Exploiting science for exercise recommendations is funnier than it sounds especially when it’s an opportunity to use the word “ass” — as in “sleepy ass syndrome” which I have mentioned on occasion, and will be the focus of that post.
“Sleepy ass syndrome” is a colloquial denomination for a pathology first described in detail by Dr. Stuart McGill first under the more scientific-sounding moniker of arthrogenic neuromuscular inhibition. However, McGill notes that ‘clinical evidence’ predated his study, and in Low Back Disorders, explicitly credited Czech neurologist and physical therapist Vladimir Janda (1928-2002) for the origin of the idea.
Before saying a few more words about Janda, note that “clinical evidence” refers to anecdotal empirical data (that is, data that is not collected and analyzed in a systematic fashion) collected by physicians and physiotherapists during their practice . It’s the kind of data that makes for great watercooler conversations, but cannot be used to prove anything — although it can motivate hypotheses, which will become important very soon.
Janda pioneered the idea that chronic pain may result from muscle imbalances pulling joints out of alignment. According to Janda, local causes of pain, such as tissue failure to maintain joint alignment, are actually the endpoints of chains of causes that have a systemic origin — namely, a failure of the central nervous system (CNS) to maintain proper control around the joints.
Janda’s approach is often described as “holistic”, and as a result Janda has become a “functional movement” bullshitter darling (see item 37 of this list of 40 things to do to become a functional movement guru). But that’s merely guilt by association, and we should not let it fool us. Case in point, McGill’s approach is also “holistic” and he is also a functional movement bullshitter darling (see the same item in the previously mentioned list).
In any case, whatever holistic is supposed to mean is irrelevant here. What is not irrelevant at all is why Janda’s approach can be easily abused due to its focus on general hypotheses over systematic data collection and analysis. It can be grasped in a pinch with how McGill credits Janda with the idea behind the sleepy ass syndrome:
Dr. Vladimir Janda proposed the crossed-pelvis syndrome in which those with a history of chronic low back troubles displayed characteristic patterns of what he referred to as […] weak gluteal and abdominal wall complex and tight hamstrings and hip flexors. […] Although I have difficulties integrating the terms weak and tight from a scientific point of view, Janda’s insights were generally true.S. Mc Gill, Low Back Disorders 3rd Ed., p. 150
Once again, the culprit is semantics. ‘Weak’ and ‘tight’ are everyday terms that can be used without much care for whatever they are supposed to mean in the context of Janda’s approach, and Janda never tried to reconstruct those terms using systematic data. Now, that’s where Analytic Fitness™ can really shine. It is, after all, more or less the proper application of the philosophy of science to exercise science, broadly construed. And the applied philosophy of science is often just about semantics (as I demonstrated with stability1 and stability2.)
And thus today, I invite you to an exploration of
weak asses inhibited gluteal and abdominal wall complexes and what we can do about it.
A theory of sleepy asses
McGill’s misgivings about Janda’s ‘weakness’ and ‘tightness’ is that they are theoretical constructs still lacking a systematic scientific replacement. Regular readers of this blog already know similar constructs, for instance, “irradiation” and “functional exercise”. McGill’s opinion about the scientificity of Janda’s weakness and tightness is mistaken. It’s an honest mistake, mind you, and stems from a lack of familiarity with Carnap’s philosophy of science, which McGill cannot in good conscience be blamed for. Again, regular readers of this blog already know Carnap (otherwise, you can check this post and that one later).
In a nutshell, unbeknownst to himself, McGill has already proposed an explication of Janda’s ‘weakness’ and ‘tightness’ and, therefore, a scientific interpretation thereof. In fact, his remark that “[Janda’s] insights were generally true” may be taken as summing up just that. In this part, I’ll sketch just enough of McGill’s explication of ‘weakness’ and ‘tightness’ to later capitalize on it for exercise selection.
The Tonic-Phasic Hypothesis
In Training From Scratch (Special): What muscle does that work? I mentioned the hypothesis put forward by Dr. Vladimir Janda (1928-2002) that the CNS is keeping some muscles under higher tension than others. Vladimir Janda proposed differentiating muscles between tonic and phasic muscles with a hypothetical distribution given by the muscle map below with the tonic muscles and the phasic muscles that have an affinity for shortened (contracted) and lengthened (relaxed) states.
Janda’s hypothesis is an empirical hypothesis based on repeated observations. The ‘tightness’ referred to is easily described phenomenologically. For instance, try to do a Cossack lunge after sitting for a while: your “ouch” when the stretch reflex kicks in in the extended leg is the phenomenological counterpart of Janda’s ‘tightness’.
Janda’s hypothesis is not about the observations, but about their underlying reason. For instance, Janda hypothesized that the tonic–phasic distinction is explained by the evolutionary process of hominization and in particular, the development of bipedal locomotion. For instance, the tonic muscles are those you need to walk on all fours. Since every human being still begins their life moving around that way, the ‘older’ tonic muscular system is also the one that our CNS learns to control first, resulting in an affinity for shortened states.
Unfortunately, like every evolutionary explanation, Janda’s hypothesis is not directly testable. It’s a well-known problem with evolutionary hypotheses, so I’ll let it slide. An unfortunate consequence is that a bunch of quacks can routinely abuse Janda’s work by dropping the name of the syndromes he described and exploiting the fascination of the general public for evolutionary hypotheses, two staples of functional movement bullshitters (see Item 10 and 30 of the aforementioned list).
Again, this is merely guilt by association. After all, eugenics claims support from Darwin’s theory of natural selection, and that’s not a reason to ditch it. So let’s forget the quacks, and get back to McGill.
Weakness & Tightness, explicated.
Janda’s approach meshes well with post-Bergmark biomechanics, in particular, the insight that musculoskeletal mechanical stiffness (stability2) is often maintained below the critical value for stable equilibrium and requires corrections from the CNS (see this post for a short discussion of this point). It is then unsurprising that McGill and Janda knew one-another and collaborated multiple times, as shown in the picture below, where Janda is “assessing [McGill’s] gluteal activation while apparently providing mild entertainment to the audience” (McGill, 2016, Low Back Disorders 3rd Ed., p.150).
From the work of Bergmark, McGill, and countless other biomechanists, we can identify the (hypothetically) phasic muscles of the torso (rectus abdominis, obliques, glutes and some of the spinal erectors) with muscles whose stiffness is predominantly active, that is, controlled by the CNS.
Similarly, we could propose that the (hypothetically) tonic muscles of the torso are muscles whose stiffness is predominantly passive in Bergmark’s sense, i.e. governed by the Lidell-Sherrington reflex (the stretch reflex) which does not require control from the CNS.
Based on the above, McGill’s “difficulties integrating the terms weak and tight from a scientific point of view” do not appear that serious. In fact, the integration of weak and tight from a scientific point of view is much closer at hand than McGill realizes and it would be an interesting little exercise in philosophy of science to work it out systematically. But we do not need to work out this integration in detail to derive practical recommendations, and that’s what I will do now.
Waking up your ass
Simple-minded solutions to the problem of sleepy asses are misguided and don’t work but they are on the right track.
Two commonly proposed solutions to cure the sleepy ass syndrome are isolation exercises and motor patterns on unstable platforms. The first solution is inspired by bodybuilding and the so-called ‘mind-muscle connection’ (more on that soon). The second solution is inspired by ‘functional exercise’ and the incorrect notion that compromising balance improves stability. I’ve discussed the problem with the first solution in this post and with the second in this one, so let’s just say that they are misguided.
But I’m feeling unusually charitable today so I’ll concede that both bodybuilders and ‘functional exercise’ advocates are onto something. And quite ironically, in spite of diametrically opposed approaches to fitness and exercise, it’s pretty much the same thing, but they both get it wrong. And yet, if we un-muddle it, we can have something pretty solid.
The usual suspects: muscle & movement
When bodybuilders talk about ‘training muscles’, functional exercise zealots reply with ‘training movement’. Now, I don’t particularly care for being Captain Obvious here, so let’s just say that you can’t do one without the other and leave it at that. What is more interesting is that they are sharing a concern: increasing the training effect without increasing the risk of overuse injury.
Take, for instance, the bodybuilders’ fetish for exercise variety. Insisting that some exercises are redundant misses an important point behind exercise variety: varying joints angles even so slightly may help prevent overuse injuries. In principle. The problem in practice is the discrepancy between the movement mechanics of an isolation exercise (often constrained by a machine) and the function of the joint in more natural scenarios.
The bodybuilder’s solution to this transfer conundrum is the so-called ‘mind-muscle connection’: the notion that isolation exercises improve proprioception of the isolated function and that this improved proprioception can transfer from isolation exercises to more ‘functional’ movements.
While there an internal logic to the argument, there is an obvious rejoinder: if the goal is to improve proprioception in complex movement why not use complex movements in the first place? For instance, exercises such as the Turkish Get-Up are perfectly appropriate to improve proprioception and work the shoulder, hip and ankle joints in a variety of angles (see this recent analysis of the shoulder action in the TGU).
The simple existence of an alternative to a bodybuilding-based approach to proprioception is not in itself an argument against it. However, the risk of ingraining poor motor patterns due to the mechanical constraints of the machines is. Furthermore, a ‘bodybuilding’ approach to rehab may actually be a bad idea due to the possibility of failure at low loads (cf. aside below).
Functional exercises zealots usually endorse variety as a side effect of their fetish for taxonomies and lists of things-to-do-instead-of-what-everybody-else-is-doing (items 13 and 14 of the aforementioned list). Their main injury prevention tool is compromising stability1 in order to limit the load that can be imposed on the body as a whole. Limiting the load is in turn an effective way to ward off tissue failure (provided that stability2 is not compromised), while instability also provided heightened proprioception.
Unfortunately, there’s a double itch. First, unstable platforms impose too drastic reductions of the working load. The reduction is such that the training stimulus from resistance exercises performed on unstable platforms soon becomes insufficient to promote strength and power adaptations (see this meta-analysis). Eventually, this precludes strength-endurance adaptations that prevent injuries, defeating the main purposes of ‘functional’ movement.
Second, an unstable platform dangerously compromises the mechanical stability (stability2) of ankle, knee, and shoulder joints. This may be enough to disrupt otherwise stable equilibria during exercise, but it’s probably not that common — otherwise, the functional gurus would already have split between advocates and opponents of unstable platforms, which has not happened. More likely to happen is a delayed failure. Quite ironically, this implies that unstable platforms are not a good idea for rehab for the very same reason bodybuilding isn’t (cf. aside below).
There are actually science- and evidence-based alternatives to mechanically constrained machine-based isolation exercises and mechanically unstable balance exercises. In recent years loaded variations of activation exercises have become popular for developing proprioception together with strength and stability. One example is the Barbell Hip Thrust, a loaded version of the Glute Bridge or Back Bridge, proposed inter alia by McGill for “grooving gluteal-dominant hip-extension patterns” (McGill, 2016, p. 234).
The problem with loaded activation exercises is much the same as with any bodybuilding exercise: specificity. Now, there is some empirical evidence that Barbell Hip Thrusts transfer better than Barbell Front Squats to 10- and 20-meter sprints while Front Squats transfer better to jumping performance. But the issue of activating sleepy glutes is not the same for athletes and for regular Janes and Joes: athletes would want their glutes to assist with performance while regular Janes and Joes would merely want them to work the way they should.
Still, we can take a hint from the Barbell Hip Thrust study and in particular the hypothesized mechanism whereby they transfer better than Front Squats to running performance. Dubbed by the authors ‘Force Vector Theory’, the hypothesis is, in a nutshell, that Barbell Hip Thrusts share a force vector with the horizontal component of running, while Front Squats don’t (and symmetrically, Front Squats share a vertical force vector with jumping while Hip Thrusts don’t).
The Force Vector Theory is an interesting hypothesis that can be put to work for rehabilitating sleepy asses in everyday activities. One of the most common movement pattern messed up by sleepy asses is standing up, for instance from a seated position. Re-learning how to stand is actually a big part of back rehabilitation protocols because people with sleepy asses often use the lower back to generate momentum to get off their chairs.
A standard rehabilitation protocol would include bodyweight Glute Bridges followed by repetitions of the movement of standing off a chair while trying to repeat the feeling of gluteal activation learned with the Glute Bridges. That’s effective but it has several limitations:
- It’s dead boring. Who would cheer at the prospect of sets of Glute Bridges followed by sets of raising from a chair?
- It’s not effective for adults in the long run. Adults do not seem to adapt to motor patterns through sheer repetition without loading (see this review).
Someone with seriously impaired and/or painful back function also has a tolerance to load at the nadir. Standing from a chair for repetitions would actually provide a strength-endurance overload for that person but if that’s not you, you need to add resistance to improve abdominal and gluteal function in the activities of everyday life.
One solution for people with really poor proprioception is to use Glute Bridges for proprioception and then raise off a chair holding a weight. I’ve used this strategy in the past with people whose proprioception was seriously deficient, and it worked great. It’s more effective than sticking to bodyweight but it’s still dead boring.
However, I’ve learned since then about how the CNS anticipates possible shifts in loads and how it can be exploited for exercise and nowadays I would probably try and implement something more like that (missing the jug at the first rep is entirely optional).
Conclusion (for today): Stabilize like a boss
Vladimir Janda’s notions of ‘weakness’ and ‘tightness’ are a reminder of how robust anecdotal evidence can be.
Vladimir Janda proposed the notion of CNS-mediated muscle ‘weakness’ in the 1960s based on clinical evidence. It took about 40 years for experimental research to catch up and introduce an explication (in Carnap’s sense) of Janda’s ‘weakness’, namely Arthrogenic Muscle Inhibition (AMI).
Some of Janda’s other insights cannot actually be investigated experimentally in the same way. But all in all, Janda’s work is an example of fringe science gone mainstream and an illustration that there is no principled criterion for telling science and pseudo-science apart. It’s also an example of how easily untested empirical hypotheses can be co-opted by quacks.
Interestingly, from the standpoint of the philosophy of science, the methodological difference between McGill and Janda is not that big. The empirical standing of the Bergmark-McGill biomechanical model of the spine and of Janda’s hypothesis of the co-existence of two muscular subsystems are equivalent: both are rooted in observations, supported by empirical evidence, but ultimately untestable.
Still, both lend themselves to practical applications whose success will ultimately be their real empirical test. We can mine them for exercise recommendation and contribute to their empirical test by acting upon those recommendations.
And it does not have to be boring, either. After all, you can’t spell ‘functional’ without ‘fun’.
^ Later stages of hominization may have ‘exapted’ tonic muscles. For instance, the pectoralis major and the biceps help holding bipedal pre-human cubs while standing and walking, unlike chimps and gorilla cubs who can hold on to the back hair of their mothers. ‘Tight’ hamstrings (which contrary to popular belief do not contribute to back pain) became useful in high-power activities such as bipedal sprint and jump, etc.
^ This risk is at the heart of biomechanists’ objection to machine training for stabilizing joints. For instance, McGill writes the following in Low Back Disorders, 3rd Ed.: “Generally, the goal of establishing stabilizing motor patterns requires the person to support body weight and coordinate the stabilization of all joints involved in the task. In other words, it is a whole body, even the whole person, endeavor. Because neither workers nor athletes perform their tasks in this artificially stabilized manner, these types of motor patterns may not be transferable and, worse, may cause inappropriate motor and motion patterns. the early instances in which machines that isolated joints can be helpful is when an injury to a specific body part requires its protection during rehabilitative training.” (p. 318)