That Fish You Caught Was in Pain

by Siegfried Othmer | October 12th, 2006

Boy with FishThis week we were treated to the news that in addition to his other exotic and high-living habits, the esteemed leader of North Korea likes his sashimi cut from living fish. And this week also the topic of whether fish feel pain graced our newspaper. Elizabeth Braithwaite, a behavioral biologist at Edinburgh University, wondered why that question had not arisen before now. Perhaps anglers really don’t want to know. I thought it was because the question had already been answered.

Braithwaite points out that fish have the same architecture for nociception that we do: A-delta and c-fibers. It would be remarkable indeed if fish so equipped did not have the rest of the alarm system we call pain also in the repertoire of their nervous system. Indeed, fish react aversively under noxious stimuli, just as one would expect. And if they are then given pain-killing medication, they respond to that in the expected fashion as well.

So all the evidence would seem to be in place that allows an external observer to conclude that fish feel pain. They just act as if they do, and nervous system architecture supports our observation. But ultimately we are rendering a judgment about the subjective experience of another animal, and we don’t really have a clue what that might be like. And even the matter of subjective experience presupposes a conscious entity to appraise that experience. Perhaps that’s the more serious sticking point when it comes to fish. What level of consciousness might a fish possess? What level of awareness?

Not long ago, doctors felt free to take issue with their patients who complained of pain. Maybe they believed them; and maybe they didn’t. They felt entitled to be more skeptical of any observation that rested on subjective experience. And it wasn’t long ago that doctors held to the belief–against all evidence to the contrary–that infants did not feel pain. There is no better case than this to illuminate the human capacity to give belief priority over mere observation. It made all the difference, I’m sure, that there was no tangible, measurable concomitant of pain to contradict the fond and firm belief. Once again we are stymied by subjective experience, something we cannot see or put a caliper to.

In 2005 Christopher deCharms reported on functional MRI signals in the anterior cingulate that scale with felt pain severity, and these signals were found to be responsive to feedback control. At his recent lecture at the ISNR meeting in Atlanta, deCharms thought this work might set the stage for the large-scale deployment of fMRI feedback for pain management. This is, of course, entirely unnecessary–as well as being completely unaffordable. The fMRI measure should merely be used to document our various other biofeedback techniques for pain management so that these methods will finally be taken more seriously.

The only niche that really needs to be filled here is to make the subjective visible, or at least to come as close as we can. Then we can all go on and confidently use the techniques we already know. Is there any chance at all that fMRI feedback might turn out to be superior in pain management? Not bloody likely. It is too inefficient. But even if it were systematically effective it would be entirely irrelevant as a practical remedy for cost reasons. Therapeutic efficiency is not an issue with totally disabled chronic pain patients with unlimited time on their hands. Monetary efficiency, on the other hand, matters greatly, particularly in an environment where the State of California is cutting off chronic pain care through the Workmen’s Comp system, and the VA is even cutting back on traumatic brain injury treatment for young returning veterans.

A few weeks ago, Mark Jensen showed promising initial results on pain management in Complex Regional Pain Syndrome at the Society for Applied Neuroscience annual conference in Wales. Straight-forward protocol-based neurofeedback training was used. Significant pain reduction was achieved in individual neurofeedback sessions, and such pain reduction tended to sustain longer than with hypnotherapy in the same pain patients.

The relevant point for present purposes is that we don’t need to train the “locus of pain” in the brain any more than we need to train the presumed “locus of origin” of a migraine in the brainstem, or to target the seizure focus in seizures, in order to be clinically effective with feedback techniques. It is sufficient to train on a correlate of the attribute we are seeking to influence toward improved regulation. The nervous system makes the connection; we don’t have to.

As for the fish, perhaps the issue of pain sensation can be resolved as above with a measurement under challenge with functional MRI. That only leaves the small matter of consciousness…

DeCharms, R.C., et al (2005). Control over brain activation and pain learned by using real-time functional MRI. Proc Natl Acad Sci USA. 102(51):18626-31

Leave a Reply