With the popularity of the strength training posts, I’ve gotten many questions on both my and Matt’s comments on unstable training. People want to know why we both generally oppose the use of such devices as swiss balls, bosu balls, and dyna discs. Matt has once again been kind enough to let me post his excellent paper on the subject. So what follows is a research based paper by Matt Andre discussing the use of unstable training. It is long, but well worth the read, and is completely cited for those research sticklers.

“The physical education of the human race ought not to be confined to the humble object of preventing disease. Its aim should be loftier and more in accordance with the destiny and character of its subject – to raise man to the summit of his nature; and such will be its scope in future and more enlightened times” (Green, 1986, p. 183). This prediction, stated by S. D. Kehoe in the 1860s (according to Harvey Green), stands true today. Both exercise and sports scientists alike strive to determine the best way to achieve the pinnacle of human performance, not just for health’s sake, but simply to see what the human body is physically capable of. However, these goals have led some Americans down a path that will actually take them further away from the objective of reaching their highest physical potential.

Back in the 1860s, during the period of “Muscular Christianity,” the ideals of Kehoe and others were becoming quite popular (Green, 1986). Kehoe was interested in teaching people that the use of Indian Clubs was the best way to achieve not only optimal health but also a fit and capable body (Green, 1986). Other fitness gurus of that era pushed their exercise cure-alls of gymnastics, calisthenics, stretching, throwing, light dumbbell exercises, skill training, several different forms of isometrics, rubber pulley exercises, special exercise machines which claimed to do everything, and more (Green, 1986). This is interesting because every new training idea or invention was supposedly much better than the one that came before it, and was all that anyone needed to achieve their health and fitness goals.

Today, it is well documented in the scientific literature that progressive resistance training aimed at improving strength, power, and muscular hypertrophy, as well as endurance training or moderate aerobic exercise combined with improved eating habits, can lead to a plethora of health benefits in addition to very significant increases in all aspects of human performance for all age groups, including the elderly and children Behm, 1995; Brooks et al., 2005; Fitzgerald et al., 2004; Fujita et al., 1995; Ratamess et al, 2009, Rogers et al., 2008; Smith et al., 2009). However, it is noted that certain characteristics of exercise are important for achieving these benefits, in that the exercise must be progressive, specific, and must have some variation (Ratamess et al, 2009). Understanding these basic rules are important, because it helps one determine whether or not to implement a specific training program or tool.

Unfortunately, the idea of having a one-size-fits-all magical cure-all has not gone away, and people still search for the newest, greatest training method or device that will help them achieve all of their health and fitness goals with less time and energy spent. The idea of a modern training concept that will trump all those which came before it leads us to the focus of this paper. Numerous fitness enthusiasts and fitness professionals are advocating the use of unstable surface training (UST) as a means of achieving all physical and health goals. Unstable surface training may include the use of stability balls (AKA Swiss balls), BOSU balls, wobble boards, and several other devices that create an unstable platform with which to work from. Additionally, UST is being combined with resistance training exercises such that one will perform a traditional exercise with weights wile standing, sitting, or lying on an unstable surface. Upon entering any corporate gym n America or performing a simple Google search, one will quickly discover the high revalence of UST and its combination with resistance training.

One idea behind using UST is to increase trunk muscle activation during exercises that attempt to isolate the trunk musculature. However, a very recent review written by a Professor and practitioner of osteopathy makes a compelling argument that attempting to isolate the “core” musculature does not appear to offer any proven benefits, despite claims that this type of training will reduce lower back pain, improve rehabilitation, improve movement patterns and motor function, and prevent injuries (Lederman, 2010). In fact, it is suggested that core stability exercises are no more effective at strengthening the trunk and preventing injuries than any other type of exercise or therapy (Lederman, 2010). While claims exist about the preventative benefits of core stability training, hard scientific evidence is lacking (Lederman, 2010).

If we assume that Lederman’s (2010) argument is true, then it is unnecessary to address the use of UST for trunk isolation movements in this paper. Regardless, there appears to be a larger problem at hand. Proponents of performing resistance training exercises on an unstable surface claim that you can work your core, improve balance and coordination, rehabilitate injuries, “prehabilitate” (prevent injuries), build muscle, burn fat, and get stronger all at the same time. All of these claims for this fitness cure-all can be found on bosu.com and other websites that sell these types of products. Additionally, these claims have been found in numerous articles (not peer-reviewed) found on “ptonthenet,” a website for personal trainers that claims to be evidence-based yet never shows a single scientific reference. Additionally, “expert” writers on the website make claims about training “truths” that have been consistently rejected in the scientific literature.

Another interesting aspect to this story is that UST has already been thoroughly examined in peer-reviewed literature, yet advocates of UST never appear to discuss these studies. It appears that many advocates of UST may be unaware of the facts that (A) “ptonthenet” does not provide peer-reviewed scientific literature and (B) many of the claims that practitioners make about UST have actually been tested by many different scientists in many different laboratories. Therefore, it is the purpose of this paper to discuss the existing peer-reviewed literature as it pertains to UST and find out if its current popularity amongst practitioners is justified. Deciphering this evidence will help to determine if UST devices are the newest cure-all, or if the American people should be focusing on other proven methods for health and performance, such as traditional progressive resistance training, as is outlined in the recently updated ACSM Position Stand on Resistance Training (Ratamess et al., 2009).

Justification of UST Research
In 1995, Behm authored a paper entitled “Neuromuscular implications and applications of resistance training,” in which he discussed the adaptations that occurred within the body when someone began (and continued) resistance training. Some of these adaptations, which are necessary for gains in muscular strength, power, and hypertrophy, include the need to decrease antagonistic coactivation, increase agonist muscle activation as monitored by EMG, exceed a minimal threshold of intensity (% 1RM), improve synchronization, improve rate of force development, and increase inter- and intramuscular coordination to ensure proper movement patterns (Behm, 1995). These concepts are still held to be valid today (Ratamess et al., 2009). If a participant performs elbow flexion (i.e. a biceps curl), the agonist muscle isbiceps brachii and the antagonistic muscle is triceps brachii. An early adaptation to resistance training involves teaching the body to increase activation of the agonist (in this example, biceps brachii) and decrease activation of the antagonist (triceps brachii), which will lead to less torque at the elbow joint and greater performance (Behm, 1995). This “greater performance” due to a decrease in antagonistic coactivation will then lead to improvements in muscular strength, power, and hypertrophy (Behm, 1995). Suggestions for a minimal threshold of intensity differ depending on the goal, but, for strength especially, it is recommended that intensity exceed 80% of one repetition maximum (1RM; the most weight that can be lifted for one repetition) after a few weeks of training (Behm, 1995; Ratamess et al., 2009; Rhea et al, 2003). Training at too low of an intensity does not cause the need for adaptation at the genetic level, which means that improvements in muscular strength and hypertrophy will not be achieved (Behm, 1995). Additionally, to achieve increases in muscular power, rate of force development must be improved (Behm, 1995). All of these factors from Behm’s paper (1995) are somewhat interdependent, and one factor can certainly impact another. For example, if a stimulus were to lead to improper recruitment patterns and ruin coordination, then this could impact agonist muscle activation, rate of force development, and basic performance variables.

It is interesting that Behm (1995) noted the importance of these neuromuscular factors, because these same factors would come into play during his later research involving UST. In 2002, Behm and others conducted a study that addressed the effects of surface stability on muscle force and muscle activation. The authors (Behm et al., 2002) noted that even though UST was being utilized since before WWII by physical therapists, especially in the form of Physio balls (i.e. Swiss balls, stability balls), there was no scientific research to validate the claims that these therapists were making, including the claim that “increased stress on the neuromuscular system caused by UST will lead to increases in neuromuscular coordination, thereby leading to strength.” Behm et al. (2002) investigated isometric force output, muscle activation and EMG activity of the knee extensors, plantar flexors, and their antagonists under both stable and unstable conditions in eight, male, recreationally-trained subjects. Peak force was reduced by 70.5% in knee flexors and 20.2% in plantar flexors when knee extension and plantar flexion were performed on unstable surfaces (Behm et al., 2002). Muscle activation was reduced by 44.3% and 2.9% in unstable conditions for the knee extensors and plantar flexors, respectively (Behm et al., 2002). The reason that knee extensors saw a greater decrease for both variables was because the condition was more unstable than the plantar flexion condition, proving that the more unstable the surface, the greater the reduction in peak force output and agonist muscle activation (Behm et al., 2002). Additionally, antagonistic muscle activation was significantly increased for both movements (Behm et al., 2002).

Interestingly, the authors (Behm et al., 2002) stated in their introduction that neural adaptations (ex. decreased antagonistic coactivation, increased agonist EMG) played the most important role in strength gains when one first begins training; they also discussed this heavily in their 1995 paper. If this is so, and the study (Behm et al., 2002) shows an increase in antagonistic coactivation (negative adaptation) and a decrease in agonist EMG (negative adaptation), then it appears that the authors have demonstrated that UST should be contraindicated. Despite proving the maladaptations caused by UST, the authors then say, “Swiss balls may permit a strength training adaptation of the limbs, if instability is moderate, allowing the production of overload forces” (Behm et al., 2002). This statement clearly makes no sense, as the maladaptations would certainly outweigh any potentially positive effects of UST. Interestingly, Behm appears to be one of the largest proponents of UST from the research world, yet he successfully demonstrated that UST is ineffective, at best.

Currently, many corporate gyms, such as many corporate branches of Gold’s Gym, require their personal trainers to be certified through the National Academy of Sports Medicine (NASM). The authors of the NASM textbook (Clark et al., 2004) claim that all persons should begin their resistance training career with stabilization training (comprised mainly of UST) to improve neuromuscular efficiency. This is very important, because it means that a large proportion of personal trainers (often referred to as “fitness professionals”) have been taught to use UST as a mandatory part of their training protocols.

The authors of the text claim that if a trainer has their clients utilize UST, then they will improve “proprioception,” learn better movement patterns, get stronger, decrease stress on the joints, and not become injured as a result of strength training (Clark et al., 2004). They (Clark et al., 2004) also claim that UST is a “superior” way to alter body composition, because they theorize that “additional” muscles utilized during UST will cause the body to burn more calories than their traditional resistance training counterparts (it is interesting that these “theories” are stated as “fact” in the text). These “additional” muscles that the authors refer to are antagonists, and this increased coactivation leads to ruined neuromuscular efficiency and improper movement patterns, the opposite of what the authors (Clark et al., 2004) claim. This increased antagonistic coactivation will increase stress on the joints. Additionally, it has been noted in the literature (Ivy & Portman, 2004) that intensity (% 1RM) determines how many calories are burned during an exercise, and if UST causes one to be weaker, it could be assumed that UST will not burn as many calories as the stable version of the same exercise at a higher intensity. With all of these negative adaptations, it appears that beginning with UST will lead to an increased risk of injury with a decreased opportunity to achieve common fitness goals (positive change in body composition, increased muscular strength, increased muscular hypertrophy, etc.). An inability to achieve fitness goals despite hiring a personal trainer may lead some clients to develop feelings of helplessness, and, as a result, quit exercising.

This information is quite frightening, considering that (A) this is the textbook used to certify many employed personal trainers, (B) people pay a lot of money for this NASM certification (which appears to be based on false theories), and (C) people who hire a NASM trainer assume that they are getting a well-educated, high quality fitness professional. This “fitness professional” could potentially be using contraindicated training methods simply by doing what they were instructed to do. It is suggested that gym owners place more scrutiny on the certification process that they have their employees go through.

[NOTE: As a personal trainer, I was certified through NASM (my boss required it). I had all of my clients do what NASM had recommended. When I began researching UST, my goal was to prove that it was useful. However, after seeing all of the evidence against it, I apologized to my clients and changed my ways.]

The Remaining Body of Evidence

It is clear that UST has been utilized for a long time, and is still utilized by fitness professionals today, despite initial research indicating potential maladaptations. With the popularity of the NASM certification and the widespread acceptance of UST as a viable training method, it would appear that UST deserves attention in the scientific literature. Fortunately, after Behm and his colleagues performed their study in 2002, a host of scientists began to conduct more research to determine the effects of UST (if any) on several different performance variables.

Since Behm and others conducted their 2002 study, researchers (Anderson & Behm, 2004; Cressey et al., 2007; Drake et al., 2006; Hamlyn et al., 2007; Kavcic et al 2004; McBride, 2006; McBride et al., 2006; McBride et al., 2009; Nuzzo et al., 2008) have found similar results from the combination of UST and traditional resistance training exercise. Like their 2002 paper, Anderson & Behm (2004) again looked at isometric force output. This time, the comparison was done with an isometric bench press on a stable surface and the same press performed on a Swiss ball (Anderson & Behm, 2004). The authors (Anderson & Behm, 2004) observed a 59.6% reduction in force, eliminating the possibility for adaptation.

In a comparison of rehabilitation exercises typically utilized to activate the trunk musculature, researchers (Drake et al., 2006) found that the trunk is significantly more activated and recruited during stable floor exercises than during the same exercises performed on a stability ball. Another rehabilitation study (Kavcic et al., 2004) looked at spinal stabilization and activation of spinal stabilizer muscles, and found that on a list of 8 exercises aimed at activating the muscles that stabilize the spine, the only one using an unstable surface (a Swiss ball) was the least effective. The other 7 floor exercises were superior (Kavcic et al., 2004).

Decreases in isometric force output and muscle activity of agonists, accompanied by increases in antagonistic activation, have been found with isometric squats on unstable surfaces compared to stable isometric squats (McBride et al., 2006). McBride and others (2009) found that dynamic squats under several different loads not only showed a decrease in agonist activation and increase in antagonist activation on unstable surfaces, but also that unstable squatting led to a decrease in erector spinae (spinal stabilizer muscle, part of the “core”) activation. Additionally, as the intensity increases, so do the differences between stable and UST muscle activation (McBride et al., 2009). This is interesting, as UST proponents would have people believe that the core musculature, including the erector spinae muscles, experience greater activation levels on unstable surfaces.

The idea that the trunk musculature is better activated on unstable surfaces appears to have some evidence (Lehman, 2007). However, these studies use untrained individuals, very low intensity (too low to elicit any worthwhile adaptations, such as improvements in muscular strength or hypertrophy), and often compare UST versions to inferior exercises with shorter ranges of motion, such as floor crunches (Lehman, 2007). It could be assumed that the vast majority of exercisers participating in a well-designed program would quickly advance beyond an un-weighted floor crunch. If the studies which show greater trunk muscle activation with UST use a load that is too low to elicit any positive adaptations, then the use of these training modalities should be questioned, especially considering the maladaptations associated with UST.

In response to these studies, research was conducted to compare stable surface training with heavy deadlifts, squats, and power cleans, to UST, and a much higher degree of trunk muscle activation with the traditional, stable versions of these foundational exercises, in addition to higher peak force output, peak power, and velocity, was observed (Hamlyn et al., 2007; McBride, 2006; McBride et al., 2009; Nuzzo et al., 2008). This decrease in trunk muscle activation caused by UST starts at a smaller but significant level around 50% 1RM (too low for adaptation), and increases dramatically as one approaches 100% 1RM (McBride, 2006; McBride et al., 2006; McBride et al., 2009; Nuzzo et al., 2008). This means that with submaximal weight, there is some decrease in performance, and with weight that is heavy enough to elicit adaptation, there is a very large and substantial decrease in performance.

One study (Nuzzo et al., 2008) showed that a maximal effort isometric contraction of the trunk musculature on a stability ball had less trunk muscle activation than a stable squat performed at 50% 1RM. Similar results have been observed by other researchers (Hamlyn et al., 2007; McBride, 2006). Hamlyn and colleagues (2007) found that a squat with no external resistance whatsoever had the same trunk muscle activation as maximal effort UST exercises attempting to isolate the trunk musculature. These studies show that a basic strength training program aimed at improving the entire body already utilizes the core musculature more than the same exercises under unstable conditions, as well as UST exercises aimed at only working the trunk musculature.

In response to these studies, one group of researchers (Koshida et al., 2008) conducted a study comparing a stable bench press to a bench press on a stability ball (UST). The researchers only used 50% 1RM and observed an approximate decrease of 6% in force and 10% in power and velocity (Koshida et al., 2008). Although these results are to be expected, the authors’ conclusion was not: “Such small reduction rates of muscular outputs may not compromise the training effect” (Koshida et al., 2008). Interestingly, the authors (Koshida et al., 2008) cite evidence in the introduction to their paper that a minimum of 80% 1RM is needed to elicit adaptations in muscular strength, so it is quite odd that they chose to remain well below this minimal intensity threshold. If other studies (Hamlyn et al., 2007; McBride, 2006; McBride et al., 2009; Nuzzo et al., 2008) show that these performance variables worsen as the intensity increases, then one can infer that these so-called ”minor” maladaptations in Koshida’s (2008) study would become even more severe as the intensity increased to the level necessary for adaptation. Performance variables have been measured in response to long-term UST training (Cressey et al., 2007). NCAA Division I soccer players performed their normal strength and conditioning programs, except that one group performed the final exercise of each training session on an unstable surface (Cressey et al., 2007). After ten weeks of training, the UST group saw performance decrements in bounce drop jump, countermovement jump, 10- and 40-yard sprint times compared to the group which did the same exact workout except for the last exercise (Cressey et al., 2007). They also found no change in predicted power output in the UST group despite a significant improvement in the other group (Cressey et al., 2007).

Other performance variables that UST proponents are excited about include balance, coordination, and proprioception. While it may seem like UST will improve these things, there is no hard evidence of this. In fact, one study (Schilling et al., 2009) found that balance was not improved by UST. Schilling and others (2009) found that several weeks of UST with the hopes of improving balance showed no improvements in performance on balance tests, and, in some cases, became worse than the sedentary control groups at performing these tests. However, subjects did report that they thought their balance had improved (Schilling et al., 2009), which leads to the primary basis behind the justification of UST: people think it works.

The results of Schilling et al.’s study are not surprising, since it has already been demonstrated in the literature that proprioception, coordination, core stabilization, and balance are all skill specific (Drowatzky & Zuccato, 1966; Stanton et al., 2004; Willardson, 2004). Drowatzky and Zuccato (1966) found that participants could perform well at one balance, coordination, core stabilization, or proprioception task, yet do poorly at another. Additionally, dynamic and static balances are not related (Drowatzky & Zuccato, 1966). Ades et al. (2003) had 42 women between the ages of 65 and 88, who all had disabling cardiovascular heart disease, perform basic resistance training exercises on a stable surface, with free weights and machines, targeting the legs and arms, at 80% 1RM for 6 months, and found significant improvements in strength, physical function, flexibility, balance, coordination, proprioception, endurance, power, and activities of daily living. The control group, which participated in stretching, calisthenics, and yoga, saw no improvements, not even in flexibility (Ades et al., 2003).

Ades et al. (2003) demonstrated that stable resistance training, with no balance training, “core” training, or UST, was not only able to make elderly and disabled women stronger and more powerful, but also improved balance, proprioception, and coordination to the point where they were able to perform activities of daily living which are often hindered by aging. Also, in another study (Judge et al., 1994), UST did not improve performance of activities of daily living in the elderly, but basic strength training on leg machines did. It appears that those who argue that UST is necessary for balance, proprioception, coordination, and basic function are incorrect; however, traditiona resistance training alone will improve those areas, even in the elderly (Ades et al., 2003; Judge et al., 1994).

It is known that performing a task makes you better at the skills related to that task, including balance, coordination, core stabilization, or proprioception (Drowatzky &; Zuccato, 1966; Stanton et al., 2004; Willardson, 2004). UST can lead to improper recruitment patterns and ruin motor skills, so it certainly should not be combined with anyone trying to improve a sports skill on a stable surface (Willardson, 2004). Additionally, it wouldn’t be beneficial for an elderly person who needs to navigate objects and avoid falling; potentially, UST could ruin stable-ground performance and lead to more catastrophic falls among the elderly.

If someone intends on performing on unstable surfaces (ex. surfing), then it would seem necessary to train on an unstable surface. However, performing the actual activity (ex. surfing) should be enough proprioceptive training, whereas the weight room can be used to improve strength and power. Theoretically, UST wouldn’t “ruin” skill performance for a surfer, but this has not been demonstrated in the literature. For those who perform or function primarily on a stable surface, there should not be any relevant benefits of UST, as learning to balance on an unstable surface may not be necessary for the majority of people. UST will potentially make people really good at UST, while ruining stable surface performance.

Other studies finding UST to be ineffective addressed endurance performance measures (Stanton et al., 2004), muscle activation during a muscular endurance task (Wahl & Behm, 2008) and replacing a desk chair with a stability ball (Gregory et al., 2006). Stanton et al. (2004) found that runners who added Swiss ball training to their workout routine were unable to improve running performance measures, including posture, compared to the group who did nothing but run. They did improve their ability to perform UST, but this did not help running performance at all (Stanton et al., 2004). Overall, this is interesting, because it may have been thought that any new and/or additional stimulus should help athletic performance, but in the case of UST and running, it did not.

Wahl and Behm (2008) found that soleus muscle activation was decreased by 51% during an isometric wall-sit exercise for local muscular endurance in the group utilizing UST during the exercise. It is not clear why the authors (Wahl & Behm, 2008) tested local muscular endurance on an unstable surface. As with the other studies, this example begs the question, “What is the point of the exercise?” Also, “What adaptation are you trying to elicit?”

Some might think that UST can improve posture. The idea of replacing an office chair with a stability ball to improve trunk muscle activation and posture while sitting is an interesting topic. One proponent of UST (not a physiologist or a biomechanics expert) claims that using a stability ball for a chair at the office gives you a workout, increases performance, and improves posture (Bumgardner, 2007). Unfortunately for Bumgardner, this theory was previously tested and disproved in the scientific literature (Gregory et al., 2006) and has yet to be proven in an actual peer-reviewed study. This is another example of a UST claim that not only has no proof to justify it, but has in fact been disproven.


UST is a good training method if you want to improve performance on an unstable surface at the cost of a decrease in force output, power output, and agonist muscle activation, a decrease in sports performance, skill performance, and coordination on a stable surface, and an increase in antagonistic coactivation (thereby increasing pressure on the joint). In other words, UST has not been proven to be effective at anything except for UST performance.

UST has been shown in the literature to be fully contraindicated for human performance, especially if used in conjunction with dynamic movements. Society will have to accept that achieving physical fitness requires hard work with the basic tenets of exercise and nutrition. Those who need to work on improving movement patterns could perform exercises such as those outlined in the study by Kibler et al. (2006). Those who would like to improve core strength and spinal stability should perform traditional, functional resistance training exercises, such as Olympic lifts (ex. power cleans), squats and deadlifts.

Additionally, health and fitness professionals need to take more responsibility for their actions by choosing to have an evidence-based practice, where their exercise recommendations are fully based on proven scientific concepts. It seems that many practitioners are fighting to find reasons to justify UST, when instead they should be concentrating their efforts on finding the best possible way to help their clients achieve their goals. If there is a mountain of research showing the maladaptations related to UST, then why do people continue to try and prove that UST may have some slightly positive outcome (that most likely will not be able to outweigh the numerous contraindications)? As was outlined in Green’s book (1986), many fitness trends have come and gone throughout American history, all promising improved health and physical function like never seen before; as with most, unstable surface training is a trend that is better left forgotten.

“Those who wish to make their core stronger should do what we have always done and squat.” Jeffrey M. McBride, Presentation: 2009 NSCA National Conference

Ades, P. A., Savage, P. D., Cress, M. E., Brochu, M., Lee, N. M., & Poehlman, E. T.
(2003). Resistance training on physical performance in disabled older female
cardiac patients. Medicine and Science in Sports and Exercise, 35(8), 1265–1270.
Anderson, K. G., & Behm, D. G. (2004). Maintenance of EMG activity and loss of force
output with instability. Journal of Strength and Conditioning Research, 18(3),
Behm, D. G. (1995). Neuromuscular implications and applications of resistance training.
Journal of Strength and Conditioning Research, 9(4), 264–274.
Behm, D. G., Anderson, K., & Curnew, R. S. (2002). Muscle force and activation under
stable and unstable conditions. Journal of Strength and Conditioning Research,
16(3), 416–422.
Brooks, G. A., Fahey, T. D., & Baldwin, K. M. (2005). Human bioenergetics and its
applications. Fourth edition. McGraw-Hill: New York, NY.
Bumgardner, W. (2007). My chair – an exercise ball.
Clark, M. A., Lucett, S., Corn, R., Cappuccio, R., Humphrey, R., Kraus, S. J., Titchenal,
A., & Robbins, P. (2004). Optimal Performance Training for the Health and
Fitness Professional. National Academy of Sports Medicine: Calabasas, CA.
Cressey, E. M., West, C. A., Tiberio, D. P., Kraemer, W. J., & Maresh, C. M. (2007). The
effects of ten weeks of lower-body unstable surface training on markers of athletic
performance. Journal of Strength and Conditioning Research, 21(2), 561–567.
Drake, J. D. M., Fischer, S. L., Brown, S. H. M., & Callaghan, J. P. (2006). Do exercise
balls provide a training advantage for trunk extensor exercises? A biomechanical
evaluation. Journal of Manipulative and Physiological Therapeutics, 29(5), 354–
Drowatzky, J. N., & Zuccato, F. C. (1966). Interrelationships between selected measures
of static and dynamic balance. Research Quarterly, 38(3), 509–510.
Fitzgerald, S. J., Barlow, C. E., Kampert, J. B., Morrow, J. R., Jackson, A. W., & Blair, S.
N. (2004). Muscular fitness and all-cause mortality: prospective observations.
Journal of Physical Activity and Health, 1, 7–18.
Fujita, Y., Nakamura, Y., Hiraoka, J., Kobayashi, K., Sakata, K., Nagai, M., & Yanagawa,
H. (1995). Physical-strength tests and mortality among visitors to healthpromotion
centers in Japan. Journal of Clinical Epidemiology, 48(11), 1349–1359.
Green, H. (1986). Fit for America: Health, Fitness, Sport and American Society.
Pantheon Books: New York, NY.
Gregory, D. E., Dunk, N. M., & Callaghan, J. P. (2006). Stability ball versus office chair:
Comparison of muscle activation and lumbar spine posture during prolonged
sitting. Human Factors, 48(1), 142–153.
Hamlyn, N., Behm, D. G., & Young, W. B. (2007). Trunk muscle activation during
dynamic weight-training exercises and isometric instability activities. Journal of
Strength and Conditioning Research, 21(4), 1108–1112.
Ivy, J. L., & Portman, R. (2004). Nutrient Timing. Basic Health Publications: North
Bergen, NJ.
Judge, J. O., Whipple, R. H., & Wolfson, L. I. (1994). Effects of resistive and balance
exercises on isokinetic strength in older persons. Journal of the American
Geriatrics Society, 42(9), 937–946.
Kavcic, N., Grenier, S., & McGill, S. (2004). Quantifiying tissue loads and spine stability
while performing commonly prescribed low back stabilization exercises. Spine,
29(20), 2319–2329.
Koshida, S., Urabe, Y., Miyashita, K., Iwai, K., & Kagimori, A. (2008). Muscular outputs
during dynamic bench press under stable versus unstable conditions. Journal of
Strength and Conditioning Research, 22(5), 1584–1588.
Lederman, E. (2010). The myth of core stability. Journal of Bodywork and Movement
Therapies, 14, 84–98.
Lehman, G. L. (2007). An unstable support surface is not a sufficient condition for
increases in muscle activity in rehabilitation exercise. Journal of Canadian
Chiropractic Association, 51(3), 139–142.
McBride, J. M. (2006). Performance enhancements: New techniques and technology.
National Strength and Conditioning Association National Meeting (presentation).
McBride, J. M., Cormie, P., & Deane, R. (2006). Isometric squat force output and muscle
activity in stable and unstable conditions. Journal of Strength and Conditioning
Research, 20(4), 915–918.
McBride, J. M., Larkin, T. R., Dayne, A. M., Haines, T. L., & Kirby, T. J. (2009). Effect
of absolute and relative loading on muscle activity during stable and unstable
squatting. In NSCA 32nd National Conference and Exhibition, Las Vegas, July 8-
McGuigan, M. R., Newton, R. U., Kraemer, W. J. (2006). Resistance training for better
health in older adults. International Journal of Sport and Health Science, 4, 19–
Nuzzo, J. L., McCaulley, G. O., Cormie, P., Caville, M. J., & McBride, J. M. (2008).
Trunk muscle activity during stability ball and free-weight exercises. Journal of
Strength and Conditioning Research, 22(1), 95–102.
Ratamess, N. A., Alvar, B. A., Evetoch, T. K., Housh, T. J., Kibler, W. B., Kraemer, W.
J., & Triplett, N. T. (2009). Progression models in resistance training for healthy
adults. Medicine and Science in Sports and Exercise, 41(3), 687–708.
Rhea, M. R., Alvar, B. A., Burkett, L. N., & Ball, S. D. (2003). A meta-analysis to
determine the dose-response for strength development. Medicine and Science in
Sports and Exercise, 35(3), 456–464.
Rogers, C. J., Colbert, L. H., Greiner, J. W., Perkins, S. N., & Hursting, S. D. (2008).
Physical activity and cancer prevention. Sports Medicine, 38(4), 271–296.
Schilling, B. K., Falvo, M. J., Karlage, R. E., Weiss, L. W., Lohnes, C. A., & Chiu, L. Z. F. (2009). Effects of unstable surface training on measures of balance in older adults. Journal of Strength and Conditioning Research, 23(4), 1211–1216.
Smith, S. R. (2009). Beyond the bout – New perspectives on exercise and fat oxidation.
Exercise and Sport Sciences Reviews, 37(2), 58–59.
Stanton, R., Reaburn, P. R., & Humphries, B. (2004). The effect of short-term Swiss ball training on core stability and running economy. Journal of Strength and Conditioning Research, 18(3), 522–528.
Wahl, M. J., & Behm, D. G. (2008). Not all instability training devices enhance muscle activation in highly resistance-trained individuals. Journal of Strength and Conditioning Research, 22(4), 1360–1370.
Willardson, J. M. (2004). The effectiveness of resistance exercises performed on unstable equipment. Strength and Conditioning Journal, 26(5), 70–74.

Get My New Guide on: The Science of Creating Workouts


    1. Pete on June 16, 2010 at 12:00 am

      Interesting stuff – it's amazing how easy it is to believe what you hear from the so called "experts" when you don't know the relevant scientific literature. Unfortunately, science is all too often ignored when people are trying to sell you something, and marketing can be far more persuasive than a scientific journal article.


    2. Sam on July 1, 2010 at 6:43 pm

      I don't practice training on unstable surfaces, but wouldn't co-contraction of the synergists involved in running be a good thing? (Lower extremity abductors and adductors?)Have you addressed stability training in a frontal plane only? Also, inability of the quadriceps and hamstrings to co-contract upon ground contact is beginning to be thought of as a major cause for ACL injuries in soccer and other "cut and change direction" athletes. Would instability training be beneficial for these athletes? Cross country courses also offer unstable surfaces to run on…Of course, I can also see from your resources that doing a one legged squat on solid ground would probably activate synergists more than UST…and running on grass should be the perferred regimen vs. UST if concerned about the instability factor. Practice where you compete? I love reading this stuff.

      • Greg on July 23, 2012 at 11:50 am

        "Also, inability of the quadriceps and hamstrings to co-contract upon ground contact is beginning to be thought of as a major cause for ACL injuries"

        FYI, squats, deadlifts and power cleans, done properly, all require co-contraction of the quads and hamstrings, so this concern is addressed by a traditional program of barbell training.

    3. Anonymous on July 25, 2010 at 11:27 pm

      Are we training the brain or muscles? Sometimes we want to train the nervous system. The force output argument changes when the athlete is stable and skilled at UST. Using non-UST developed athletes in research would likely spoil the data. No serious athlete would scrap heavy weights to do UST. Variety is important.

    4. chris on October 18, 2010 at 10:49 pm

      It doesn't look like this paper ever got published?

    5. Greg on January 4, 2011 at 1:04 am


      I've read your blog previously (love it) but never noticed this post. I am happy to see that you have cited some of my research papers (Lehman GJ) David Behm (Behm DG) has recently published a few reviews and recommendations that you would find interesting. David's work is top notch in this area. He consistently beat me to publishing something and did it better).

      I you would like to read my blog or link to it please see it at thebodymechanic.ca

      Thanks for your work, I really appreciate the topics you cover.

      Best of luck with your research.


    6. slm on February 27, 2011 at 2:31 pm

      I definitely understand the importance of debunking some of the myths of using stability balls, especially since most of them are driven by commerce more than fact.

      Having said that I have used stability balls in my personal training and also with clients for 12 years. I would say from my experience the most useful part of the stability ball is how it can help facilitate movements that could not be performed as easily with other props or equipment. Prone Pike lifts, for example, are easy to set up for myself and clients and is one of my favorite exercises. Squats against the wall using the ball to lean against, allow people new to exercise, a way to feel how extended the spine needs to be. There are many more examples of exercises, but mostly I wanted to state that it may not be the best for all the claims stated in your paper, but there are other useful ways to use the ball during exercise

    7. Rob Smith on July 8, 2011 at 3:28 pm

      interesting post. i just read your article on the myths of running, almost all of which i had come across in some shape or form and happened to agree with, but i had been under the impression that UST was beneficial. if i have understood you correctly, you are saying that doing squats, for example, on an unstable surface while recruiting more muscles does not get close enough to RM for it to cause an adaptation. however – and i think you address this to some extent with your comments on surfing – surely there is some benefit to activating the stabilizing muscles in the lower leg and foot for running? you could say that trail running is a more effective way of doing this, but it is harder to control and you run the risk of injury if you are starting out with weak or atrophied muscles. i just use these balance boards for that – to improve my balance.

    8. Greg on July 23, 2012 at 11:46 am

      I'm always amazed when scientists make statements in their conclusions (or abstracts!) that are not supported by their own data. The whole point of the scientific process should be to distance ourselves from these biases. If all else fails the peer review process should weed this stuff out.

      If a 10% decrease in power and velocity "may not compromise the training effect" (per Koshida et al. 2008), then I have a great idea for a new supplement. It's called "reverse creatine." It decreases power and velocity by 10%. Although it has no known benefits, I can comfortably report based on current research that it "may" not compromise your training. Would you take it?

    9. […] You can’t fire a cannon from a canoe! […]

    Leave a Reply