Sunday 1 July 2012

Fernando Cervero: Cellular and Molecular Mechanisms of Pain

    Abstract: Pain is a sensory and emotional experience that in humans also has a strong cognitive component. We can identify the elementary neurobiological mechanisms at cellular and molecular level that mediate injury-related responses of the nervous system, yet the link between these mechanisms and the conscious perception of pain remains elusive. The challenge is precisely to identify this link.

    Cervero, Fernando (2012) Understanding Pain: Exploring the Perception of Pain. MIT Press

Comments invited


  1. Cervero said that the sentient observer must prescribe or attribute the concept of feeling to non-communicating humans or non-human animals. You (Prof. Harnad) mention that a toaster does not feel, but we cannot immediately make that conclusion in the above case. Could not our notion of "feeling" merely be a byproduct of our nervous system, even present in other vertebrates without our knowledge? We as a species constantly attempt to distance ourselves from the rest of the animal kingdom. What right do we have to assume in the negative about such a clearly adaptive function as affective experience as a reinforcement or punishment mechanism?


      1. Both our feeling and our feeling that others feel is a product of our nervous system.

      2. Yes most (perhaps all) animals feel, regardless of whether they can communicate; and toasters don't. (Where does communication stop and mere interaction begin, as between predator and prey?)

      3. But it is certainly not an explanation of how and why we feel rather than just do to say feeling is a reinforcement/punishment mechanism:

      4. What gets reinforced and punished is doing, and we can already implement such a mechanism in unfeeling robots: What's the feeling for?

    2. Dr Harnad, on what basis do you confidently affirm that most (perhaps all) animals feel in your proposition #2? If there is one thing I understood of the Turing test, is that there is no way to empirically test your proposition! Perhaps you and others should be more prudent when addressing this issue.


      @Sebastien Tremblay

      1. The Turing Test is for robots, not animals.

      2. The other-minds problem applies to every other entity but oneself: it does not apply only to nonhuman animals nor only to non-speaking humans or animals

      3. Where prudence is needed is not in solving the theoretical other-minds problem (because you know as well as I do that cows, sheep, dogs, cats, horses all feel pain).

      4. Where prudence is needed is with snails and lobsters and octopuses, where it is not as obvious that they feel, yet if we injure them, assuming they cannot feel, we do far more damage if our theoretical inference that they cannot feel is incorrect than we do if we do not injure them and our theoretical inference that they can feel is incorrect. (This is a variant of Pascal's Wager.)

      5. As we go down the phylogenetic scale, my affirmation is based more and more on prudence than on what is obvious higher up (just as it is more prudent to assume that a person is innocent until proved guilty, rather than the reverse, and along the same lines that it is better to free 10 guilty people than to hang an innocent one...)

    4. @Stevan Harnad

      Thank you for your answer.

      Although I agree with the general idea of your proposition #4, I would disagree that the value of “damage”, whatever you mean by it, is relative to the quality of “feelings” a creature is alleged to possess. If we found out, per example, that feelings (i.e. consciousness) is solely dependent upon the expression of a gene that only half of the human population is equipped with, would you argue that the death of a human from the other half is less “damage” than the death of a member from the “conscious” half?

      Secondly, I would agree that the observation of animal behavior (without any study of their nervous system) would lead one to the intuition that cows, sheep, dogs and cats all have feelings that are hardly attributable to animals like lobsters and octopuses. The problem with this intuition is that it is largely biased by the fact that we, humans, are mammals that share a behavioral repertoire with the first group of animals that allows us to easily project feelings upon them. It is a lot more difficult to do the same type of anthropomorphism with octopuses and lobsters. If I would have to draw a line between feelers and non-feelers, I would certainly restrain myself from concluding anything based solely on this intuition.

      I understand your position of “prudence” to be like considering every animal a “feeler”, until the proven the opposite, in the scope reducing any absolute value of “damage”, just like in our system of justice. Although I agree with the principle, your position should not be oblivious to the fact that we humans attribute higher values to human lives than non-human animals lives (i.e. we would rather save our child, than our dog on a forced-choice scenario), and by extension, to human needs compared to other animals needs. That being said, would you deprive yourself a having a house knowing that building that house would probably result in the death of thousands of ants, worms, plants, birds’ habitat, etc.? Would you deprive yourself from eating proteins and nutriments that are part of a healthy nutrition?

      If it is not clear that “feelings” is the currency of “damage”, and if our attribution of feelings to other animals is based on a biased intuition, I do not see how your position is tenable.

      @Sebastien Tremblay

      (1) Yes, the death of an "organism" that was just as unfeeling as a teapot would matter incalculably less than the death (or suffering) of a feeling organism.

      (By the way, how would we ascertain that your hypothetical gene makes organisms feelingless? If they acted just as if they felt, how would we know? And even if they acted differently, how would we know they didn't feel? [These are mostly counterfactual zombie speculations, not ways of answering substantive questions.])

      (2) Lobsters are very different, but I have no doubt they feel: Do you? They're not like us, but close enough to err on the side of mercy. (So what is the point you are trying to make?)

      (3) We favor our kin over non-kin. By the same token, we favor our kind over other kinds. But if we are not talking about philosopher's extreme forced-choice koans ("Which one would you save from a burning building?") but about real-life decisions in which the trade-off is much less extreme or balanced, then the answer is obvious:

      If it's a choice between dying or eating a plant (even if it feels), eat the plant. With eating animals, there is (in all but extreme cases) no such trade-off: If you eat them, it's just because you prefer the taste, not because you need to, for survival and health.

      Your question about the house, and logging, etc. is an ecological one. No, don't freeze in the winter instead of building shelter at the expense of trees and their other users. But how much of the ecological devastation we are wreaking on the planet and its creatures is really a subsistence trade-off like that, rather than the taste for a cushier dwelling.

      (We're verging here on the question of the meaning of life. And if you ask, I'll tell you what it is, because I think I know…)

  2. From the perspective of our summer school, I found it of interest to note that certain lobotomy patients, as well as patients using opiates, feel a “numbed pain,” as it were, that is to say a sensation “of pain” that is not painful. This dissociation of pain as a feeling from the experience of suffering could have rather profound implications for consciousness studies. Does this mean that the brain areas responsible for generating the feeling of pain, and the brain areas responsible for the unpleasant quality of the painful feeling, are indeed distinct? What does this imply from a computational standpoint?

    1. Very interesting thought! Does this dissociation exist in certain animals? Would there be anyway of knowing if there was? It seems to me most people agree that animals experience pain as a feeling, but if they experience suffering is another question. If we found different parts of the brain in humans that are responsible the generating of the feeling of pain and the unpleasant quality of the painful feeling, could analogous brain structure be examined in higher mammals, for example chimps?
      Also, and this is probably stretching it, but what about out of body experiences? How do these relate to consciousness? If one only remembers seeing themselves, were they not conscious of there actions? ( in the first person narrator kind of sense)

      Izabo Deschênes

  3. posted on FB during the talk

    Cervero: Nociception is detection/action (doing); pain is feeling.

    And according to Cervero, conscioussness seems to be the link between détection/action and feeling, which is similar with what i understood from Damasio yesterday : the link between action programms and emotions appears to be what conscioussness is (the concert/orchestra metaphore).

    Dr. Cervero's talk brought an (arguably...) interesting question to mind: if feeling *is* consciousness, then does "feeling" less mean one is less conscious? His talk also, for obvious reasons, makes me question the desirability of feeling (and of, thus, of consciousness)

    Feeling less means one is conscious of less, not that one is less conscious. Feeling/consciousness is all or none: An entity either feels (as an organism does) or it does not (as a toaster does not). The rest is just about what and how much you feel. The hard part is explaining how/why you feel anything at all.

    Cervero said that the sentient observer must prescribe or attribute the concept of feeling to non-communicating humans or non-human animals. You mention that a toaster does not feel, but we cannot immediately make that conclusion in the above case. Could not our notion of "feeling" merely be a byproduct of our nervous system, even present in other vertebrates without our knowledge? We as a species constantly attempt to distance ourselves from the rest of the animal kingdom. What right do we have to assume in the negative about such a clearly adaptive function as affective experience as a reinforcement or punishment mechanism?

    Yeah, I guess I really just don't understand why humans evolved to have consciousness/the capacity to feel. That's probably one of the biggest reasons I was so excited to attend this conference-maybe it would shed some light on this question.
    It makes sense to me that humans would have pain, in physical terms, because that lets you know that something is off-balance/not right in your body. BUT what really confuses me is the necessity for emotional pain, say, heartbreak. Maybe this just comes as a by-product of the capacity to feel physical pain? I don't know. Does anyone have other ideas? Or can point me towards literature?

    As for adaptiveness, heartbreak, empathy, and other emotional pain promote prosocial, altruistic, behaviour, which serves to strengthen chances of survival at individual and societal levels. Both Jackson and Damasio touch on some of these points, so you could start with them and the links given on the course site:

    1. NICO SJ
      Quoting from above: "Feeling less means one is conscious of less, not that one is less conscious." I understand your ( Turing Consciousness ) point about consciousness being all or none. Point gladly taken. However, what does being conscious of less amount to, specifically? "Conscious of less" makes me immediately think of meditation, but that is incompatible with conscious = feel, since it brings in executive function (which Dr. Tallon-Baudry made clear is NOT consciousness).

      Feeling less means feeling less: Turn down the volume, lower the lights and I feel less. But I still feel, And that's what needs to be explained.

      ‎"Feeling less means feeling less: Turn down the volume, lower the lights and I feel less." Do you mean that you feel less compare to the previous state you were in? Otherwise, I'm not sure how I should interpret this statement. And, I'm pretty sure I won't agree with this.

      If I lower the volume of my sound system, it's not that my feeling is reduce but rather that the experience I'm feeling is completely different. My experience of the music playing is changed altogether. (It's anecdotal but in some cases - mostly with noise music - artists strongly suggest that you listen to their records loudly.) In a way, I feel less of what the music should feel like (I can't no longer hear some of the details, feel the abrasive texture the composer wanted me to feel at some point); but how is that different from detection? Turning down the volume is simply making me unaware of some aspects of the composition. It's not that I'm feeling less, instead I'm no longer able to detect what was meant to be detected.

      30 juin 13:33 à 1 juillet 00:36


      Detecting is doing, not feeling. Feeling is felt detection.

      You can feel fewer things, or feel them less intensely, just as you can see fewer things, or see them less clearly or brightly. What is not a matter of degree, though, is whether you feel anything at all.

      Feeling itself (i.e. that you are feeling, not what you are feeling) is all-or-none: no degrees, like intensity.

      Or, if you like, it is like a threshold, but not a detection threshold (which concerns how intense a stimulation must be to be detected) but a threshold below which there is no feeling at all (a teapot, a toaster). At moments when that happens (perhaps in delta sleep), we are not there. It is not that we are feeling what it feels like to not-feel (that is self-contradictory), but there is simple no feeling going on at all.

      Is it clearer now what is all-or-none?

  4. Cervero briefly mentioned in his talk that we are only aware of certain external realities - some specific light and sound waves, but that we are completely unaware of the electromagnetic waves that are present around us (cell phones etc.) - however that these waves are just as real as the ones that we can sense. If we assume that humans evolved to be aware of certain things because it was evolutionarily advantageous for us to do so, do you think we would ever evolve to become aware of these waves if they are indeed as cancerous and harmful as many have suggested?

    1. Hmm... well the evolutive processes would work if let's say such waves made us sterile or killed us before we had the chance to reproduce, but only if some of us were able to escape the harmful radiation by detecting it and steering clear of it.

      What seems more likely is that we'll pay more attention to it and develop technology to measure such radiation.

      But to me, this was an extremely interesting point (but I'm not sure why it's interesting...). We, humans, have good eye sight, hear well, and smell taste and feel touch alright. So that's the kind of information we're processing all the time, that's what we go on to make decisions, etc.

      I think the combination of sight and hearing is especially important for us because it allows us to situate ourselves in time and in space. This would allow us to make more complex decisions that other animals are not forced to make as MacIver demonstrated so elegantly with the fog analogy.

      For me, it brought to light that if we're to define consciousness and to deliberate whether machines or other animals have it, it is important to recognize what kind of information different senses bring in and what kind of elaboration it allows.

  5. As said on facebook, does this theory brings something more then the portillon (gate control of pain) theory?

    Expectations: top down mediation. Perception: Bottom up mediation.

    Natural intuition makes us rub the skin where pain is felt. This stimulates big fibers which modulates pain perception. Bottom-up pian modulation mechanisms.

    Expectations mediates our attention the same way. Telling a child that a puncture is going to go well + making him talk about what he did this morning, or making him focus on his joy of the day will also modultate pain perception. Top-down pain modulation mechanisms.

  6. In his talk, Dr. Cervero assumed that pain has two components: a psychical adjunct (consciousness) and a protective reflex (nociception). His description of the processes involved in nociception were really informative, but I wonder whether being in pain necessarily comes with the psychical adjunct. Could there be pains that are not phenomenally conscious? For instance, suppose that I am sleeping - but not dreaming - and someone puts my hand on a slowly-heating object. If I am not mistaken, I should normally exhibit typical pain behavior in such a case: at some point, I will move my body away from the source of harm and I might cringe a little bit too. Do I count as being in pain? (Peter Carruthers defends a similar view concerning non-conscious pain in his 2000 book on consciousness.)

    1. More likely: the burning will wake you up...

  7. Xavier Dery ‏@XavierDery

    Again, "We don't know" a lot, but to me it indicates honest, sound science. 'Not knowing' now can be on the way of finding out. #TuringC

    11:17 AM - 30 Jun 12 via Twitter for Android