It got me think to the quote that I cited at beginning of my thesis, which summarizes my feeling about pain and science in general: “We’re not going to get rid of chaos and complexity. They’re not going to go away. But we can find a way to live with them.” Al Held (Painter, 2001).
While I agree that sometimes we don’t see the forest for the trees, I think it’s equally reductive to try to understand a forest without knowing how trees work, so to speak. Is someone who exclusively examines cellular mechanisms missing an important part of the overall story? Probably so, but I would argue that someone who exclusively studies behavior without attempting to elucidate the cellular or network-level mechanisms that give rise to behavioral effects is missing an equally important part of the same story.
I think part of the difficulty is that our ability to develop new tools and technology is improving more rapidly than our ability to interpret animal behavior - at least it seems that way to me. (Speaking for myself, I would certainly rather try to evaluate change on a molecular level than try to determine whether a mouse is in pain, stressed, depressed, or just being uncooperative.) I don’t have the historical perspective that some late-career investigators do, but I wonder if many years ago, before we had a decent handle on genetics and neuronal tools like DREADDs and optogenetics, someone would have written a similar opinion article claiming researchers put too much emphasis on behavior and lamenting the lack of consideration for small-scale mechanisms. Currently we have a bunch of new and exciting cellular and circuit-level tools, so as a community we’re swinging towards that direction. Perhaps in time, we will swing back the other way as we run into technological road blocks or develop new & better ways to study behavior.
In a perfect world, maybe we (either as individuals or in collaboration) would simultaneously study all levels of a problem, from the single cell all the way up to the whole animal - or a whole population of people. However, if we focus on the areas with the best tools at the time, perhaps all these aspects will come together in the end as these tools continually change? It certainly doesn’t help the frustrating feeling that we’re not able to adequately look some of the more difficult aspects of a problem, or that we’re not doing enough to solve the real-world problems that real people are experiencing, but I suppose I personally take the long view that over time we as a community will accumulate enough knowledge to effect positive change.
Excellent points @liz
I especially agree with this:
In a perfect world, maybe we (either as individuals or in collaboration) would simultaneously study all levels of a problem, from the single cell all the way up to the whole animal - or a whole population of people. However, if we focus on the areas with the best tools at the time, perhaps all these aspects will come together in the end as these tools continually change?
I think studying all biological levels in an integrated way is crucial. Right now we do science in a bottom up way, each group doing their own thing. I think though that introduces so variables and potential confounds that it may be hard to integrate all those results. I like Allen Institute-style science, where they make a concerted and organized attack on a problem at all levels, and everyone talks about the approaches and coordinates. I wish we could do the same for pain neuroscience. Maybe one day.
And this 1000x 
(Speaking for myself, I would certainly rather try to evaluate change on a molecular level than try to determine whether a mouse is in pain, stressed, depressed, or just being uncooperative.)