Is it all just too bloody complicated?

I’ve just got back from the ESMAC-SIAMOC meeting in Rome. We’ve been entertained royally for three days in the aulas and cloisters of the Thomas Aquinus University. It has once more been a fantastic opportunity to network and exchange ideas and on one level I come back rejuvenated and inspired.

I say “on one level” because on another level I’ve also come back somewhat disappointed – disappointed because there was little in the scientific programme which left me feeling I understood things better than I did before I arrived. A large number of papers could be summed up by the conclusion, “we understand this area less after performing this research than we did before we started”.  I don’t think it is just ESMAC-SIAMC that suffers in this way. I see it at most of the conferences I attend and in a lot of papers that I read (and, if I’m being honest, in some of the papers that I write).

Just two areas illustrate this. One is in the advanced and complex modelling that is so often the focus of contemporary biomechanics. We learnt (or had confirmed) in Rome that the results are highly dependent on the details of how the individual anatomy is parameterised and of the calibration processes used to define joint centres.  The overall conclusion is that we are less confident in the output of our simulations and models after we’ve performed this research than we were before. Of course it is important to know what the limitations of our research. At some stage, however, we will have to acknowledge those limitations and accept the conclusion that the biological complexity of the human neuromusculoskeletal system is just too great for us to stand any chance of applying these techniques usefully (at least not beyond the constraints of healthy people exercising tightly controlled tasks).

The other field is that of measuring spasticity. Seven or eight years ago I was really excited about the prospect of instrumenting clinical tests to quantify spasticity more rigorously. The results I’ve seen reported are really quite disappointing in that it seems that spasticity is a rather complex and badly behaved phenomenon that simply refuses to be measured.  I have little faith any longer that spasticity is a purely velocity dependent response  (Lance, 1980) and the additional complexity that is introduced when displacement, acceleration or even jerk might have to be considered removes any hope that we will ever understand how these components interrelate within the current paradigm.

One of the “advantages” of research leaving us less clear of what is happening than we were before is that it opens up the conclusion that “further research is required to better understand these phenomena”. Research thus begets research and the university departments rub their hands in glee at the prospect of more research grants, papers and citations. For many of us it leads to increased job security. We have a vicious circle that delights and thrives in creating complexity and chaos.

This is particularly bizarre in orthopaedic and rehabilitation fields (and perhaps more widely across health sciences) in that the tools we have to treat our patients are generally extremely blunt. Selective dorsal rhizotomy, intrathecal baclofen and botulinum toxin are the only tools we have to manage spasticity. At a clinical level the only decision we need to make is which, if any, of these to use. If we want our research to be clinically useful we need to concentrate on the simple questions that need to be answered before we turn our focus to the more complicated ones that don’t.

The small number of presentations that did impress me posed a research question in such a way that the answer actually improved my understanding of a given issue. Almost all of these resulted in me having a clearer, simpler view of the world after the presentation. This doesn’t necessarily require simplistic techniques. The walk-DMC scale that Kat Steele and Mike Schwartz proposed in their prize winning paper (page 25 of Abstract Book) uses a sophisticated technique. It is a technique, however, that has been appropriately selected to answer a well posed research question (Can we quantify the effect of disordered motor control on walking in children with cerebral palsy?). Once the appropriate techniques has been selected the answer is simple (Yes, at least on the basis of the preliminary analysis they presented).

One of the most ancient tests of the scientific quality is Occam’s Razor, that science (and our thinking in general) should be as simple possible but no simpler. It would be interesting to perform an audit of the presentations at any contemporary conference against this criterion. I suspect the results would be quite sobering.


Lance, J. (1980). Pathophysiology of spasticity and clinical experience with baclofen. In R. Feldman, R. Young & W. Koella (Eds.), Spasticity: disordered motor control (pp. 485-495). Chicago: Year book medical publishers.


An ideal treadmill?

We’ve recently been measuring the oxygen cost of walking in a group of amputees. The measurements we’ve made seem too good to be true. When we compare our results with those in the literature our amputees appear to be using considerably less oxygen to walk for a given distance. The differences are so substantial that I’ve asked our research fellow to look for possible explanations and one of the differences with those other studies is that we are making measurements with people walking overground whereas all the other studies have examined treadmill walking. This raises the old chestnut as to whether treadmill and overground walking can be considered equivalent or not.

In a sense the answer to this question has been known since 1632. In his Dialogue Concerning the Two Chief World Systems, Galileo Galilei proposed the hypothesis that has since become known as the principle of Galilean relativity that:

any two observers moving at constant speed and direction with respect to one another will obtain the same results for all mechanical experiments

He illustrated this with a thought experiment based on a ship, in the modern world I think the example of a railway carriage is more appropriate. The hypothesis states that if you are in a sealed railway carriage then there is no physical experiment you can perform that will allow you to know whether you are stationary or travelling at constant speed (or the speed at which you are travelling). If, for example, you drop a weight, it will always fall vertically (as you observe it). Galileo stated this as a hypothesis, Newton went a little further and stated it as a principle, Einstein went even further and used this thought experiment for his theory of general relativity. Many experimentalists over the years have sought to disprove it, all have failed (except for those conducting experiments at near the speed of light).

Given that the body acts as a physical system (at least as far as the biomechanics of walking is concerned) we can use this principal to explore treadmill walking. Consider walking up and down that sealed railway carriage. There is no way you can know what speed the train is travelling from how it feels to walk. If you put an oxygen mask on and always walk at the same speed within the carriage you will always make the same measurement (or would if Oxygen consumption measurements were at all reliable!). Now assume that the train is travelling backwards at a speed that is identical to that at which you are walking forwards. Again we can argue that you will be consuming exactly the same amount of oxygen as you would if the train were stationary. From the perspective of someone watching from outside the train however you are walking on the spot. This is effectively a treadmill, a highly impractical treadmill, but a treadmill none the less. Under these circumstances it is clear that the oxygen consumption (or any other biomechanical variable) will be the same as for ordinary overground walking. I’ll refer to this as the ideal treadmill and reiterate that we can use one of the most widely accepted principles in the whole of physics to state quite categorically that walking on an ideal treadmill is biomechanically identical to walking overground.

I don’t go to the gym, I much prefer to run through the Cheshire countryside close to my home if I want some exercise, but those who do tell me that running on a treadmill is quite different to running overground. I think most people feel it is harder to run on a treadmill). Is this proof that Gailieo, Newton and Einstein and every other physicist who has ever tested this hypothesis are wrong – of course not. What it shows is that the treadmills are not ideal as I’ve defined it above. The important feature is that the belt does not continue to move at constant speed. When you land on the belt you exert a forwards direct force on the treadmill that tends to slow the belt down and when you push off you exert a force in the other direction that tends to speed the belt up. You are running on a non-ideal treadmill.

This has important implications for research because how much the belt varies in speed as you run on it will depend on all sorts of characteristics of the treadmill – whether the belt itself can stretch, the characteristics of the motor, whether there is any control system to help regulate speed. We can’t just assume that all treadmills are the same from a biomechanical point of view. Some I would guess, may be quite close to ideal, some, it is obvious, are very far from ideal. Many researchers have published papers comparing treadmill with overground walking but I don’t know of any of them that make the specific point that the results are valid only for the treadmill that they have chosen to perform the experiments on and should not be applied to treadmills in general. Following on from this we should not compare results from different papers on treadmill walking unless we have good reason to believe that the treadmills are equivalent.

One way of protecting against this might be to use comparative measures. We could, for example, report energy cost for amputees as a percentage of that for able-bodied controls walking on the same treadmill. This relative measure may be less dependent on the type of treadmill than the absolute measures. We’ve measured such controls in overground walking and can identify one study from another centre with equivalent data for treadmill based walking. Interestingly the differences are almost as large in the relative measures as in the absolute measures. There could be many reasons for this but one might be that the amputees and able-bodied cohort respond differently to the non-ideal nature of the treadmill. Thus an able-bodied person with intact musculo-skeletal anatomy and full proprioception might be able to adapt more easily to non-ideal treadmill walking than an amputee. As with all other measures this difference may be specific to the particular treadmill and it is dangerous to assume that this as a result applicable to walking on treadmills in general.

In summary the laws of physics dictate that the biomechanics of walking on an ideal treadmill (on in which the belt speed is constant) are the same as the biomechanics of overground walking. Any differences in biomechanical measurements between overground and treadmill walking must thus represent deviations from ideal behaviour (I’m a biomechanist so I’ll completely ignore the possibility of any perceptual or other psychologicl effects of course!) and it is likely that these vary from treadmill to treadmill. It would be interesting to know if anyone has compared results from different treadmills to investigate how significant this effect is.

What are the hip adductors for?

Various things have made me think about muscles this week and an issue that’s puzzled me for some time. Why do we have such big hip adductors? Edith Arnold’s paper (2010) is probably the most authoritative we have on the relative strengths of muscles based on muscle volume and fibre length measurements from 21 cadavers. I’ve tabulated the data below. It’s got a few surprises. The gluteus maximus is the largest muscle but several others are considerably greater capacity to generate force (because they have more shorter fibres), the soleus, for example, has nearly twice the force generating capacity.

Muscle strength

Adapted from data contained in: Arnold, E. M., Ward, S. R., Lieber, R. L., & Delp, S. L. (2010). A model of the lower limb for analysis of human movement. Annals of Biomedical Engineering, 38(2), 269-279.

What I want to focus on today though is the force generating capacity of the adductors. Summing the peak forces we get 2000N for adductors brevis, longus and magnus working together which is considerably greater the gluteus maximus and up there with the other big muscles. Of course the purpose of muscles is to generate moments and we have to take the moment arm into account as well. The adductors have some of the largest moment arms in the lower limb so their moment generating capacity is even larger than the simple peak force might suggest.

But why do they need to be so large? In both walking (Schwartz, 2008, see figure below) and running (Novacheck 1998) at a range of speeds there is a continuous abductor moment at the hip throughout stance. Indeed because the hip joint is so lateral with respect to the centre of mass of the trunk its very difficult to see how there can be anything other than an internal abductor moment at the hip.

hip adductor moment

Schwartz, M. H., Rozumalski, A., & Trost, J. P. (2008). The effect of walking speed on the gait of typically developing children. J Biomech, 41(8), 1639-1650.

Anderson and Pandy (2001) in their simulation of human walking found low levels of activation in the adductor magnus. This was confined to a short period  around foot contact presumably to supply the small adductor moment immediately after heel contact as seen above. Unsurprisingly in later analysis  (2003) of the same data they concluded that the adductors make a negligible contribution to the vertical component of the ground reaction. Liu et al. (2006) who were tracking real data, found that the adductors don’t make any contribution to the vertical or fore-aft component of the ground reaction either at normal walking speed (2006) or a range of walking speeds  (2008). 

So what’s going on? The adductor magnus is the seventh biggest muscle in the human body and yet it doesn’t seem to do anything during walking or running. There is absolutely no doubt of course that it does come into action during a range of 0ther activities, especially in sport, but how much?  Taking the argument above that it is most likely to be needed when the centre of mass of the trunk is lateral to the hip joint then  there will be relatively few occasions when this occurs in most sports (and I’d suspect even fewer in  non-sporting activities). To get to the size they are the hip adductors must have conferred some evolutionary advantage – but what?

Having written this I’ve been out for a 15km run and guess what – it is my groin (adductors) that is aching. Something really doesn’t add up!


Anderson, F. C., & Pandy, M. G. (2001). Dynamic optimization of human walking. J Biomech Eng, 123(5), 381-390.

Anderson, F. C., & Pandy, M. G. (2003). Individual muscle contributions to support in normal walking. Gait Posture, 17(2), 159-169.

Arnold, E. M., Ward, S. R., Lieber, R. L., & Delp, S. L. (2010). A model of the lower limb for analysis of human movement. Annals of Biomedical Engineering, 38(2), 269-279.

Liu, M. Q., Anderson, F. C., Pandy, M. G., & Delp, S. L. (2006). Muscles that support the body also modulate forward progression during walking. J Biomech, 39(14), 2623-2630.

Liu, M. Q., Anderson, F. C., Schwartz, M. H., & Delp, S. L. (2008). Muscle contributions to support and progression over a range of walking speeds. J Biomech, 41(15), 3243-3252.

Novacheck, T. F. (1998). The biomechanics of running. Gait Posture, 7(1), 77-95.

Schwartz, M. H., Rozumalski, A., & Trost, J. P. (2008). The effect of walking speed on the gait of typically developing children. J Biomech, 41(8), 1639-1650.

Why we walk the way we do

This isn’t really a proper post. It’s just a notification that I’ve finally recorded the last screencast in the current Why we walk the way we do series. This series of videos now forms what I see as a complete and biomechanically rigorous explanation of healthy human walking (at least for kinematics in the sagittal plane – adding kinetics, muscle activation and the other planes is a future project). In thinking these through over the last four years I’ve turned up a few surprises and I’m now convinced that there are serious flaws in most of the published explanations of walking. This latest video is no exception. It looks at the nature of the transition from swing to stance and argues that David Winter’s early view (1992) that the foot is placed “gently” is more appropriate than more recent theories that view foot contact more as a “collision”. Please leave a comment if you find the video useful or equally if you want to argue against my ideas – they are just ideas.

I’ve now tidied up the videos page on this blog-site so the videos are much easier to view from there. Alternatively you can find them on my YouTube channel.

Over the weekend I’ve also added tidied up the Resources page and made a GPS/MAP calculating spreadsheet available as well as the muscle length modelling software (for Vicon systems only) that is already there.


Winter, D. A. (1992). Foot trajectory in human gait: a precise and multifactorial motor control task. Phys Ther, 72(1), 45-53; discussion 54-46.

Taking it slowly

Hi, I’m back, refreshed from a family holiday in France and Spain and invigorated by an excellent ESMAC conference in Glasgow. Thanks to so many people that used the opportunity to say how much they liked the blog – I suppose I better keep going. The view counter passed 10,000 earlier today so let’s see if we can keep it ticking over.

I’ve no doubt that, for clinical gait analysts, the most important paper published over the last decade is Mike Schwartz’ study on the effect of walking speed on gait variables (2008). It’s the only paper that I maintain a link to on my desktop and I rarely interpret any patient data without referring to it. If you haven’t already done so then download it now and do the same (let’s see if we can knock Tom Novacheck’s [1998] review of running biomechanics off the top of the Gait and Posture most downloaded papers table and replace it with a genuine scientific study!).

walking speed

In the study quite a lot of kids were asked to walk at a range of walking speeds. The resulting gait trials were divided up into five groups by walking speed and the average gait variables for the different groups were calculated. The darker the blue in the figure above the faster the walking with the middle trace representing self-selected walking speed. You can see that the gait traces change quite considerably with walking speed even when there is nothing wrong with the child.

We were looking at data from a patient with a rare genetic disorder today. I think if I’d looked at the same data ten years ago I’d have made all sorts of pronouncement on his gait impairments. Now I just look at Mike’s paper and can say, “Yep, he’s walking slowly”, not only that but, “he’s walking slowly in exactly the same way as anyone else would walk slowly”.  It might be worth trying to work out why he’s walking slowly but there is no evidence in the gait data of any specific impairment that is affecting his walking.

I was chatting about this in the group and talking about how we walk with different gait patterns at different speeds and one of my colleagues asked quite, “Why?”. It’s one of those simple questions that caught me completely unawares and started me thinking.

Kinematically, there is absolutely no reason why anyone shouldn’t walk more slowly by having exactly the same pattern of movement but simply going through that pattern more slowly. You could thus walk slowly in a way that gives exactly the same gait graphs (after time normalisation, see previous post on this). The answer of course is that walking is not primarily driven by the kinematics but the kinetics. The way that energy flows between the segments and the way this is mediated by muscle activity depends very much on how fast the segments move. Kinetic energy is proportional to the square of speed and forces and moments act to produce accelerations.  Walking slowly efficiently requires quite different dynamic mechanisms to walking quickly efficiently.

Although this answers the question at one level it only does so partially. What would be really interesting would to be to look at how the curves change with speed from the context of how we understand the process of normal walking and see if we can explain why the gait pattern varies the way it does. Anyone who can do this easily and comprehensively has a better understanding of normal walking than I do. I’m going to have to go away and think about this.

One thing that I think would be quite instructive would be to try and do this practically. Stick some markers on yourself and record yourself walking normally. Then try and see if you can walk slowly but in the same kinematic pattern (after time normalisation). I wouldn’t mind betting that it’s not possible. Even if it is possible to match the kinematics this will require quite different kinetics and muscle activations. You may even be able to feel which muscles you are having to use differently. I suspect there’s be  a huge amount to learn from such an exercise.


Novacheck, T. F. (1998). The biomechanics of running. Gait Posture, 7(1), 77-95.

Schwartz, M. H., Rozumalski, A., & Trost, J. P. (2008). The effect of walking speed on the gait of typically developing children. J Biomech, 41(8), 1639-1650.