Can pain be a classically conditioned response?

Well, what do you think? For those who are clinicians, how does your thinking about that question influence your clinical reasoning? Have you ever found yourself saying (or thinking) along lines like this?

“Well, he still has pain, and it’s 6 years since his injury. I think that, now, bending forwards is associated with pain for him. We need to uncouple the pain from bending forwards, so that he can bend without pain.”

If you have thought like this, you’re not alone. We recently ran a survey (Madden and Moseley) to find out how many clinicians think this way – specifically, how many clinicians endorse an associative learning framework that implies that pain can be elicited by classical conditioning. What do we mean by that? Well, think back to Pavlov and his dogs (Pavlov 1928). Pavlov rang a bell whilst feeding his dogs and, after a while, the dogs salivated in response to the sound of the bell – even when the food wasn’t there. He called that a classically conditioned response.

We had the impression that clinicians think that a similar process occurs with pain – which is not implausible. We know that, after an acute injury, a variety of non-noxious signals (the position of the body in space, visual information, context, etc. – e.g. when bending forwards) are repeatedly paired with nociception (warning messages from the damaged tissue – which commonly elicit pain). It seems possible that, after enough pairings of those two types of information (non-noxious info + nociception) the non-noxious signal alone (e.g. signal of bending forwards) might elicit pain, even without the nociception. In other words, pain that persists after healing of damaged tissue could be a classically conditioned response.

We put this idea to 1090 practising healthcare clinicians, representing 57 different countries, in an online survey. (You may have been one of them. If so, we are very grateful to you for completing the survey; thank you.) First, we filtered out those who thought that pain cannot occur without nociception, because that would logically preclude them from endorsing a classical conditioning framework for pain. Encouragingly, 86% of our respondents (accurately) stated that pain can occur without nociception. Hooray for pain science education! Of that 86%, 96% believed that pain can be a classically conditioned response to a non-noxious stimulus. Now, that is an astonishingly high percentage. It works out to approximately 94% of all the survey respondents endorsing a classical conditioning framework for pain.

We were interested in what these clinicians thought their views were based on, so we asked them whether or not they thought there was evidence to support their view. Of those who endorsed the classical conditioning framework for pain, 98% believed that their endorsement was based on scientific evidence. But here’s the catch: there is no conclusive, published scientific evidence that pain can be a classically conditioned response to a non-noxious stimulus – we reported and discussed this in a systematic review and meta-analysis that was recently published in Pain Medicine (Madden, Harvie et al. 2015). What an interesting situation! How, we asked, did this come about; how did so many clinicians come to believe that there is evidence to support the idea that non-noxious stimuli are capable of eliciting pain? We thought carefully about this, and made some suggestions in the paper itself (which makes for pretty easy reading, and which you can find here). If you’re a clinician, and if you responded to this survey, please consider reading the paper and reflecting on your beliefs – and let us know what you think informs them (you could use the comments section below). Our ideas on the matter are merely speculative, and we’d be keen to hear yours.

The how did this happen is indeed interesting, but let us not forget to ask what all this means. I am in the midst of the academic debate about whether or not pain can be a classically conditioned response, and I think that there are sensible, carefully considered arguments put forward by those who support the idea and by those who don’t. Certainly, there is no conclusive supporting evidence at this stage, but the paucity of studies should not be construed to mean that there is evidence that pain cannot be elicited by classical conditioning. There are simply not enough studies to say. To me, the most important message of this survey is that those who work most closely with people in pain – the clinicians themselves – see the classical conditioning framework as plausible, and that they think it explains the clinical pictures they see in their patients. In effect, the classical conditioning framework for pain seems sufficiently relevant for clinicians to have adopted it in the absence of supporting scientific evidence. To my mind, if it seems that relevant in the clinic, it should be investigated in the laboratory. Only once we have innovatively-designed, rigorous research will we be able to properly state whether or not pain can be a classically conditioned response in humans.  If it can, then hooray! we can support clinicians in their application of the framework so as to optimise clinical strategies for treating and preventing persistent pain. If it cannot, then it will be important to more thoroughly interrogate the reasons why so many clinicians endorse the classical conditioning framework – what are they seeing that we are not, and what better explanations can be developed?

Tory Madden

Tory MaddenTory arrived from South Africa to start her PhD at BiM.  She is a physiotherapist who worked clinically before turning her focus toward research.  She is interested in pretty much anything related to pain and neuroscience, thanks to some particularly inspirational teaching by Romy Parker during her undergraduate training at the University of Cape Town.
Tory’s research looks at classical conditioning and pain.  She is also an associate editor for BiM.  She tries to spend much of her spare time exercising to compensate for the vast quantity of chocolate that lives in her bottom desk drawer.  Luckily, she loves trail running as much as she does food.

References

Madden, V. J., D. S. Harvie, R. Parker, K. B. Jensen, J. W. S. Vlaeyen, G. L. Moseley and T. S. Stanton (2015). “Can pain or hyperalgesia be a classically conditioned response in humans? A systematic review and meta-analysis.” Pain Medicine. DOI: http://dx.doi.org/10.1093/pm/pnv044

Madden, V. J. and G. L. Moseley “Do clinicians think that pain can be a classically conditioned response to a non-noxious stimulus?” Manual Therapydoi:10.1016/j.math.2015.12.003

Pavlov, I. P. (1928). Lectures on Conditioned Reflexes. New York, Liverlight Publishing Company.