Is it time for a conceptual revolution in neuroscience?

A recent article by Russell Poldrack and colleagues begins with an apt quote from Rutherford D. Rogers, the former Yale University Librarian:

“We’re drowning in information and starving for knowledge”

They chose this quote to reflect what some brain researchers (myself included) might call a major roadblock in contemporary neuroscience – a huge amount of data covering a conceptual landscape riddled with ambiguities. Although there has been great technological advancement in neuroscience – for instance, the relatively rapid development of functional magnetic resonance imaging and optogenetics – the conceptual foundation and integration (or what the authors call the ‘semantic infrastructure’) necessary for bridging the brain-psychology chasm is not as strong.

There are too many examples of this to note. I would point, for instance, to concepts like reward or creativity. The authors point out ‘working memory’ (Is it holding information in one’s awareness? Manipulating information in awareness? Does it involve both of these ideas or even others?). They then discuss the longstanding tendency, particularly within cognitive neuroscience, to equate the type of task performed in the laboratory with the mental construct. For instance, they point out, the Sternberg Item Recognition Paradigm is typically used to investigate aspects of working memory (and inferences to working memory ability or impairment are justified in most cases), but not all of the task conditions within the paradigm can be equally associated with working memory (e.g. there are cognitive and motor aspects that would likely not be associated at all with working memory in terms of brain functioning). And so, calling it the Sternberg working memory task, as many do, conflates the task with theoretical assumptions about which many experts might not agree.

The issue of fuzzy concepts and inconsistent usage of terminology is, of course, not new to science as a whole. Nor is it new to neuroscience – from Bennett and Hacker’s ‘Philosophical Foundations of Neuroscience’ to other recent articles on the subject, such as Bilder’s ‘Neuropsychology 3.0: Evidence-based science and practice’. Every branch of science (and perhaps every rigorously studied topic, including those not using the scientific method directly) has had to come to terms with its terms. Poldrack and colleagues point out the relatively well-known example of Gene Ontology – which provided explicit descriptions of various gene-related concepts (e.g. regarding biological components or processes). What is new, is the creation by the authors of an informatics resource which aims to identify particular usages of all terms and associated sub-terms, effectively allowing anyone to perform an ‘intelligent’ search of such concepts.

Their Cognitive Atlas is an open collaborative project (inspired by projects like Wikipedia). It will avoid imposing a single ontology, which would cause considerable consternation across subfields, and will embrace the current messy state of affairs which represent the agreements and dissenting opinions of experts.

Overall, this seems to me to be a great approach to the problem of fuzzy concepts. I wonder though, should this be a ‘cognitive’ atlas? I’ve often found the term ‘cognitive’ to be too fuzzy in and of itself and wonder how (and if) this will impact this project. If the authors are looking to help bridge the divide between brain function and psychology in general, I suspect it would be best to include all brain-related concepts (including those used in all of neuroscience, psychology, psychiatry and also those in non-human animal studies). It just might turn out, for example, that the difference between the ‘reward’ concept across disciplines, and between studies involving humans and other animals, might not be all that large. Or better yet, the conceptual differences in one area might reveal a limitation or unexplored terrain in another.

When we finally escape the conceptual quagmires within neuroscience, the next great challenge will be to align the neuroscientific concepts with everything else.


Selected references
Bilder RM (2011). Neuropsychology 3.0: evidence-based science and practice. Journal of the International Neuropsychological Society : JINS, 17 (1), 7-13 PMID: 21092355

Cromwell HC, & Panksepp J (2011). Rethinking the cognitive revolution from a neural perspective: How overuse/misuse of the term ‘cognition’ and the neglect of affective controls in behavioral neuroscience could be delaying progress in understanding the BrainMind. Neuroscience and biobehavioral reviews PMID: 21345347

Marshall, P. (2009). Relating Psychology and Neuroscience: Taking Up the Challenges Perspectives on Psychological Science, 4 (2), 113-125 DOI: 10.1111/j.1745-6924.2009.01111.x

Poldrack RA, Kittur A, Kalar D, Miller E, Parker S, Sabb F and Bilder RM. (2011) The cognitive atlas: toward a knowledge foundation for cognitive neuroscience. September 2011|Volume5|Article17| 1


About davejhayes
This entry was posted in fuzzy concepts. Bookmark the permalink.

6 Responses to Is it time for a conceptual revolution in neuroscience?

  1. quasihumanist says:

    Psychology is usually classified as a social science, and rightly so, because its terms cannot be defined outside of a social and cultural context. For example (and this example is essentially due to the philosopher Wittgenstein), it makes no sense to say that person A is in love with person B for only a duration of 30 seconds, even though it is certainly possible to imagine that person A has the brain state (okay, it is not a simple brain state and this example is too simplistic, but..) of being in love with person B for approximately that amount of time.

    To the extent that neuroscience draws from psychology, it is going to have this feature. To get rid of this feature, you basically have to turn neuroscience into brain biology and go down the path of defining terms in such a way that they are based on biological features rather than more general human concepts.

    It is productive to map and explain the conceptual quagmire you are in, but some swamps are good for the environment and shouldn’t be drained.

  2. Rich says:

    I think part of the problem stems from the lack of recognition among neuroscientists that they are really studying psychological concepts. For example, ‘reward’ and its cousin ‘reinforcement’ is a difficult (but essential) concept in psychology. Many of the mistakes that neuroscientists make have been made previously by psychologists some 20 or even 50 years ago, and the pitfalls have been well-navigated since then. Naming experimental effects or tasks after the theoretical construct used to explain performance in the task is a beginner’s mistake (e.g., latent inhibition), and shouldn’t be repeated by any knowledgeable scientist.

  3. Drawing a line between psychology and physiology may never be completely possible. However, it seems that agreeing on vocabulary and descriptions would be the first way to try to make sense out of a wealth of data.

  4. Rich says:

    Yeah, I think there are heaps of instances where neuroscientists and especially psychiatrists interested in neuroscience assume that just because the task is called a ‘working memory’ memory task, or even worse, a ‘theory of mind’ task, that it only engages those psychological processes. The mistake is compounded when such tasks are used in imaging genetics studies which attempt to identify ‘working memory’ genes by correlating genotypes with ‘working memory’ activation from functional MRI data.

  5. davejhayes says:

    Thanks for the interesting comments and discussion. I do think it is the primary job of neuroscience to map brain function (and therefore biology) in terms of behaviour (including thoughts and emotion). To do this, I think it’s fair to talk about, for instance, reward-related brain function, but not to confuse the psychological concept of reward with that of a biological correlate. There is no ‘reward’ in the brain, strictly speaking – they cannot be mapped one to one. Nonetheless, beginning with the neuronal correlates of a psychological concept is not a bad place to start.

    I believe that the gap will be better bridged (as is already happening) as more information helps us connect concepts at different levels – such as the concepts of water and H2O. Yes, they are the exact same entity, but at different levels of understanding, and as such, are immediately associated with different concepts. Understanding reward-related activity in the brain will lead to a finer understanding of the psychological concept of reward and vice versa.

    Thanks again for the interesting comments.

  6. Pingback: Science, Philosophy, Reason, and Life Without Memory | Social Behavioral Patterns–How to Understand Culture and Behaviors

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s