Central pain masquerading as peripheral: re-examining the mechanical presentation

When pain is predictably provoked by mechanical stress, and eased by its alleviation, we quickly implicate a mechanical, or at least peripheral, nociceptive mechanism, and apply diagnoses like mechanical low-back pain that justify our favoured peripherally directed interventions. While the logic is attractive, what if central processes could mediate this presentation? Centrally mediated pain masquerading as peripheral.


We recently investigated the idea of centrally-mediated mechanical symptoms (Harvie et. al 2015 PDF). The study involved twenty-four people with the type of persistent neck pain problems seen in everyday practice, and all with pain on rotation. They performed head rotation to their first onset of pain (P1), in three virtual-reality conditions where the amount of rotation that they saw did not match reality. Instead, the viewed rotation was more or less than was actually occurring, creating an illusion of movement that was different to actual movement. Remarkably, pain with movement depended not only on how far people actually moved, but how far it appeared they had moved (see figure and explanation in caption below).

Virtual Reality Neck Pain Expt

Mean (circle) and 95% confidence interval (error bars) for the range of motion to first onset of pain presented as a proportion of the mean range of rotation for the neutral condition. When the visual feedback suggested less movement, the first onset of pain (P1) was delayed by 6%, when the visual feedback suggested more movement, P1 7% sooner.

That pain with movement can be reliably modulated by the (visual) suggestion of more or less movement (i.e. by a non-mechanical input) is significant, and prompts us to reconsider the mechanical presentation.

In the past, perceptions such as pain were simply considered a read-out of incoming information. However, it has become clear that we could not make sense of the world if sensory information was not first filtered and arranged by our subconscious. In the case of visual perception, for example, the infinite array of colours, edges and shapes are arranged by our subconscious into the meaningful objects that we see and understand. Certain rules seem to govern this process — such as the way objects are arranged according to continuity of lines, colour and motion. The rules that govern the construction of pain, while only recently receiving attention, appear to involve the brains analysis of information relating to bodily danger. Nociception is the most obvious signal of danger to the body — but not the only one. Specific movements for example, might also become signals of bodily danger because of their meaning derived from association with injury. This would explain how (visual) signals of movement may have come to be a contributor to pain in these people with neck pain.

While ample research supports the idea that signals of threat influence pain, this study suggests specifically that information about the body in space (whether visual, proprioceptive or vestibular) that has been associated with an injury, might be relevant signals of threat. Indeed their influence may even result in a clinical pattern that appears mechanical, but is in fact centrally driven.

The treatment of threatening pain-associations is an ongoing field of study. In the meantime I think that there are a few things we can do to better align clinical practice with the threat-based understanding of pain that this finding aligns with. Firstly, we can expand our minds and clinical assessments to identify both nociceptive and non-nociceptive sources of threat (guaranteed we wont treat something we don’t assess!). Secondly, we can leverage our skills in education and behaviour therapy to encourage thoughts and actions that counter threat.

Some for some ideas about countering threat/threatening associations in the clinic:

  • Nijs, J., Girbés, E. L., Lundberg, M., Malfliet, A., & Sterling, M. (2015). Exercise therapy for chronic musculoskeletal pain: Innovation by altering pain memories. Manual therapy, 20(1), 216-220.
  • Moseley, G. L. (2003). A pain neuromatrix approach to patients with chronic pain. Manual therapy, 8(3), 130-140.
  • Russek L, McManus C. A Practical Guide to Integrating Behavioral and Psychologically Informed Approaches into Physical Therapist Management of Patients with Chronic Pain. Orthopedic Physical Therapy Practice. Jan 2015;27(1):9-16.
  • Moseley, G. L, Butler D.S., Beams T.B., Giles T., J. (2012). The graded motor imagery handbook. Noigroup publications.
  • Moseley, G. L, Butler D.S. (2015). The Explain Pain Handbook: Protectometer. Noigroup publications.
  • Pain-related fear: Exposure-based treatment of Chronic Pain’ Editors: Johan W.S. Vlaeyen, Stephen J. Morley , Steven J. Linton , Katja Boersma , and Jeroen de Jong , 1st ed., IASP-press, 196 pages, 2012

Daniel Harvie

Dan Harvie BiMDr Daniel Harvie is a PhD graduate of the Body in Mind research group at the University of South Australia. His PhD thesis investigated a new theoretical model relating to how chronic pain develops – the Imprecision Hypothesis. Daniel is now part of the CONROD injury research centre at Griffith University, Queensland. Here he is focussed on better understanding chronic pain and developing brain-based treatments, with a specific focus on whiplash associated disorder. He has a clinical physiotherapy background, including a clinical masters, and has taught Pain Sciences and Clinical Reasoning to physiotherapy students.


Harvie, D. S., Broecker, M., Smith, R. T., Meulders, A., Madden, V. J., & Moseley, G. L. (2015). Bogus Visual Feedback Alters Onset of Movement-Evoked Pain in People With Neck PainPsychological Science, 0956797614563339. PDF


  1. John, thanks for the references – I have been re-reading Pamela Lyon’s (and yours and Cohen’s) papers on evolutionary responses. I think this gets to the heart of what I am trying to understand. Perhaps it questions the definition of language – specifically a ‘language of protection’. The left hemisphere is highlighted as the ‘seat’ of language however perhaps pain as an output involves communication that begs a response – a call to action when protection is key, involving core self, and right hemisphere dependent for output. Thoughts?
    Symonds LL. Right lateralized pain processing in the human cortex: an fMRI study. J Neurophysiol 2006.
    BF Skinner vs Noam Chomsky – I vote for Chomsky.

  2. John, what do you think are the selection pressures for decreased response to predators in the right visual field for many organisms right up to birds?

    In terms of the interaction between various networks, the response to salient information seems to have a theme, how dangerous is this really and what do I do about it? with characteristic alterations in behaviour from reflexive response to fine integrated movements (and all the gradients in between depending on the multitude of temporal inputs and predictive coding at multiple levels). As well, there are underlying drivers that operate. In humans, there doesn’t seem to be an asymmetry of response to threat depending on which side the threatening input is provided but you do see a shift in midline responses with people who are under ongoing ‘threat’ with unrelenting pain as an output (CRPS & tension type headaches as examples in which they have a leftward bias in line bisection tests). Does the balancing act in understanding rely only on understanding the continuum of human experiences or also the continuum of pressures leading to our species? Thanks for questioning.

    John Quintner Reply:

    Stu, of course I cannot answer your question. In fact, it may well be unanswerable.

    But I have taken an extract from Pamela Lyon’s latest paper (2015) that could be helpful to you in the context of this thorny discussion on aversive conditioning:

    “In the psychological literature assignment of an existential value to the sum of information concerning current circumstances, internal and external, is called valence, which has been described as ‘a basic building block of emotional life’ (Barrett, 2006).

    Valence refers to the attractiveness, acceptability or tolerability of a stimulus (Lyon, 2014). Usually such stimuli are rendered in terms of ‘positive’ or ‘negative,’ ‘attractant’ or ‘repellant,’ in addition to neutral stimuli to which the organism is indifferent, but these characterizations are not always stable.

    Aversive conditioning depends on the capacity of previously neutral or even positive stimuli to become negative for an organism under particular circumstances. Thus, all sensory signals have a context-dependent valence—positive, negative or neutral—which influences the available response.”


    Barrett LF. Valence is a basic building block of emotional life. J Res Pers 2006; 40: 35-55.

    Lyon P. Stress in Mind: A Stress Response Hypothesis of Cognitive Evolution: In: Caporael LR, Griesemer JR, Wimsatt WC eds. Developing Scaffolds in Evolution, Culture, and Cognition. Cambridge: MIT Press 2014:

    Lyon P. The cognitive cell: bacterial behaviour reconsidered. Frontiers in Microbiology 2015 doi: 103389/fmicb.2015.00264.

  3. John Quintner, Rheumatologist says

    Tori, the most compelling argument we can make against pain being a CR is that as such it would confer no evolutionary advantage. One could even argue that if that were the case, it would be evolutionarily disadvantageous. By contrast, avoidance and fear of the UCS (i.e. the CRs we have been stressing) do have evolutionary advantage. That is self-evident.

    Kind regards

    Brian and John

  4. Tory Madden says

    Hello John and Brain

    Thanks for giving such a thorough response to my question! I must say that I was hoping for some neurophysiological reasons, but I will respond in brief to those you have raised.

    This is certainly a thorny, but fascinating question – at least, I think it is fascinating.

    As you say, most fear conditioning research considers pain to be a US and fear the consequent CR. I agree that that is a sensible way to view things, if you are inclined to measure fear. I also think it is worth considering that the fact that many fear conditioning studies do not demonstrate pain as a CR may be because (a) they do not measure pain or (b) they are not designed to elicit pain.

    I also find it helpful (for myself) to conceptualise the pain-fear relationship in steps. For example, if I experience certain nociceptive input in a context such that the combination of that nociceptive input and the context is sufficient to elicit pain, that pain is a multidimensional experience, involving sensory and affective aspects. I may also have an affective response to that (mutidimensional) pain – fear, perhaps. If I am an effective associative learner, then that fear will probably be directed toward the stimulus to which I attributed the pain. Certainly, my conscious experience is of pain, but it is possible that considering the conscious experience alone fails to take in the whole story. In fact, there was a stimulus that elicited the nociceptive activity (as the sufferer, I think that stimulus is the cause of my pain, but I only have my conscious experiences to go on). Then the nociceptive activity served as a stimulus to the brain, as did the context. (Of course, I am oversimplifying with the ‘context’ idea, but conceptually I think it works.) Pain was generated, and that is what I experienced.
    Now, fear conditioning studies usually consider pain to be the stimulus, but viewing things in this way means that pain is relegated to response, and nociceptive input (and possibly other things – we might say threat signals) is positioned as the stimulus that was primarily responsible for eliciting the response.

    I acknowledge that this is an unconventional take on the roles of various stimuli/responses in the classical conditioning framework, and it does tend to some feathers. To me, though, it still seems plausible, albeit undemonstrated. However, the proof is in the pudding!

    Incidentally, I read the Price quote that you used differently to how you seem to. You paraphrased him as describing fear/distress as a ‘reaction to pain’. I think his wording indicates that he considers the fear/distress to be part of the pain itself. In fact, I think that fear could serve as either – as an inherent part of the pain experience, or as a reaction to that pain experience.


  5. John Quintner, Rheumatologist says

    Thank you for your question, Tory.

    Our proposition that there may be sound neurophysiological reasons why pain cannot become a CR was intended to be expressed very tentatively. We hoped that an interested neuroscientist would respond to that claim.

    Both Marin [2001] and Price [2002], for example, indicate that the brain processes nociceptive input in such a way that it gives rise, not just to the experience of pain, but also to the subjective affective response of fear. Therefore, from the perspective of the clinician/observer, pain appears to be a response to existential threat, in accordance with the IASP’s working definition of pain.

    However, in aversive conditioning the CR is invariably fear or anxiety and never pain. From this knowledge we speculated that there might be reasons why only those brain activities relating to fear are involved in the conditioning procedures. We are not neurophysiologists so our speculations may be without foundation. Others might be able to clarify the situation.

    We believe that there are stronger conceptual grounds for dismissing the claim that pain may be a CR and that pain ought not to be conceptualised only as a response.

    Regarding the lived experience of the person, pain is perceived as a stimulus. We are convinced that in Pavlovian conditioning the experience of pain functions as an unconditioned stimulus (UCS) with withdrawal, if possible, and with fear as an unconditioned response (UR).

    In conditioning theory it is generally conceded that for conditioning to take place the subject must not only attend to both the UCS and the neutral stimulus, but also to comprehend the relationship between them. One of the required characteristics of an UCS is that it is followed by an automatic response (UCR), without any prior learning.

    Another of the characteristics of UCRs is that by classical conditioning they may become CRs to a previously neutral stimulus. In Pavlov’s investigations the significant response to meat powder (UCS) in a dog’s mouth was salivation (UCR). As a result of the conditioning procedures presentation of various neutral stimuli came to be followed by salivation (CR).

    All of the above is well known, of course, but let us return to the experience of pain.

    First, we note that the experience of pain is followed by certain responses occurring automatically. To quote Price [2002]: “Part of the affective dimension of pain is the moment by moment unpleasantness of pain made up of emotional feelings that pertain to the present or short term future such as distress or fear.” These reactions to pain are of course unlearned. Pain, therefore does have the characteristic properties required of UCS with, among other things, fear being an UCR.

    Second, we may note that when a stimulus such as an electric shock of sufficient intensity to cause pain is used in classical conditioning, the CR to a previously neutral stimulus is fear, and not pain. This again supports the claim that pain can be an UCS with fear as an associated UCR, which may become a CR. Pain cannot become a CR because in a classical conditioning context it functions as a stimulus and not as a response.

    To further support our claim that pain is an UCS in Pavlovian conditioning terms, let us repeat that it is widely accepted that for conditioning to take place the subject must attend to both the UCS and to the neutral stimulus.

    To what does a subject attend when electric shock is used as a stimulus to evoke pain? Many researchers refer to the shock as the UCS but, again, to what is the subject attending? As Heft [2007] reminds us: “Perceiving is an act of selection and selectivity in perceiving stems in part from a developmental history that is a collaboration with other individuals, if sometimes only incidentally.”

    In the context of this discussion, we are strongly of the view that the pain being experienced is the focus of the subject’s attention. The literature on overshadowing [for example, see Wasserman & Miller, 1997] convinces us that the experience of pain would likely take precedence over all else that the subject could possibly be aware of at the time. Consequently we believe that the experience of pain functions as the UCS. Furthermore, on this ground alone we call into question the hypothesis advanced by Moseley and Vlaeyen [2015] that posits chronic pain as a CR.

    Heft H. The social constitution of perceiver-environment reciprocity. Ecological Psychology 2007; 19: 85-105.

    Marin S. Neurobiology of Pavlovian fear conditioning. Annual Review of Neuroscience 2001; 24: 897-931.

    Moseley GL, Vlaeyen JWS. Beyond nociception: the imprecision hypothesis of chronic pain. Pain 2015; 156: 35-38.

    Price D. Central neurological mechanisms that interrelate sensory and affective dimensions of pain. Molecular Interventions 2002 Oct; 2(6): 392-403.

    Wasserman EA, Miller RR. What’s elementary about associative learning? Annual Review of Psychology 1997; 48: 573-607.

  6. John Quintner, Rheumatologist says

    Tory, thanks for returning to this discussion. We are now preparing a detailed response to your question.

  7. Tory Madden says

    Hello John and Brian, I have just returned to this discussion to consider some of the issues raised in it. I am particularly interested in your view that there are ‘sound neurophysiological reasons’ why pain cannot be a conditioned response. Would you mind explaining your reasons for my benefit?
    Many thanks,

  8. John Quintner says

    Daniel, before comments on this post are closed, we would like to briefly remark upon this extract from the paper by Moseley & Vlaeyen (2015):

    ‘Our hypothesis considers the nociceptive input as the US for which there are no special conditions necessary to evoke pain as the UR. The multisensory and meaningful events that routinely coincide with the nociceptive input are the CS. Once the nociceptive and non-nociceptive inputs are associated, a process termed “acquisition,” not just the initial multisensory event will elicit the painful response but also events that share some features with the multisensory event. This process is called stimulus generalization’ (pp. 35-36).

    We will comment on this hypothesis section by section.

    1. The first sentence is acceptable. If you give someone an electric shock it will hurt.
    2. ‘The multisensory and meaningful events that routinely coincide with, or preempt, the nociceptive input are the CS.’

    Moseley and Vlaeyen need to explain exactly what they mean by multisensory and meaningful events as being the CS. Taken at face value, we believe that this assertion is unsustainable.

    If a tone precedes a painful stimulus it will eventually become a CS with fear (in cognitive terms, anticipation of being hurt) as the CR.

    In the context in which the tone is presented, there can be a multitude of stimuli associated with the room, e.g. the ambient temperature, the type of chair on which the subject is seated, the apparatus being employed, the personal characteristics of the researchers, and so on.

    It has been very clearly established that unless the subject is attending to (i.e. focusing upon) the tone it will not become a CS, it is also clearly established that other contextual stimuli will not trigger the CR.

    Pavlov himself demonstrated that when two potential conditioned stimuli (e.g. a tone and a light) are presented simultaneously, with one being slightly stronger than the other, the stronger of the two will become the CS. The weaker stimulus will be far less effective as a CS.
    In learning theory this phenomenon is referred to as overshadowing – i.e. what is being focused upon becomes the CS.

    Are Moseley and Vlaeyen implying that because ALL of these stimuli routinely coincide with pain they become a single complex CS, or multiple CSs? But the separate elements can’t each become CS. One will become dominant, for whatever reason, and the others will become far less effective or nor effective at all (i.e. overshadowed).

    Thus we don’t understand what they are referring to as multisensory and meaningful events. It is very difficult to come to grips with their argument because they make no reference to relevant research.

    But before completely dismissing this aspect of their argument we need to comment upon conditioning and generalisation when humans are the subjects in classical conditioning research.

    In animals the about-to-become CS is perceived for example as a tone of a certain frequency, but with human subjects the about-to-become CS has both perceptual and conceptual aspects.

    A study conducted some time ago illustrates this. An individual with an obsessive-compulsive disorder had a fear of a diversity of triggers (e.g. dogs, petrol, knives) because of their conceptual relevance, i.e. they were potentially life threatening.

    A further conceptual elaboration was in terms of contamination, which of course might also be life threatening. In respect to contamination, when the patient was told that her sister-in-law was ill with diarrhoea, the presence of her sister-in-law later elicited disgust and triggered compulsive-obsessive behaviour (conceptual generalisation), as did a photo of her sister-in-law (perceptual generalisation).

    3. ‘Once the nociceptive and non-nociceptive inputs are associated, a process termed “acquisition,” not just the initial multisensory event will elicit the painful response but also events that share some features with the multisensory event.’

    Again we have some difficulty following their argument. It seems they are again referring to stimulus generalisation. If that is indeed the case, the above sentence could be translated as follows: ‘Once conditioning has taken place, stimuli similar to the CS may also become CS, (but of lesser magnitude)’. This is what is meant by stimulus generalisation.

    We have already dealt with their assertion that the multisensory event will elicit the painful response as a CR. To reiterate, there is absolutely no evidence that pain could become a CR and no such claim has ever been made before.

    There is a very extensive literature on aversive conditioning and on stimulus generalisation where shock is used as the UCS. In that large body of research the only CR reported is fear.

    We still incline to the view that there may be very sound neurophysiological reasons why pain, and other sensations, cannot ever become CRs.

    Brian Griffiths & John Quintner

    Reference: Mosely GL, Vlaeyen JWS. Beyond nociception: the imprecision hypothesis of chronic pain. Pain 2015; 156: 35-38.

    John Quintner Reply:

    Addendum: The claim that pain may become a conditioned response has very significant theoretical and practical implications. We have taken that claim seriously and believe that it deserves to be seriously discussed on Body in Mind. We are somewhat disappointed that such a discussion has not as yet been entered into.

  9. John Quintner says

    Daniel, I must admit to having difficulty in my efforts to fully understand the paper by Zaman et al. [2015] in Neuroscience and Biobehavioral Reviews.

    Could you please shed light on the issues that I raised above, as well as those that follow in this comment?

    For example, throughout the paper they refer to bodily sensations as CSs. It is not at all clear what they mean by that.

    In the second half of section 1.3 (Fear learning, perception and chronic pain) they make this remarkable assertion: “Thus, due to their association with pain these bodily sensations become aversive and with time elicit sensations of pain.”

    Is this equivalent to saying that if a particular sound is paired with a pain-evoking stimulus a sufficient number of times, the sound will become a conditioned stimulus (CS) and, because it is predictive of the presentation of a pain-evoking stimulus, will become aversive (unpleasant and anxiety provoking)? Then, it is said that given sufficient time the CS itself will itself evoke pain. What possible basis is there for making such a claim?

    In Section 4 (Chronic pain theories) they assert: “The more threatening the US is perceived (i.e. due to catastrophic misinterpretations of pain, avoidance-behavior, etc.), the more likely that CSs themselves become aversive, and obtain US properties.”

    What they seem to be saying here is that a noxious stimulus (e.g. a laser beam or an electric shock) can be perceived as more or less threatening depending upon contextual factors – expectations, suggestions, instructions, etc. The more threatening the pain-evoking stimulus (US) is perceived to be, the more aversive is the CS. The claim that an aversive CS can obtain US properties does not make any sense to me.

    As I remarked above, their theoretical model hinges on the belief that pain may become a CR (conditioned response). But the extensive literature on aversive conditioning in fact presents counter-evidence. As has often been stated, the conditional component of the experience of pain is fear or anxiety. I am therefore still puzzled as to why Zaman et al. [2015] and Moseley and Vlaeyen [2015] are proposing that pain itself may be a CR.

  10. John Quintner says

    Daniel, thanks for your response. On a quick read through the paper by Zaman et al. [2015] my attention was captured by the following: “The idea is that pain induces long-term memories through conditioning mechanisms, resulting in the reorganization of limbic structures, including medial prefrontal areas and subcortical structures such as the dorsal and ventral basal ganglia, amygdala and hippocampus.” Here the authors have fallen into the logical trap of reifying pain and then according it agentive properties in terms of changing neural circuitry. There may be more to come.

    John Quintner Reply:

    Daniel, here is another statement from the same paper that strikes me as rather odd: “Ample evidence supports the notion that pain perception is not a stable construct, but a noisy inferential process prone to errors.” [4. Chronic pain theories, p. 121] One either has the experience of being in pain or one does not have it. Where does the possibility of error lie?

    I do admit to finding it difficult to follow the thread of the argument in this paper. Perhaps others will find it easier to understand.

  11. A practical example from my own practice:
    I have a client who was told “not to take her arms overhead as it hyper extends your back”.
    There are a couple of questions that arise regarding the clients reselling of this story:
    1. Is this in fact what the practitioner said, or is it the client’s perception of what was communicated?
    2. Was there any real emphasise during that exchange on how this particular movement realities to actual functioning?

    It seems to me that it has become fashionable to sight the EP perspective, but I’ll be brave enough to state my own personal experience regarding this EP perspective in practice:
    If you are not ready to embody it in your own living, then don’t expect it to make any real difference in how your client actually experiences the way the information affects their functioning in a practical sense.
    Change arises from conscious repetition, conscious repetition is driven by a need; a practical awareness of this need to change. (Be out of pain, for example)
    Somehow my experience of hearing my clients stories regarding treatments they have received from PT or Osteo brings back the old medival perspective of using imagery to manipulate and instill a pattern of dependence on the treatment. I dare to argue that this is part of the practitioner’s training!!
    EP can’t be learned from a set list formula!
    Let’s research how the whole person learns. But that may lead to change, a dengeruous thing in the face how we have build the machinery and economy of the treatment of pain.

    Daniel Harvie Reply:

    Thanks Mastaneh. Following on from you ‘researching how the whole person learns’: Adding experiential, vicarious and other types of learning may indeed be superior which is why EP integrates graded exposure and and why you probably won’t use it in isolation.


  12. Systema practice solved a lot of these conundrums for me as a pain therapist. If I see a moving fist my brain computes very quickly the vector: ah oops a threat to my face. Let us have a startle-flinch now thank you! My SF occurs in a soma that has experienced a sound spanking as a child. In that moment two things happen: I sweat a cold hard brick of stored memory data that says buddy this is gonna hurt like THAT last time way back when. And two, as my face and neck retreat, something around c6-c7 mechanically does not like to flex and rotate quite so much because of that old whiplash injury caused by seeing Angela D. bend over with bright green underpants way back in middle school. Suddenly, a fist causes a data surge of co-morbid affairs. Central and peripheral. Cognitive and mechanical. The important thing that helps me roll with these uncertainties is to have a priority system to tell me case by case where to begin and how to proceed. This I obtained by studying another art called Applied Kinesiology where a hypothesis is testable immediately: is the pain I am feeling now because of the blunt force, diiminished ROM, or a pair of green underpants in remote past? Always it is ALL three because we keep the body in mind but with muscle testing I can determine which one is THE priority Tx and there collapse the stack.

  13. John Quintner says

    Daniel, your paper is very interesting and quite challenging.

    In the abstract, you conclude ” … visual-proprioceptive information modulated the threshold for movement-evoked pain, which suggests that stimuli that become associated with pain can themselves trigger pain.”

    Would it be fairer to conclude that it is the “expectancy” or “belief” rather than the “stimulus” that can be the trigger? As you know, the importance of expectancy has been recently reviewed by Atlas and Wager [2012].

    I also wondered whether your subjects reported that they were free of pain at the commencement of each study. This was not mentioned in the protocol.

    Reference: Atlas LY, Wager TD. How expectations shape pain. Neurosci Lett 2012; 520: 140-148.

    Daniel Harvie Reply:

    Hi John, this is cut and pasted from my facebook response to you with a little extra at the end.

    What our study shows is that when you give information that suggests more or less movement that they are actually performing, their apparant movement-evoked pain-threshold shifts. I will try to address your questions regarding our best efforts to explain this effect.

    As you say, ‘conditioned pain’ is not a proven construct, nor does it have an agreed definition. In a systematic review of this idea (led by Tory Madden and currently under review), we have defined it simply as when a CS elicits pain, either alone or in combination with other sub-(pain)threshold stimuli. A clinical example that might illustrate this is dysynchiria – where the visual suggestion of touch causes pain without actual touch. I wouldn’t doubt that there are other contributors, however, it is the visual suggestion of touch, perhaps because its prior association with noxious input, that results in a vision-evoked pain response.

    What has been shown in many experiments is what you might call ‘classically conditioned hyperalgesia’ (nocebo would be an example), where a CS increases the painfulness of an already painful stimulus. If such hyperalgesia can be conditioned in a 30-minute experiment, it doesn’t seem such a big stretch to consider that 3-months (or 3-years) of pairing of movement, touch and pain might result in neurones associated with movement and touch developing some unusual links with neurones governing pain.

    Whether the CS exerts its influence via expectancy/fear/belief is another question. Since our subjects had no awareness of the experimental manipulation, we would contend that whatever the process, it is subconscious. Of course at the subconscious level stimuli are evaluated with respect to what they mean and predict etc. Therefore I would say that it is not ‘one or the other’.

    John Quintner Reply:

    Daniel, thanks for your response. Although I am neither a neuroscientist nor an academic psychologist, it does to me seem like a “big stretch” of the imagination to consider that “3-months (or 3-years) of pairing of movement, touch and pain might result in neurones associated with movement and touch developing some unusual links with neurones governing pain.” Could you please provide evidence to support the concept of “neurones governing pain”?

    The other important issue that has arisen is a journalist from Scientific American has interpreted your study as heralding a new era in Pain Management: “On the horizon: virtual reality therapy that treats chronic pain.” Do you agree with her that “Many cases of chronic pain are thought to be the result of obsolete associations between movement and pain”?

    Daniel Harvie Reply:

    Rather than get into the evidence around ‘neurones governing pain’ I will simply correct myself and say ‘neurones governing nociceptive processing’.

    On your last question – I think obsolete associations are an excellent candidate for explaining some chronic pain. 2015 heralded two new theories, by top scientists, regarding this very thing: ‘the Imprecision Hypothesis’ and what I refer to as the ‘failed perceptual discrimination model’. I think they both present good cases…but I am looking forward to more interrogation.

    Zaman, J., Vlaeyen, J. W., Van Oudenhove, L., Wiech, K., & Van Diest, I. (2015). Associative fear learning and perceptual discrimination: a perceptual pathway in the development of chronic pain. Neuroscience & Biobehavioral Reviews, 51, 118-125.

    Moseley, G. L., & Vlaeyen, J. W. (2015). Beyond nociception: the imprecision hypothesis of chronic pain. Pain, 156(1), 35-38.

  14. The article in Psychological Science is regrettably not open access so one is not able to readily determine the basis of the authors claim stated in the article abstract that, “Our hypothesis was clearly supported…” The absence of statistical qualification in the abstract leaves the reader uncertain. A reported ‘clear’ difference of ‘13%’ without a qualifier of experimental error of measurement encompassing variation in device positioning for example, movement habituation, measurement error, and statistical significance?

    Clearly, ANY visual distraction (from pain) – ‘bogus’ or otherwise, during pain modulated movement may lead to changes movement. The experimentally induced lack of concordance between what is seen and what is done is one distraction. Alternatively, distracted by a life threatening event or an attractive passerby, it would seem doubtful that pain would present a significant limitation to ensuing movements of avoidance or attraction. Therefore, in the apparent absence of visual controls (other images / no images – distressing / relaxing etc) and pain controls (pain free subjects under the same experimental regime) how is visually dependent cortical confounding excluded?

    BiM Reply:

    Hi Manfred. We have added the PDF for this article into the blog post. For future reference, any articles published by the BiM team can be found on our Articles Page where we also provide PDFs where possible.

  15. An effect related to to the working of the placebo in other areas.
    Belief strongly modulates physological reaction in people with pain; good information helps a lot (explain pain), as well as addressing perceived threats as identified by Daniel above. The memory presence of an accident often looms large in the clinic for the client, sometimes dramatically affecting both perception and action in relation to painful symptoms.