One can grant there is a causal dependence between brain activity and experience, but it is a leap to suggest that we can identify brain activity with experience.

Argument 1: Empirical thought experiment

The two come apart in the following thought experiment:

Suppose we found an incredible empirical correlation between brain activity and experience, such that a brain scanner technology exists which can determine one’s experience from a brain state measurement.

Now suppose one day we suddenly had a conflict (the brain scanner says that the subject thinks he’s drinking coffee, but we see him eating ice cream and he assures us he thinks he is eating ice cream). In such a case, we would conclude the brain scanner is wrong, not the person. This demonstrates that, at least in cultures remotely similar to our own, conventional means of determining thoughts (e.g. first-person experiential reports, when subjects are not identified as hallucinating or having a reason to lie) have a logically prior status over any kind of scientific method.

Argument 2: Conceptual analysis

What it means to be a thought is not the same thing as what it means to be a brain state.

  • Brain states are connected by casual relationships. They, categorically, cannot be ‘correct’ or ‘incorrect’, while thoughts can be.
  • Mental states occupy the space of reasons - the meaning of certain thought is identified by its relationships to other thoughts. Structured by normative relationships (can be correct or incorrect).
  • If a brain scanner with amazing correlation says that someone is telling a lie by analyzing brain states, it may very well be that they are lying 100% of the time that brain scanner says they’re lying, but it is not the brain state itself that makes them lying or not.