The Identity Theory and Mind-Brain Correlations

By Brandon Towl

In another thread (“What’s a good reading against type-identity theory for neuroscientists?”), I mentioned that I had a forthcoming paper in Philosophical Psychology that puts pressure on the IBE argument for identity theory.  Below is the abstract.  If anyone would like a draft of the paper, please email me (btowl[AT]artsci[dot]  — I’m sure that there is more to this issue, and I would welcome any debate, questions, refutations, etc.

Abstract: One of the positive arguments for the type-identity theory of mental states is an inference-to-the-best-explanation (IBE) argument, which purports to show that type-identity theory is likely true since it is the best explanation for the correlations between mental states and brain states that we find in the neurosciences.  But given the methods of neuroscience, there are other relations besides identity that can explain such correlations.  I illustrate some of these relations by examining the literature on the function of the hypothalamus and its correlation with sensations of thirst.  Given that there are relations besides identity that can explain such correlations, the type-identity theorist is left with a dilemma: either the correlations we consider are weak, in which case we do not have an IBE to an identity claim, or else the correlations we look at are maximally strong, in which case there are too few cases for the inductive part of the strategy to work.



  1. I would assume the hypothalamus is correlated with feelings of thirst because it causes other changes in the conscious parts of the brain. Similarly, activity in the retina is correlated with visual experience, but there isn’t grounds for an identity inference. Correlations are but one part of an identity story, but more must be added before we would want to claim identity (and frankly in neuroscience few people would talk about things this strange way).

    Perhaps an example where it is really well worked out, such as the family of voltage-dependent potassium channels, would be a good case study. Such high-level things as sensations of thirst aren’t worked out neurally, so the philosophy will be that much more tenuous. there isn’t exactly a standard method for things like that, yet.

    But for voltage-gated ion channels? You might have a recipe for a nice analysis of how things work at a mature level.

    This is from reading the abstract, so it may be completely in left field, just my thoughts as I take a break.

  2. Brandon N Towl

    Actually, I think you are right-on in many ways!

    So, the common ground

    1) I agree that in neuroscience few people would talk this way. Indeed, I find it weird that philosophers do (sometimes) speak this way. They are my target, not so much the neuroscientists.

    2) Also agreed that correlations are only part of the identity story. But getting correlations is often a much more complicated affair than what is presented in the average gloss of the research.

    Those things said, The connection between neuroscientific findings and the metaphysical theory called identity theory is a lot more indirect and problematic than most people recognize.

    I agree that there might be some cases where the story is much more straightforward/ well developed (I’ll have to take a look at the voltage dependent channels stuff). But my point is not that we *never* find identities– we may well may. But I believe these have to be taken on a case-by-case basis; there probably is no “general” identity theory.

Comments are closed.