Proprioceptive exercises: standing on one leg, closing your eyes, stepping onto a board that wobbles. It is one of the few things almost all physiotherapists prescribe and almost none of them explain properly. This guide sets out what the research has actually shown, what is instead widespread belief, and how to build a progression that makes sense for your situation.
I will give the conclusion up front, because it is the part that is usually missing: the best-documented effect of proprioceptive exercises concerns the ankle, and it is real but more modest than the "50% fewer sprains" you read everywhere. On the rest, sports performance, balance "in general", unstable surfaces being superior to everything, the evidence is much weaker than the marketing of the equipment suggests.
⚠ Important: read this before continuing
The exercises and advice in this article are general guidance and may not suit your situation. Everyone has a unique clinical history: what works for others could be counterproductive for you.
Want a professional's opinion? Start with the questionnaire: you sign up and see who will follow you, a physiotherapist registered with the Albo chosen for your case, whom you message in chat. If you would rather hear a voice, the first consultation is free and we look at your situation together.
First of all: when the missing balance is not a training problem
Losing stability is not always a matter of ankles that need strengthening. Balance comes from the combination of three systems, proprioceptive, vestibular and visual, and from the way the brain puts them together. If one of these is faulty, no wobbling board will solve the problem: a diagnosis comes first.
See a doctor, without going through the exercises, if you notice even one of these signs:
- instability that appeared suddenly, over hours or days, with no trauma to explain it;
- tingling, numbness or weakness in the feet, legs or hands, especially if it is progressing or symmetrical;
- true vertigo (the room spinning), nausea, tinnitus or hearing loss on one side;
- disturbances of vision, speech or swallowing, an asymmetrical face, or an unusual and sudden headache: these are emergency-room signs;
- repeated falls with no apparent reason, particularly after the age of 65 or if you take medication that causes drowsiness or lowers blood pressure;
- marked worsening in the dark or with eyes closed, which can indicate peripheral neuropathy, common in diabetes and vitamin B12 deficiency;
- fever, weight loss, night pain or bladder and bowel control problems associated with the instability.
None of these signs automatically means something serious, but none of them is treated with single-leg balance work. The rest of the article deals with the most common situation: a joint that no longer trusts itself after an injury, or a balance that has gone rusty.
What proprioception is (and why the name is a bit misleading)
It is called many things, proprioceptive training, proprioceptive exercises, proprioceptive rehabilitation, balance training, and in practice they all point to the same family of work. "Proprioceptive gymnastics" is the name that circulates most outside clinics; here we use "exercises" because that is the term used in the studies, but if you are searching for the former you are in the right place.
Proprioception is the ability to sense the position, movement and force of your body segments without looking at them. It is why you can touch your nose with your eyes closed, or realise your foot is landing on a stone before you see it.
It depends on receptors spread across several tissues: the neuromuscular spindles in the muscle, sensitive to length and speed of stretch; the Golgi tendon organs, sensitive to tension; the capsular and ligamentous receptors of the joint; and the cutaneous mechanoreceptors, which under the sole of the foot contribute more than you might imagine. The central nervous system fuses all this information with the vestibular and visual input and produces a motor response.
The name misleads because it suggests a "sense" that can be trained like a muscle. When we work on balance, we are not really improving the receptors: we are reorganising the way the nervous system uses the information it already receives, and training motor strategies. It is a difference that sounds theoretical, but it explains why the results are much more specific to the task trained than people believe. And those strategies are not acquired in adulthood: rolling, landing, holding yourself up on one foot are learned as children, and in the studies they do not arrive with growth alone if nobody shows them: basic motor patterns, somersault included.
What is really proven
This is the core of the article, and also the point where most online content takes a shortcut.
Preventing ankle sprains is the best-documented effect. A systematic review with meta-analysis of seven randomised trials of moderate to high quality, for a total of 3,726 sporting participants (3,654 in the analysis combining them all), found a significant reduction in sprains in the groups doing proprioceptive training compared with controls, which in the included studies were not "nothing" but a heterogeneous set of comparison interventions: relative risk 0.65 (95% confidence interval 0.55-0.77). In plain terms: roughly a third fewer sprains, not half. In the clinical reading of the same data, the number of people to train to avoid one sprain is 17 (range 11-33).
The benefit is more solid in people who have already been injured. In athletes with a history of sprains the relative risk is 0.64 (0.51-0.81), with a number needed to treat of 13, but with a wide interval, from 7 to 100. On primary prevention, meaning in people who have never sprained an ankle, the pooled result is formally significant (relative risk 0.57, 0.34-0.97) but comes from two studies which, taken individually, were not; the number needed to treat rises to 33 with an interval running from 16 to 1,000: a technical way of saying that we know almost nothing precisely. The authors indeed conclude that the evidence remains inconclusive. If you have never hurt yourself, then, proprioceptive exercises remain reasonable, but nobody can promise you a certain benefit.
After an acute sprain, continuity is needed. A systematic review on postural control and lateral ankle instability observed that completing at least six weeks of balance training after an acute sprain substantially reduces the risk of relapse. The same review does, however, flag two uncomfortable points: improvements in instrumental measures of postural control were not consistent, and in people with chronic ankle instability already established the evidence is insufficient to conclude that training reduces recurrences.
It is not proven that they beat the brace. A meta-analysis that directly compared proprioceptive and neuromuscular training with bracing or taping for reducing recurrence in athletes found no difference at twelve months, and recommends choosing according to the person's preference and the experience of the therapist. It should be taken for what it is: only three studies, of variable methodological quality, and "no difference found" is not the same as "they are equivalent": with so few data, a modest difference would not show up anyway. The practical message still holds: if you are returning to play, the brace and the exercises are not enemies, and neither of them is "the" answer.
On the knee, the credit does not all go to proprioception. Anterior cruciate ligament prevention programmes work: a meta-analysis of eight meta-analyses reports a 50% reduction in the risk of all cruciate ligament injuries across the pool of athletes, and a 67% reduction in non-contact injuries in female athletes. But those programmes, in practice, are multi-component packages: strength, plyometrics, correction of landing and change-of-direction technique, as well as balance. That meta-analysis sums the results of entire programmes and does not separate the contribution of individual components: attributing that result to proprioceptive exercises alone would be incorrect. And there is a population limit to keep in mind: six of the eight meta-analyses included contained data only on female athletes, to the point that the authors themselves conclude that the data on male athletes are not enough to draw conclusions.
On falls in older people the evidence exists, with one condition. A review with meta-analysis of 88 studies and 19,478 participants found that exercise reduces the rate of falls by 21% in older people living at home; the largest effects are seen in programmes that genuinely challenge balance and exceed three hours a week, a combination that brings the reduction to 39%. The condition, precisely, is the dose. The same work found no preventive effect in residential care facilities, or in people recovering from a stroke or just discharged from hospital, while it did find one, again at home, in people with Parkinson's disease or cognitive decline.
What is not proven: general transfer. This is the point that overturns the common idea. A meta-analysis carried out on healthy people measured how much balance training improves untrained tasks: on the trained task the effect is large (standardised mean difference 0.79), on untrained tasks the effects range between -0.07 and 0.18, that is, no effect. The authors go as far as writing that generic tasks such as standing on one leg have little value as an overall measure of balance. It is a meta-analysis of six studies and 102 subjects, so it is small: it should be taken as a strong warning, not as a law. But the warning is clear.
What follows, in practice? That the phrase "proprioceptive exercises improve performance in any sport" has no evidence behind it up to the mark, and that training balance "in general" in the hope that it will then transfer to your sporting movement is a gamble. It is better to choose exercises that resemble the task you care about: the position, the direction of the perturbation, the speed.
An important piece of advice: before trying these exercises, make sure you have received a correct diagnosis. People often think they have one problem when the real cause of the pain is another. Doing the wrong exercises can make the situation worse instead of better.
The five levers of progression
Before the list it helps to understand the controls. All proprioceptive exercises can be made harder by acting on five independent variables. Changing one at a time is what distinguishes a programme from a collection of exercises.
1. Base of support. Two feet side by side, then together, then in line (tandem), then a single foot. It is the simplest lever and the most underrated.
2. Visual information. Eyes open, then gaze fixed on a point, then eyes closed. Removing sight forces the system to rely on the other channels. Caution: if you get much worse with your eyes closed, speak to a professional first (see the signs above).
3. Surface. Stable ground, then a mat or cushion, then a board or hemisphere. Caution: in the studies cited here the unstable surface is never the isolated variable compared with the others, so there is no proof that "unstable" automatically means better. In my experience it is simply different, and it is the lever people tend to overuse.
4. Motor task. Still, then in motion, then with a second task (counting backwards, following an object with your eyes, bouncing a ball). The dual task brings the situation closer to real life, where balance never has all of your attention.
5. Perturbation. First self-generated (move your arms, shift your weight), then external and predictable, then external and unpredictable: a light push from a partner, a ball thrown from different directions.
A rule I use in the clinic: one lever at a time, and you move on when the current task stays clean and controlled, not when it is merely tolerable. The durations that follow are practical conventions, not dosages proven by the studies: use quality as the criterion, not the stopwatch.
Basic level
These are the proprioceptive exercises that build the vocabulary of movement, and for many people they are already sufficient.
Two-legged with eyes closed. Standing, feet at hip width, close your eyes beside a wall or a table. Notice the sways and how you correct them with the ankle. Stop when you need to open your eyes or lean on something.
Feet together, then tandem. Bring your feet together until they touch, then place them in line, heel of one against toes of the other. Each step reduces the base of support and increases the lateral demand.
Weight shifts. Standing, slowly move the load forward and back and then from side to side, exploring the limits within which you can stay without lifting your feet. It helps you discover where your zone of control ends.
Standing on one leg. The classic. Soft knee, level pelvis, gaze forward. At first with a finger resting on the wall: gradually removing it is already a progression.
Intermediate level
One-legged with eyes closed. The combination of the first two levers, and the sharpest jump in difficulty of the whole scale. Always near a support.
On a soft surface. Cushion, rolled-up mat, foam pad. First on two feet, then in tandem, then on one leg. The yielding surface disturbs mainly the information coming from the sole of the foot.
With upper limb movement. Standing on one leg, move your arms diagonally, pass an object from one hand to the other, look upwards. Every movement shifts the centre of mass: it is a perturbation you create yourself, so it is controllable.
Star excursion. On one leg, touch the floor with the other foot as far away as possible forwards, inwards and outwards-backwards, without moving the supporting foot. It is also a test: if one leg reaches much less far than the other, you have found the asymmetry to work on.
Tandem walking. One foot in front of the other along an imaginary line, forwards and then backwards. It introduces the dynamic component.
Advanced level
It makes sense to get here if your goal requires it: contact sports, changes of direction, uneven terrain. It is not "better" by definition.
Wobble boards and hemispheres. Wobble board, two-dimensional balance board, BOSU. They are tools, not goals: use them for the type of oscillation you want to train, not because they look harder.
External perturbations. A partner gently pushes you from random directions while you hold the position. It introduces unpredictability, which is what almost all self-run programmes lack.
Single-leg landings and hops. Small hops forwards, sideways, diagonally, with a controlled and stable stop. This is where the work starts to resemble sport.
Sport-specific movements. Reproduce your movement, a shot, a change of direction, a contact, under demanding balance conditions. Given the specificity we discussed above, this is probably the part that pays off most.
For the ankle
It is the joint on which proprioceptive exercises have the best evidence, and that is no coincidence: lateral sprain is among the most common musculoskeletal injuries. There is also a dedicated physiotherapy clinical practice guideline, published in the Journal of Orthopaedic & Sports Physical Therapy (2021 revision), which addresses first-time lateral sprain and chronic ankle instability together, framing them as deficits of stability and movement coordination.
Alphabet with the foot. Seated, draw letters in the air with the toes. It is mainly mobility and fine control, a good starting point in the early phases.
Single-leg progression. Stable ground, then cushion, then board. Change surface only when the previous one is stable and pain-free.
Wobble board. Standing on the board, try to keep the edge away from the floor for as long as possible, then try to control the circle clockwise and anticlockwise.
Hops and landings. Small single-leg hops in various directions, in the advanced phase: this is the task that most resembles the moment when the ankle usually gives way.
For the knee
The knee is at the centre of a chain: above it is the hip, below it the ankle, and the control you see at the knee is often the result of what happens elsewhere.
Single-leg mini squat. On one leg, lower yourself slightly, making sure the kneecap stays aligned over the second toe, without the knee collapsing inwards. Check yourself in a mirror or film yourself: the internal sensation in this exercise is often misleading.
Controlled step-down. From a low step, lower yourself slowly, lightly touching the floor with the heel of the other foot, then come back up. It is the classic exercise that reveals the problem is not strength but eccentric control.
Lunge with perturbation. In a stable lunge position, a partner gently pushes your pelvis or shoulder from different directions.
Single-leg stance with the free leg moving. While the other leg traces trajectories, the supporting knee must stay aligned and the pelvis level.
For the hip
The hip governs the pelvis, and a pelvis that gives way to the side dumps everything onto the knee.
Single-leg with attention to the pelvis. On one leg, check that the free side of the pelvis does not drop. If it drops, it is a sign that the work is for the glutes more than for proprioception.
Single-leg deadlift. On one leg, bring the torso forwards and the free leg backwards, keeping the pelvis closed and the back long. It combines balance, hip control and the posterior chain.
Clock drill. On one leg, touch the numbers of an imaginary clock on the floor with the other foot, returning to the centre each time.
Programming: what we know and what we don't
On the ideal dose of proprioceptive exercises, the research is less generous than we would like. The review on balance training in young people does find moderate to large effects on static and dynamic balance, but when it tries to explain which dosage produces them, the meta-regression finds, in adolescents, no variable capable of predicting the result: neither the length of the programme, nor the frequency, nor the volume. The authors themselves suggest that the decisive variable is something else, not measured in the studies: probably intensity. In the secondary analyses, which the authors themselves invite readers to interpret with caution because they look at one variable at a time, the largest effects were seen with twelve weeks of programme, two sessions a week and short sessions (4-15 minutes): indicative guidance, not a prescription. On falls in older people, by contrast, the indication is clear and concerns quantity: programmes that challenge balance and exceed three hours a week achieve the largest effects.
What can be said honestly:
- after a sprain, the threshold that appears in the literature is at least six weeks of continuous work: it is not a detail, it is the difference between a programme and a few sessions;
- for prevention, short but regular sessions built into the warm-up are the most realistic format, because it is what people keep doing;
- in older people at risk of falls, quantity matters more than the refinement of the exercise, and it should be built with a professional;
- concentration is part of the exercise: doing it while looking at your phone changes the task, and as we have seen, the results are specific to the task.
Mistakes I see most often
Confusing difficulty with usefulness. Climbing straight onto a BOSU because it looks serious: it is the most common mistake in proprioceptive exercises done by self-taught people. None of the studies cited here showed that an unstable surface is the superior choice: they simply did not ask that question. Treat it for what it is: one of five levers, and not the most important one.
Training balance "in general". In light of the specificity of the results, months spent standing on one leg do not necessarily prepare you for a change of direction on wet grass. If you have a goal, the exercise has to resemble it.
Stopping too early after a sprain. When the pain disappears, the control problem can remain. This is where relapses begin, and it is the reason the literature speaks of weeks, not days.
Expecting it to resolve chronic instability. If the ankle has been giving way for years, the evidence that balance work alone reduces recurrences is insufficient: an assessment is needed, because there may be mechanical, strength or joint components to address.
Not measuring anything. Without an initial test, how long you can stay on one leg with your eyes closed, how far you reach in the star excursion, and what the difference is between the two sides, you will not know whether you are improving or just passing the time.
How telerehabilitation can help
The added value of a physiotherapist on this topic is not showing you the exercise: you can find that anywhere, including in this article. It is working out which task to train in your case, recognising the compensations that make you look good without being so, and deciding when to change lever. It is also the reason proprioceptive exercises done blind deliver less than those chosen after an assessment.
With Ri-Hub you can do it by video call, at 29.90 euros per session: the physiotherapist watches the quality of your footing in real time, has you do the right tests and builds a progression around your real goal, whether that is getting back to running or no longer being afraid on the stairs.
In summary
Proprioceptive exercises work, but not for everything and not as much as common sense promises. For the ankle they reduce sprains by roughly a third, with the clearest benefit in people who have already been injured and with a minimum duration measured in weeks. For falls in older people, the exercise that challenges balance is effective, provided it is done enough. For the knee they count as part of a broader programme, not on their own. And on "improving balance in general" the research calls for caution: you improve mostly at what you train.
It is a less spectacular result than what circulates online, but it is more useful: it tells you where to put your time. Choose the task you need, train that, measure it, and change one variable at a time.
You might also be interested in:
Sources
- The effectiveness of proprioceptive training in preventing ankle sprains in sporting populations: a systematic review and meta-analysis, Journal of Science and Medicine in Sport, 2015
- Proprioceptive Training for the Prevention of Ankle Sprains: An Evidence-Based Review, Journal of Athletic Training, 2017
- Systematic review of postural control and lateral ankle instability, part II: is balance training clinically effective?, Journal of Athletic Training, 2008
- The effectiveness of proprioceptive and neuromuscular training compared to bracing in reducing the recurrence rate of ankle sprains in athletes: A systematic review and meta-analysis, Journal of Back and Musculoskeletal Rehabilitation, 2018
- Specificity of Balance Training in Healthy Individuals: A Systematic Review and Meta-Analysis, Sports Medicine, 2016
- Exercise to prevent falls in older adults: an updated systematic review and meta-analysis, British Journal of Sports Medicine, 2017
- Meta-analysis of meta-analyses of anterior cruciate ligament injury reduction training programs, Journal of Orthopaedic Research, 2018
- Effects and Dose-Response Relationship of Balance Training on Balance Performance in Youth: A Systematic Review and Meta-Analysis, Sports Medicine, 2018
- Ankle Stability and Movement Coordination Impairments: Lateral Ankle Ligament Sprains Revision 2021, Journal of Orthopaedic & Sports Physical Therapy, 2021
Last check of the sources: August 2026.
Stay in touch with us
Join our dedicated Facebook group: we reply to all patients, give practical advice and clear up the most common doubts.
Need a consultation?
Start with the Ri-Hub questionnaire: a few questions about your case, you sign up, and you see who will follow you: a physiotherapist registered with the Albo, whom you message in chat whenever you like. If you would rather talk it through by voice, the first consultation is free, lasts twenty minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to continue.
