I used a bicycle simulator that requires steering to keep the two CoGs above their respective AoS.\u003c/p\u003e\u003cp\u003eBicycle BC skill is a continuum defined by perturbations that distort the balance. These perturbations will be called BC challenges. An important BC challenge are the upward reaction forces that result from pedaling, deform the upper body and perturb the steering. The impact of BC challenges is modulated by the cycling speed, and this follows from the mechanical properties of the bicycle: as the bicycle's speed increases, so do the bicycle's self-stabilizing forces [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eIn the bicycle BC training, I gradually increased the challenges of the BC task, slowly approaching the challenges of bicycle BC on the public roads. For that, I used the increase in BC challenge level with pedaling force (due to the upward reaction forces) and its decrease with speed (due to the self-stabilizing forces). This was possible thanks to the bicycle simulator, which contains a motor that drives the rear wheel. At the end of the training, when participants could maintain BC with a high pedaling force at a low speed, the bicycle simulator was used to practice a safe start-and-stop technique that had to be used when cycling outdoors.\u003c/p\u003e\u003cp\u003eFor a lost motor skill that must be restored, it is important to know to what extent it depends on (1) plasticity in the CNS (responsible for the mapping of sensory feedback onto control actions), and (2) plasticity in the organs that are responsible for registering feedback and executing actions. Plasticity in the CNS depends on the signaling properties of neurons, and these change over a much shorter timescale (hours, days) than the organs that are responsible for sensory registration and motor execution [\u003cspan additionalcitationids=\"CR17\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. To the best of my knowledge, plasticity in the sensory organs has not yet been demonstrated, and plasticity in the muscles with functional consequences (strength/power increase) operates at a timescale of at least weeks. To dissociate central (CNS-dependent) from muscular contributions to a possible improvement in BC skill, the training included sessions on exercise (spinning) bicycles on which the functional threshold power (FTP) could be monitored.\u003c/p\u003e\u003cp\u003eI hypothesize that the training effects generalize to bicycle BC on the public roads and that there is a separate CNS-dependent contribution to this effect.\u003c/p\u003e"},{"header":"Study setup and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eParticipants\u003c/h2\u003e\u003cp\u003eParticipants were recruited via an article in a local newspaper (De Gelderlander) about the goal of the study (March 31, 2023). This newspaper was distributed over an area with center roughly at the city of Nijmegen, the location of the intervention. The article invited participants that stopped cycling at least six months ago and wanted to get back on their bicycle. A total of 105 candidates replied, which were all invited for an information session on the Radboud University campus in Nijmegen. Based on their responses to a questionnaire, I excluded candidates that still cycled occasionally or suffered from balance-relevant sensory disorders (peripheral neuropathy, labyrinthitis, vestibular neuritis, \u0026hellip;). I selected participants based on how far they lived from the location of the intervention, the Radboud University campus. The sole motivation for this selection criterion was the desire to minimize travel time for participants that had to visit the campus multiple times (see \u003cem\u003eIntervention\u003c/em\u003e). I selected the 15 participants that lived closest to the campus.\u003c/p\u003e\u003cp\u003eThe intervention started with cycling on a stationary exercise bicycle (spinning) in the second week of June 2023. This part of the intervention was in the Radboud University Sports Center. On June 15, a second newspaper article was published in which the location of the intervention was revealed. This second article was a progress report on the study and did not ask for participation. However, after this publication, 8 additional candidates spontaneously joined the spinning sessions, and requested to participate in the study, to which I agreed. Four of these additional participants still cycled on rare occasions but considered quitting because of fear of falling.\u003c/p\u003e\u003cp\u003eA total of 23 participants started the BC training on the bicycle simulator, and 17 participants completed it (mean age 75,3 years, age range 66\u0026ndash;88 years). From the 6 participants that dropped out, 2 did so for a reason that was related to the BC training: one participant\u0026rsquo;s legs were so short that she could not touch the ground while sitting on the saddle, and one participant lacked a sufficient force to keep herself balanced on top of a stationary bicycle. Four participants dropped out for reasons that were unrelated to the BC training: a fracture due to a fall at home (two participants), bursitis, and an urgent eye surgery.\u003c/p\u003e\u003cp\u003eThe 17 participants that completed the BC training were matched to 17 candidates (the control group) that were not selected for the study because they were living farther away from the location of the intervention as compared to the experimental group. The matching criteria were gender and age.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eBicycle simulator\u003c/h3\u003e\n\u003cp\u003eThe bicycle simulator is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The core elements of the simulator are three rollers with a width of 1.75 m. This is likely to be close to the width of an average Dutch bicycle path; prior to 2022, the advised minimum width was 2 m, but 40 percent of the bicycle paths did not meet this requirement [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. The two front rollers are covered with a conveyer belt that is supported from underneath by a low-friction platform (not visible on Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The bicycle\u0026rsquo;s front wheel rests on the conveyer belt and the rear wheel rests on the rear roller. The rollers\u0026rsquo; rotational inertia is close to the linear inertia of the average Dutch rider-bicycle combination, and this ensures that the acceleration response to a pedal stroke feels natural. A torque motor is attached to the axis of the rear roller, and this motor is essential to assist weak participants in the first stages of the training. Finally, the bicycle simulator has a virtual-reality (VR) component that projects an animation of a bicycle lane (not visible on Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), which is controlled by the cycling speed and the cyclist\u0026rsquo;s lateral position on the left-right axis of the simulator, which was measured using a motion capture system.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIn between the front and the rear rollers, there is a platform on which the rider can place his feet and the trainers can stand. The rider has one trainer on each side, supporting him by means of gentle pushes on his shoulders. As the rider\u0026rsquo;s BC skills improve, the trainers provide less support and increase their distance from the rider.\u003c/p\u003e\u003cp\u003eThe rider wears a safety harness with which he is connected to a support construction above the rider\u0026rsquo;s head (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The safety harness does not prevent the rider from falling, but in case he would fall, his knees would not touch the floor before the suspension becomes active.\u003c/p\u003e\n\u003ch3\u003eIntervention\u003c/h3\u003e\n\u003cp\u003eThe intervention was over a period of 11 weeks. In that period, the participants engaged in (1) spinning, (2) BC training on the bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator. The three components of the intervention were also administered in this order, with spinning and BC training overlapping in time for the last 8 weeks.\u003c/p\u003e\n\u003ch3\u003eSpinning\u003c/h3\u003e\n\u003cp\u003eThe first three weeks of the intervention only involved a group spinning training, which was given by two professional spinning coaches and one amateur. Participants trained at an individually determined power level, controlled by the one-minute functional threshold power (1minFTP).\u003c/p\u003e\u003cp\u003eOne spinning session lasted 45 min., excluding warming-up and cooling-down, but including a 5 min. break in the middle. Every session started with a 3-minute block in which the 1minFTP was assessed. The remainder of the session had the structure of an interval training. The high intensity blocks lasted at most 1 min. and were alternated with low intensity blocks at 70 percent of 1minFTP that lasted at most 2,5 min. Increasing intensity was achieved either by increasing the resistance (and keeping the RPM fixed) or by increasing the RPM (and keeping the resistance fixed).\u003c/p\u003e\n\u003ch3\u003eAssessing the 1minFTP\u003c/h3\u003e\n\u003cp\u003eThe 1minFTP was used for a double purpose: (1) as a measure of leg power, to investigate its relationship with bicycle BC skill, and (2) to train at an individually determined power level. The 1minFTP is the highest power a participant can maintain for one minute while cycling at 60 revolutions per minute (RPM). The 1minFTP was assessed by having participants monitor their breath after a one-minute effort at some power level: if the participant could say only a few words at a time (a proxy for reaching the ventilatory threshold), the 1minFTP was reached.\u003c/p\u003e\u003cp\u003eAssessing the 1minFTP required keeping a constant high-power output, and this turned out to be very difficult for our participants. First, assessing the 1minFTP required that the participants could operate the bicycle computer that provided feedback about their power output. It took participants two weeks to learn to operate this computer. Second, it took several participants substantially more time to learn to produce a constant power output. Initially, all participants lowered their RPM when the resistance was increased above a certain level, thereby preventing a higher power output. Several participants continued doing this until the fourth week. Third and last, participants\u0026rsquo; performance was highly variable with respect to the maximum power output they endured, and this was both within sessions (no longer following the trainer\u0026rsquo;s instructions with respect to power output) and between sessions (lowering the 1minFTP at the beginning of a session if one was feeling tired).\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eBicycle simulator BC training\u003c/h2\u003e\u003cp\u003eBC training started in the fourth week of the intervention and was in one of the research labs of the Radboud University (Faculty of Social Sciences). The BC training was individual and was given by two research assistants. A BC training session lasted approximately 30 min. At the start, participants put on a safety harness and were connected to the support construction. The saddle height was adjusted such that the participant\u0026rsquo;s feet touched the platform in between the front and the rear rollers. For as long as the participant could not balance independently and/or showed signs of fear, the trainers gently pushed their hands against the participant\u0026rsquo;s shoulders. They gradually released the pressure as the participant\u0026rsquo;s BC became more independent, and he gained confidence.\u003c/p\u003e\u003cp\u003eThe guiding principle in the BC training was to increase the difficulty in small steps and following the participant\u0026rsquo;s progress and confidence. Every session started with a BC challenge level at which the participant was comfortable, and that level was slowly increased if the participant agreed. In the first session, the bicycle was brought up to a speed of approximately 12 km./h. by means of a torque motor. In this condition, the participant\u0026rsquo;s only task was to turn the handlebars for balance control. The BC challenge level was increased in four ways: (1) by decreasing the speed, (2) by decreasing the assistive force at the rear roller, (3) by asking participants to accelerate from standstill and to stop at any desired moment, and (4) by asking the participant to tap the support surface with both feet, place them back on the pedals, and to continue pedaling (feet tapping exercise).\u003c/p\u003e\u003cp\u003eWhen participants could balance independently while also pedaling, they practiced a safe start-and-stop technique in which the participant could always touch the floor with his feet. To achieve that, the saddle had to be set at the appropriate height, which was made easy by installing a dropper post in the seat tube. At standstill, the participant was seated on the saddle and had both feet on the ground. He started moving by applying a forceful pedal stroke with his favorite leg and continued with both legs. When stopping, the participant took his feet from the pedals and placed them on the stationary floor. When feeling adequately supported by his legs, the participant pulled the brakes, thereby stopping the rollers of the bicycle simulator. Participants were suggested to use the dropper post lever to increase the saddle height when they were up to speed and to lower it when transitioning to a stop. However, none of the participants opted for this.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eCycling outdoors\u003c/h3\u003e\n\u003cp\u003eThe participants were instructed to start cycling outdoors on their own bicycles when they could do 10 feet taps consecutively without stopping in between, and at a resistance level slightly above the resistance on the public roads. When cycling outdoors, the participants were instructed to use the safe start-and-stop technique (feet touching the road surface) that was practiced on the bicycle simulator, to start cycling in an environment without other road users (a park, a parking lot), and to gradually move to more busy environments. A few participants informed the trainers that they had started cycling outdoors before they were instructed to do so. Those participants were given the same instructions as the other participants.\u003c/p\u003e\n\u003ch3\u003eEffect evaluation\u003c/h3\u003e\n\u003cp\u003eThe effect of the intervention was evaluated according to the design of a pretest-posttest case-control study. In the experimental group, the outcome variable (see \u003cem\u003eOutcome variable\u003c/em\u003e) was measured twice, the first time in the beginning of the intervention and the second time six weeks after the end of the intervention. To correct for a possible spontaneous recovery, a control group was included that did not receive the intervention, but in which the change in the outcome variable was assessed over the same time interval as in the experimental group (see \u003cem\u003eParticipants\u003c/em\u003e).\u003c/p\u003e\u003cp\u003eThe effect of the intervention was evaluated statistically using a dependent samples t-test that compared the change in the outcome variable between the matched experimental and control group participants.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eOutcome variable\u003c/h2\u003e\u003cp\u003eThe outcome variable was obtained from a questionnaire that evaluated participants\u0026rsquo; bicycle BC skills and confidence while riding on the public roads. Because there was no existing questionnaire that measured the relevant dependent variable for this study, I developed a dedicated questionnaire, of which an English translation is added to the Supplementary Materials; the original questionnaire was in Dutch. To maximize the validity of this questionnaire, the questions were formulated in terms of easily observable criteria. The introduction to the questionnaire was the following:\u003c/p\u003e\u003cp\u003eThree environments will be described, and you must indicate whether you rode your bicycle in that environment in the past two weeks. By \u0026ldquo;riding your bicycle\u0026rdquo;, we mean starting from a standstill and stopping at a desired location. For every environment, you must indicate what applies to you: (1) I did not ride my bicycle here, (2) I rode my bicycle here, but I try to avoid it, and (3) I rode my bicycle here without problems.\u003c/p\u003e\u003cp\u003eThe three environments were the following: (1) without traffic (e.g., an empty parking lot, a quiet park), (2) with light traffic (e.g., small roads in an outside area, a bicycle lane on a quiet moment), and (3) in busy traffic (e.g., city center during traffic hours, busy crossroads with mixed traffic).\u003c/p\u003e\u003cp\u003eThe answers to every question/environment were converted into a score by assigning 0, 1, and 2 points to the respective options. Summing these scores over the three questions/environments results in a number between 0 and 6, which is the outcome variable.\u003c/p\u003e\u003cp\u003eIn the experimental group, this questionnaire was administered once in the beginning of the intervention and a second time six weeks after the end of the intervention. The questionnaire was sent as an attachment to an email, and participants were called on their phone if they had not replied after one week. For some participants, it was necessary to clarify some aspects of the questions. In the control group, the questionnaire was administered only once and at around the same time that the second questionnaire was sent to the experimental group. None of the control group participants was riding their bicycle at the start of the intervention and they were asked whether and how this had changed in the meantime.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eI conducted a pretest-posttest case-control study to evaluate the effectiveness of an intervention aimed at restoring bicycle BC skill. The experimental group consisted of 23 participants, of which 17 completed the intervention (see \u003cem\u003eMethods\u003c/em\u003e).\u003c/p\u003e\u003cp\u003eThe intervention lasted 11 weeks and involved three components: (1) cycling on a stationary exercise bicycle (spinning), (2) BC training on a bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator (see \u003cem\u003eMethods\u003c/em\u003e). Cycling on the public roads started when participants were successful in a performance-defining bicycle simulator task: making 10 consecutive feet taps without stopping in between, and at a resistance level slightly above the resistance on the public roads. At the start of the BC training, no participant could do this task, but at end of the training all participants that completed the training were successful.\u003c/p\u003e\u003cp\u003eThe effectiveness of the intervention was evaluated using an outcome variable that did not directly reflect the performance on a task that was extensively practiced during the training (e.g., the feet tapping task). Instead, the effectiveness was evaluated using an outcome variable that is much closer to cycling outdoors: a questionnaire that assesses the participant’s skill/confidence on his own bicycle when cycling outdoors (see \u003cem\u003eMethods\u003c/em\u003e). In the experimental group, this outcome measure was obtained twice, first in the beginning of the intervention, and then six weeks after the intervention. To correct for a possible spontaneous recovery of cycling skill/confidence, I used a matched control group that did not receive the intervention, but in which the change in cycling skill/confidence over the same time interval was assessed (see \u003cem\u003eMethods\u003c/em\u003e).\u003c/p\u003e\u003cp\u003eThe improvement in cycling skill/confidence was significantly larger in the experimental group (M = 3.1, SD = 2.1) compared to a matched control group (M = 0, SD = 0), t(16) = 6.0, p \u0026lt; .001, 95% CI [2.0 4.1]. Note that none of the control group participants had a spontaneous recovery of their cycling skill/confidence. The raw outcome measures of the experimental group are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The effect size (Cohen’s d) was 1.5, which is a very large effect [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. This effect was obtained with an average of 4.8 simulator BC training sessions of 30 min. each.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eIt was my original plan to dissociate central (CNS-dependent) and muscular contributions to a possible improvement in BC skill. For that, it was necessary to quantify the increase in maximum power output on the spinning bicycle, for which I used the one-minute functional threshold power (1minFTP), which was assessed at the beginning of every spinning session. However, keeping a constant high-power output (a requirement for a reliable FTP assessment) turned out to be very difficult for our participants and required several weeks of training (see \u003cem\u003eMethods\u003c/em\u003e). I conclude that I failed in obtaining a reliable quantification of the increase in maximum leg power increase in this group of participants.\u003c/p\u003e\u003cp\u003eIt is useful to report the mean 1minFTP on which participants converged after learning to produce a constant force output. This mean 1minFTP was 93.6 Watt (sd = 17.2). FTP is usually expressed relative to bodyweight, and in this participant group the average 1 minute power-to-weight ratio was 1.2 Watt/kg (sd = 0.3). For reference, it is useful to compare this value with published 1 minute power-to-weight ratios for untrained amateur cyclists [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] (defined as the 5-th percentile of the population of amateur cyclists): 3.9 for males and 3.3 for females. Thus, the participants in this study had an average 1minute power-to-weight ratio that is approximately three times smaller than that of an untrained cyclist.\u003c/p\u003e"},{"header":"Discussion and conclusions","content":"\u003cp\u003eA very large improvement in older cyclists’ BC skills and confidence can be realized by means of a short intervention (11 weeks), even if that person has quit cycling for a long time. This is remarkable given the task-specificity of the effects of different types of balance training [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], and is most likely due to the high similarity between the BC training on the simulator and cycling outdoors.\u003c/p\u003e\u003ch2\u003eLimitations\u003c/h2\u003e\u003cp\u003eThis study has three important limitations: (1) it used only an indirect measure of bicycle BC skills on the public roads, (2) it did not investigate whether the participants can deal successfully with external BC challenges, and (3) it could not distinguish between an effect at the level of the CNS and an effect at the level of the muscles. First, in the absence of an existing direct measure of BC skills on the public roads, I used a self-report measure. Although all questions were formulated in terms of observable criteria, a self-report is not a direct measure of the behavior of interest. Nevertheless, a self-report measure is to be preferred over the direct measures that were collected as a part of the BC training (e.g., the assistive force required to accelerate from standstill, the performance on the feet tapping task). These direct measures reflect the performance on tasks that were extensively practiced during training, and there is no information about their predictive value for BC skills on the public roads.\u003c/p\u003e\u003cp\u003eSecond, this study did not investigate whether the participants can deal successfully with some important external BC challenges such as an uneven road surface (e.g., curbstones, potholes) and sharp turns. It is very likely that several BC challenges require a sufficient lower body strength to prevent perturbations from affecting the steering movements. For example, (1) standing on the pedals while covering an obstacle, and (2) choosing an appropriate lean angle before entering a turn. In addition, cycling safely not only involves BC (staying upright) but also navigation (avoiding obstacles and/or following a track), which requires additional skills such as selective attention, anticipation, and planning. These have not been investigated.\u003c/p\u003e\u003cp\u003eThird and last, from the perspective of OFC, it is tempting to conclude that the improvement is mainly due to the fast plasticity in the CNS, and not to the comparatively slow plasticity in the muscles. However, to make a strong case for this claim, it is essential to have a reliable measure of maximum power output, and this turned out to be very difficult. The existing literature on the relation between, on the one hand, muscle strength, and on the other hand, the incidence of falls and balance task performance, is inconsistent: one systematic review (focusing on falls) [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] argued for a strong relation and another (focusing on balance task performance) [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] argued for a very weak one. This inconsistency is probably related to the fact that, in the population of interest, falls cannot be predicted from the performance on balance tasks [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e\u003ch2\u003eThe complicated relation between muscle strength, balance control, and falls\u003c/h2\u003e\u003cp\u003eFor several reasons, the relation between muscle strength, balance control, and falls, is a complicated one. First, many studies have been conducted in homogeneous groups (institutionalized participants, a small age range, …) and this restriction-of-range results in an underestimation of the correlation in the heterogeneous/unselected population. Crucially, restriction-of-range also operates at the individual level because most people are aware of the range of BC challenges that their BC skill allows, and they try to stay within this range. This restriction-of-range at the individual level reduces the correlation between, on the one hand, muscle strength and BC skills, and on the other hand, the incidence of falls. Experimental studies in which falls are elicited do not suffer from this restriction-of-range at the individual level, and this methodology has been applied successfully to investigate tripping-induced falls during walking [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. One such study found a convincing correlation between leg strength and the ability to recover from tripping [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eSecond, it is very likely that (1) there are large differences between muscle groups and joint ranges-of-motion with respect to their relevance for BC, and (2) this differential relevance depends on the BC task. BC during walking, such as the ability to recover from tripping, must depend on leg strength because the leg muscles are used to bring the body CoG back over its AoS. However, in bicycle BC, different muscle groups can be used to control the combined rider-bicycle and the upper body CoG: (1) by controlling the handlebars, the arm muscles can bring the combined rider-bicycle CoG over its LoS, and (2) by controlling the upper body CoG, the core muscles in the trunk and the hip bring this CoG over its AoS, the saddle. Thus, the relevance of a particular muscle group for balance control likely depends on the BC task (walking versus cycling).\u003c/p\u003e\u003cp\u003eThird and last, BC not only depends on the muscles for bringing the CoG above its AoS (the muscles’ motor role) but also for informing the CNS about the position of the CoG relative to the AoS, and this depends on the muscle spindles (the muscles’ sensory feedback role) [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. To be useful, muscle spindle output must reflect the tension in the force-producing fibers, but its precise contribution to the loss of BC is unknown.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eI have demonstrated that it is possible to realize a very large improvement in an older cyclist\u0026rsquo;s BC skills and confidence over a short period (11 weeks). This is most likely due to the high similarity between the BC training and cycling outdoors, but it does not rule out the importance of the slower process of building (and losing) muscular strength/power.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics Approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThe study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the faculty of Social Sciences (ECSS) of the Radboud University (ECSS-2020-132). All participants gave their written informed consent for inclusion before they participated in the study. This informed consent was obtained in two stages: (1) by completing an online questionnaire in which candidates requested for participation, and (2) by reaffirming this request in a second questionnaire (distributed via email) after the information session.\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eI did not ask the participants to give written informed consent for their personal or clinical details along with any identifying images to be published in this study. In line with this, no identifying information about the participants is included in this paper. Figure 2 shows a photograph showing three research assistants operating the bicycle simulator. I obtained a written informed consent for publication from these research assistants.\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eIn the Supplementary Materials, I share the dataset supporting the conclusions of this article and the questionnaire used to measure the outcome variable.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis research was funded by the Radboud Center Social Sciences.\u0026nbsp;The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.\u003c/p\u003e\n\u003cp\u003eAuthors' contributions\u003c/p\u003e\n\u003cp\u003eThis is a single author paper.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThe author would like to thank Mauritz Kleyn and Mirna Linden for their assistance with the spinning training and Maartje Floris and Loes van Leeuwen for their assistance with the bicycle simulator training and the data collection.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMobycon. Research facts and numbers \u0026lsquo;Cycling for everyone\u0026rsquo; [Onderzoek feiten en cijfers \u0026lsquo;Fietsen voor Iedereen\u0026rsquo;]. Delft: Mobycon, commissioned by the Ministery of Infrastructure and Water Management.; 2022.\u003c/li\u003e\n\u003cli\u003eVZinfo.nl. Bilthoven: RIVM; 2024 [cited 2024 25 January]. Available from: https://www.vzinfo.nl/levensverwachting/leeftijd-en-geslacht#Levensverwachting.\u003c/li\u003e\n\u003cli\u003eDe Hartog JJ, Boogaard H, Nijland H, Hoek G. Do the health benefits of cycling outweigh the risks? Environmental health perspectives. 2010;118(8):1109-16.\u003c/li\u003e\n\u003cli\u003eOja P, Titze S, Bauman A, De Geus B, Krenn P, Reger‐Nash B, et al. Health benefits of cycling: a systematic review. Scandinavian journal of medicine \u0026amp; science in sports. 2011;21(4):496-509.\u003c/li\u003e\n\u003cli\u003e\u0026Aring;str\u0026ouml;m KJ. Introduction to stochastic control theory: Courier Corporation; 2012.\u003c/li\u003e\n\u003cli\u003eTodorov E. Optimality principles in sensorimotor control. Nature neuroscience. 2004;7(9):907-15.\u003c/li\u003e\n\u003cli\u003eTodorov E, Jordan MI. Optimal feedback control as a theory of motor coordination. Nature neuroscience. 2002;5(11):1226-35.\u003c/li\u003e\n\u003cli\u003eWolpert DM. Computational approaches to motor control. Trends in cognitive sciences. 1997;1(6):209-16.\u003c/li\u003e\n\u003cli\u003eScott SH. Optimal feedback control and the neural basis of volitional motor control. Nature Reviews Neuroscience. 2004;5(7):532-45.\u003c/li\u003e\n\u003cli\u003eWolpert DM, Ghahramani Z, Jordan MI. An internal model for sensorimotor integration. Science. 1995;269(5232):1880-2.\u003c/li\u003e\n\u003cli\u003eMaris E. A bicycle can be balanced by stochastic optimal feedback control but only with accurate speed estimates. PLoS one. 2023;18(2):e0278961.\u003c/li\u003e\n\u003cli\u003eMaris E. Internal sensory models allow for balance control using muscle spindle acceleration feedback. Neural Networks. 2025;189:107571. Epub 20250516. doi: 10.1016/j.neunet.2025.107571. PubMed PMID: 40412019.\u003c/li\u003e\n\u003cli\u003eGiboin L-S, Gruber M, Kramer A. Task-specificity of balance training. Human movement science. 2015;44:22-31.\u003c/li\u003e\n\u003cli\u003eK\u0026uuml;mmel J, Kramer A, Giboin L-S, Gruber M. Specificity of balance training in healthy individuals: a systematic review and meta-analysis. Sports Medicine. 2016;46:1261-71.\u003c/li\u003e\n\u003cli\u003eMeijaard JP, Papadopoulos JM, Ruina A, Schwab AL. Linearized dynamics equations for the balance and steer of a bicycle: a benchmark and review. Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences. 2007;463(2084):1955-82.\u003c/li\u003e\n\u003cli\u003eDebanne D, Inglebert Y, Russier M. Plasticity of intrinsic neuronal excitability. Current opinion in neurobiology. 2019;54:73-82.\u003c/li\u003e\n\u003cli\u003eMagee JC, Grienberger C. Synaptic plasticity forms and functions. Annual review of neuroscience. 2020;43:95-117.\u003c/li\u003e\n\u003cli\u003eMozzachiodi R, Byrne JH. More than synaptic plasticity: role of nonsynaptic plasticity in learning and memory. Trends in neurosciences. 2010;33(1):17-26.\u003c/li\u003e\n\u003cli\u003eCROW-Fietsberaad. Updated recommendations for cycle path width 2022 [Geactualiseerde aanbevelingen voor de breedte van fietspaden 2022]. In: Waterstaat] MoIaWMMvIe, editor. Ede: CROW-Fietsberaad; 2022.\u003c/li\u003e\n\u003cli\u003eSawilowsky SS. New effect size rules of thumb. Journal of modern applied statistical methods. 2009;8(2):26.\u003c/li\u003e\n\u003cli\u003eJohnstone D. How does your cycling power output compare? 2018 [cited 2024 29-2]. Available from: https://www.cyclinganalytics.com/blog/2018/06/how-does-your-cycling-power-output-compare.\u003c/li\u003e\n\u003cli\u003eMoreland JD, Richardson JA, Goldsmith CH, Clase CM. Muscle weakness and falls in older adults: a systematic review and meta‐analysis. Journal of the American Geriatrics Society. 2004;52(7):1121-9.\u003c/li\u003e\n\u003cli\u003eMuehlbauer T, Gollhofer A, Granacher U. Associations between measures of balance and lower-extremity muscle strength/power in healthy individuals across the lifespan: a systematic review and meta-analysis. Sports medicine. 2015;45:1671-92.\u003c/li\u003e\n\u003cli\u003eBoulgarides LK, McGinty SM, Willett JA, Barnes CW. Use of clinical and impairment-based tests to predict falls by community-dwelling older adults. Physical therapy. 2003;83(4):328-39.\u003c/li\u003e\n\u003cli\u003ePijnappels M, Bobbert MF, van Die\u0026euml;n JH. Push-off reactions in recovery after tripping discriminate young subjects, older non-fallers and older fallers. Gait \u0026amp; posture. 2005;21(4):388-94.\u003c/li\u003e\n\u003cli\u003ePijnappels M, Van der Burg J, Reeves ND, van Die\u0026euml;n JH. Identification of elderly fallers by muscle strength measures. European journal of applied physiology. 2008;102:585-92.\u003c/li\u003e\n\u003cli\u003eProske U, Gandevia SC. The proprioceptive senses: their roles in signaling body shape, body position and movement, and muscle force. Physiological reviews. 2012.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Balance control, bicycle balance, motor skill training, bicycle simulator, older adults, elderly, mobility, intervention study, case-control","lastPublishedDoi":"10.21203/rs.3.rs-7765021/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7765021/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper investigates the effectiveness of an intervention that aims at restoring bicycle balance control skills in older adults that have quit cycling. The intervention lasted 11 weeks and involved three components: (1) training on an exercise bicycle, (2) a balance control training on a bicycle simulator, and (3) cycling on the public roads with a safe start-and-stop technique that was practiced on the bicycle simulator. In the bicycle simulator training, the difficulty of the balance control was increased gradually, slowly approaching the difficulty of bicycle balance control on the public roads. The experimental group consisted of 23 community-dwelling older adults that had quit cycling (N\u0026thinsp;=\u0026thinsp;19) or that were on the verge of doing so (N\u0026thinsp;=\u0026thinsp;4). The effectiveness of the intervention was evaluated by comparing the experimental with a matched control group on a posttest-minus-pretest difference in balance control skill and confidence. The intervention produced a very large improvement (Cohen\u0026rsquo;s d\u0026thinsp;=\u0026thinsp;1.5, t(16)\u0026thinsp;=\u0026thinsp;6.0, p\u0026thinsp;\u0026lt;\u0026thinsp;.001) in balance control skills and confidence on the public roads. This very large improvement does not rule out the importance of the slower process of acquiring a sufficient lower body strength. This study was retrospectively registered with identifier NCT07195526 on ClinicalTrials.gov at 19-09-2025.\u003c/p\u003e","manuscriptTitle":"Getting older adults back on their bicycle; a pretest-posttest case-control study on the improvement of bicycle balance control","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-04 15:19:57","doi":"10.21203/rs.3.rs-7765021/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"eef42fc3-555e-4952-af08-98f36a44346d","owner":[],"postedDate":"December 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-09T08:23:41+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-04 15:19:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7765021","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7765021","identity":"rs-7765021","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}