Imagine this

Exercising the brain by watching action heroes is part of an innovative pain therapy. Jennifer Trueland talks to the experts

It’s an unusual kind of therapy, where patients can be encouraged to go to the cinema and imagine they are Indiana Jones. Or where sitting in a shopping mall watching strangers go by and thinking of your hand or arm replacing theirs, is therapy, rather than voyeurism. But that can be the advice given to patients undergoing graded motor imagery. This is a sequential set of treatments for patients with serious pain – such as chronic regional pain syndrome – culminating in the use of mirrors. Physiotherapists have been making use of mirrors in therapy for generations – from simply asking a patient to perform exercises in front of the glass, to using special mirrors, including mirror boxes, to treat conditions such as CRPS and the phantom pain which can follow amputation. Some of these were discussed in a Frontline feature last year (15 April, 2009).

A three-step process

Two physiotherapists in Australia have taken this concept further. Dr David Butler and Dr Lorimer Moseley are the pioneers of graded motor imagery, and are keen to spread the word of its benefits. Research conducted primarily by Dr Moseley has shown the best results are obtained when mirror therapy is used as part of a three-stage process, to ensure the brain is properly prepared before the actual mirror comes into play. It’s about recognising that the brain – in trying to protect an injured limb or other part of the body – sets up certain defence mechanisms, which get in the way of recovery. And what’s more, it’s getting results. ‘We’re talking about a group of people who were previously non-treatable – amputees with phantom pain and CRPS 1 – and for the first time we’ve been able to do something for them,’ says Dr Butler, director of the Neuro Orthopaedic Institute, which operates in Australia and the UK. ‘Of course in a lot of cases people are sceptical – for some, they’ve maybe been discharged from treatment two, three or even five years ago and we’re asking them to come back for trials. These are the worst of the worst.’

Can you tell left from right?

The first stage is to test the patient’s laterality – the ability to tell left from right. Patients with a serious problem and who have a problem limb find it difficult, if not impossible, to pick out whether they are looking at a left or right limb, says Dr Butler. Indeed, the extent of how well or badly they can do this gives valuable clues about how deeply an individual’s problems go. An online testing station (called ‘Recognise’) has been developed to assess laterality, although a quick test can be done initially by simply using a magazine and asking people to pick out left or right hands or legs. ‘If they’re slow with identifying left or right, we can quantify it,’ he says. Once the extent of the problem has been identified, work begins to remedy it. This involves practice – and lots of it. Tools include using flash cards showing left and right limbs; over time, patients get better at picking out the relevant limb. ‘It’s about training the brain, and that involves practice,’ says Dr Butler. ‘The more you do it, the better you get, but you have to keep going with it. Sometimes people see results really quickly, but then they might go back a bit, so you can’t give up.’ Another method, which has proved successful with younger patients in particular, involves using a digital camera to photograph a certain number of right or left hands or legs (with permission, of course) each day, in different settings. This last point is important because it’s about being able to carry out the tasks in different contexts, whether that be place, time of day or even whether the patient is feeling happy or sad at the time.

Imagining movements

The next stage is motor imagery. As the name suggests, this means watching people move, or imagining movement. ‘It’s about the mirror neurones, cells in the brain which fire when we watch activities or imagine activities,’ explains Dr Butler. ‘We encourage these people to go out in the world and watch people – one of the things that people in this position often don’t do is interact or go out. But you’ve got to feed the brain.’ Going to the cinema, as mentioned above, could be part of this, as could sitting on the bus, or going to a shopping mall and simply watching passers-by. The key, however, is to imagine your right or left hand, as appropriate, is replacing that of the person watched, to exercise the brain. We all do this to an extent, he says. ‘Maybe the Indiana Jones movies are an extreme case. When you come out [of the cinema] you feel exhausted, because you’ve been working the mirror neurones in the brain.’ Once the brain is prepared, then, and only then, should mirrors come into play, he says. Dr Butler admits that even 15 years ago he too would have been highly sceptical about such an approach. Now, he says, there’s lots of data to back it up (see panel: What is graded motor imagery?).

A light bulb moment?

Dr Moseley, senior research fellow with the Prince of Wales Medical Research Institute in Sydney, has conducted many of the studies. He recalls he ‘fell into’ research after he treated (‘very badly’, he says modestly) the dean of his medical faculty. ‘He did not return for a second treatment, but instead set up a doctoral fellowship and suggested I should learn how to do research “properly”. If there was a light bulb moment, it was a light bulb with a dimmer switch that slowly came on over years of clinical practice, mixed with a preference for tough questions over easy answers.’ The initial finding of the research was that simply telling people about the biology of pain actually had substantial and measurable effects on their beliefs, their function and their pain. ‘I was really surprised at the first finding, because I didn’t think simply telling people about it would do much. That it did was exciting, because then we could think about how to do it better.’ While the clinical trial evidence shows graded motor imagery is more likely to succeed if it starts with left/right judgments rather than with imagined movements or mirror movements, the reasons for this are not clear. The theory, however, is that it’s important to start sufficiently conservatively to ‘get under the radar’ of the brain’s very sensitive protective systems, and laterality seemed a worthy candidate. Indeed, starting with imagined movements or mirror therapy can make patients worse, says Dr Moseley.

Future directions

Future work includes looking at people’s feelings about their bodies and how that plays into the pain treatment process. Another big project is due to start shortly, which will examine why some people get CRPS after wrist fracture, while most do not. So there’s lots happening, but Dr Butler observes there’s a way to go before graded motor imagery hits the mainstream. ‘I don’t think we’re there yet, but we’re making progress. For example, we ran our first course six years ago and just nine people turned up – and that was in Sydney, which is a big city. This weekend we’re running a course in Bendigo, which is quite a small rural town, but we’ve got 110 coming.’ Dr Moseley’s published research involved a very structured use of the programme, with patients following each stage for a set amount of time. This was two hours a day, for two weeks, for each component. In real therapeutic situations, this can be adapted to suit the individual’s circumstances, says Dr Butler, and, indeed, anecdotally, results have been seen with a less stringent regime. Conversely, some people need to spend longer on each stage. Indeed, there is no set time frame for the process and it can be ongoing. ‘I saw a lady who had an arm torn off in a laundry accident – she had a stump just above the elbow and a nasty little phantom pain clawing away. She did the laterality exercises for quite a long time, then the motor imagery and the mirrors, and she’s still doing all three. Her laterality slowly improved, and she’s back at work, but she keeps doing the exercises.’ So, that would make him a sort of ghostbuster? ‘I suppose you could say that,’ he says. ‘But I don’t want people to get the idea it’s some kind of magic pill. ‘It’s still hard work, for the patient and the therapist. It’s not a question of doing something once and everything gets better. It can be exciting to see something work for the first time, but the therapist has to realise the process doesn’t stop there. The programme has to be monitored, there are follow-ups.’ Having said that, David Butler believes the process fits in beautifully with the self-management approach championed in UK health systems. ‘I’d hope to see more therapists take this on. I’d like to see the tool in every unit. It’s not expensive and there are no drugs, no surgery. But at last something is happening for groups of patients where nothing has been happening for years.’  FL

Further info

For details of graded motor imagery research visit and

What is graded motor imagery?

Graded motor imagery refers to the sequence of three strategies for treating recalcitrant pain, such as phantom pain and chronic regional pain syndrome.

These are:

  • laterality reconstruction – some patients with CRPS lose the ability to recognise left from right-hand Images, so they have to be retrained to do this
  • Motor imagery – retraining the brain by watching other people’s movements or positions and copying them in your head
  • mirror box techniques – to move from imagined to actual movements of the affected limb in different contexts
Research published by Dr Moseley in 2004 and 2006, and a systematic review by Daly in 2008, showed the sequential order of therapeutic intervention seems to be important in getting a better outcome.


  • Moseley L et al ‘Graded motor imagery is effective for long-standing complex regional pain syndrome: a randomised controlled trial’, Pain (2004), 108:192
  • Moseley L et al ‘Graded motor imagery for pathologic pain’, Neurology (2006), 67:1
  • Daly AE et al ‘Does evidence support physiotherapy management of adult complex regional pain syndrome type one? A systematic review’, Eur J Pain (2008) 13(4):339
Jennifer Trueland

Number of subscribers: 0

Log in to comment and read comments that have been added