What do we know for sure about the brain

If I was going to list everything we knew for sure about the brain, it would be a very, very long list. Instead of trying to do that, I am going to focus on things we know relevant to my last post, which was quite negative about the new "brain mapping" initiative, and which generated a lot of criticism
(http://www.psychologytoday.com/blog/fixing-psychology/201304/why-brain-mapping-is-stupid-idea). The title of the post then, should maybe be less “What do we know about the brain?” and more “What are some first principles we can use to understand how the brain operates?”

1.  The brain is deeply interconnected. We used to have this simple imagining that there were distinct brain areas that sent signals in a linear direction, one to another, so that a sensory signal eventually resulted in a motor movement. We now know full well that this is not how the brain works (and we were told it was too simple over 100 years ago, http://psychclassics.yorku.ca/Dewey/reflex.htm). No area of the brain is more than a handful of neuronal connections away from every other, and this includes connections going “backwards” to so-called “earlier” processing areas. Even the popular press has picked up on this! (e.g., http://www.bbc.co.uk/news/science-environment-10925841). We used to be most confident about this linear story, I believe, when discussing the visual cortex; we even labeled the areas in the numerical order we thought they linearly processed “information”, e.g. V1, V2, V3. However it is now well known that the feedback system extends all the way back to V1, and these effects have been shown relevant to functional studies (e.g., http://www.pnas.org/content/early/2009/12/10/0907658106.full.pdf).
2. The location of “functional areas” in the brain flexibly emerge over the course of development. For more on this point, see almost any of the amazing work coming out of Leah Krubitzer’s lab (http://psychology.ucdavis.edu/labs/krubitzer/) on the effects of behavior in determining the location and size of brain regions (e.g. http://psychology.ucdavis.edu/labs/krubitzer/pubs/arealization_krubitzer_huffman_2000.pdf), the flexibility of brain areas forming in modified brains (e.g. http://psychology.ucdavis.edu/labs/krubitzer/pubs/Padberg_Recanzone_2010.pdf), or the incredible diversity in the organization of brain areas in normally developing organisms (e.g., http://psychology.ucdavis.edu/labs/krubitzer/pubs/All_Rodents_Are_Not_The_Same_%282011%29.pdf). Though some of these findings are new, the basic principles are not. Dr. Krubitzer won her McArthur fellowship 15 years ago. I emphasize Dr. Krubitzer’s work because I particularly like it, but it should be clear that the basic principles she and her colleagues have worked on will generalize fully to humans. The fact that many fMRI studies limit themselves to right handed people to avoid the most well know predictor of massive brain reorganization should be a clue to this, not to mention the extensive research on brain reorganization in people with various disabilities.
3. Naming “functional areas” based on the first thing we figure out they can do is silly. Traditionally, if we don’t name a brain area after the person who “discovered” it, we name it after the first function that we find for it. For example, the “face recognition” area, which we now know does much more than recognize faces. Or, take the weird discussions that happen involving “mirror neurons” in the “motor cortext”. A huge chunk of the discussion involves asking why there would be neurons that fire when others do something, in the part of my brain that is supposed to control my movements. This discussion is dependent upon the assumption that we somehow correctly determined the entire function of that brain region when we labeled it "motor cortex". Instead, we should be realizing that our label was wrong, and that said region of the brain does things we didn’t previously expect. That, in and of itself, is not mysterious at all.  
4.  Finding “functional areas” is often dependent on incredible amounts of suspect data manipulation. This point has been around for a long time, but was driven home undeniably a few years ago with the study that found neural activation in an fMRI of a dead fish. (I am not joking, read more here: http://blogs.discovermagazine.com/neuroskeptic/2009/09/16/fmri-gets-slap-in-the-face-with-a-dead-fish/#.UW1NGsqwWSq) Now, you can do fMRI with less processing of the data, and it works perfectly well, but it doesn’t find “areas” in the same way you might hope. See, for example, the incredibly cool work of Ken Norman over at Princeton (http://psych.princeton.edu/psychology/research/norman/index.php). If you take whole brain data during a task, say a memory retrieval task, you can then train an artificial neural network (i.e., a fancy regression equation) to distinguish trials in which people correctly recalled a word. However, when you later go and see what the network is doing, you find that it is using a ton of weak weightings scattered across a large part of the brain. Huge areas of the brain are involved in complex mental functions, and we will not find clean, task-specific areas once we start looking at the data in better ways.
5. What the brain does is dependent upon the larger bodily systems and the larger world the brain is in. There is, for example, crazy cool research over the past decades showing the unexpected importance of gut-brain connections (e.g., http://asp.cumc.columbia.edu/facdb/profile_list.asp?uni=mgw13&DepAffil=Psychiatry&view=research, http://cdp.sagepub.com/content/15/2/84.abstract, and http://blogs.discovermagazine.com/notrocketscience/2011/01/31/gut-bacteria-steer-the-development-of-the-young-brain/#.UW1SCMqwWSo). As for the roll of the larger world in brain functioning, this is what the ecological psychologists, dynamic systems folk, and radical embodied cognition folk have been studying for decades (e.g., http://psychsciencenotes.blogspot.com/2011/07/theres-more-to-us-than-our-brains-so.html, http://psychsciencenotes.blogspot.com/2011/08/what-does-brain-do-pt-2-fast-response.html, and http://fixingpsychology.blogspot.com/2011/12/but-what-about-brain.html,). While these approaches are deeply rooted in contemporary, cutting-edge neuroscience, the basic underlying principles were worked out over 100 years ago, with what little they knew about neuroscience then (e.g., the work of E.B. Holt).  Until we come up with better ways of talking about the role of the brain in these larger systems, we won’t be able to create anything like a sensible “functional map” of the brain.

Now, for some disclaimers: First, I think physical mapping of the brain is a great idea, and love pure neuroscience; it is the so-called functional mapping that I am suspicious about, especially when it is combined with particularly shoddy psychology (e.g., http://psychsciencenotes.blogspot.com/2010/03/on-why-fmri-is-bullshit.html). Second, I don’t doubt that we will learn much from the brain-mapping effort, I just don’t think we will learn what we will be promised, and I think the things we will learn could be gotten more cheaply (that was the actual theme of my prior post http://www.psychologytoday.com/blog/fixing-psychology/201304/why-brain-mapping-is-stupid-idea). Third, this is not a complete bias against cognitive psychology or fMRI work in general. As indicated above, there is fMRI work that I really like, and there is some cognitive psychology that I really like. What I don’t like are attempts to make the brain seem simpler than we know it is. The brain is a crucial component of the larger organism-environment system, and is an incredibly complex organ. It develops dynamically in both the short and long term, and that flexibility is the source of much (if not most) of the amazing things humans can do, which other organisms cannot. We do it a disservice when we promote overly simple ideas about how it works and what the best ways to study it are.

--------
Cross-posted at: http://www.psychologytoday.com/blog/fixing-psychology/201304/what-do-we-know-sure-about-the-brain