On Knowledge

by JS

I’ve been thinking a lot about the problem of knowing. If we believe current theories of the mind, knowledge is partially about belief states in the brain. You have them, I have them, but what are they? Let’s imagine for the moment that belief states are patterns of activation in the brain. The neurons propagate electrical signals which result in our believing something. We’ll leave aside for the moment the difficulty of connecting electrical signals in the brain with our feelings of believing something.

Instead lets focus on the other related problem of how electrical signals in the brain come to be associated with anything about the outside world. These meta questions aren’t new, but I claim that one is far easier to answer than the other. The question of connecting electrical signals in the brain with the feeling of believing something is the hard problem of consciousness. The other problem, associated signals in the brain with real states of the outside world is easy. Well, at least it’s easy enough that I feel comfortable researching the problem.

Of course I won’t solve the whole thing. I’m not even interested in electrical signals per se, but rather abstract computational models of what those said signals are representing. I’m also not wedded to the idea that I need to discover the secrets of human intelligence, discovering the secrets of any intelligence would be more than sufficiently satisfying. What makes this problem “easy” is that it decomposes into lots of sub-problems. You can answer questions about how brains can learn to work in narrow kinds of “worlds.” You can constrain the problem any number of ways.

But there’s still a philosophical question here. I propose two general sources of knowledge about the world. One is evolution – simply speaking, you know enough about the world to be here, or you wouldn’t be here. The other is data. Humans have massive amounts of available sensory experience. Surely our beliefs about world states are informed by that sensory experience. Can we say anything in general about the degree to which evolutionary and sensory knowledge (leaving out social knowledge for now) contribute to having correct belief-state/world associations in the brain?

This is the question I’ve been thinking about today. I’ll post some more thoughts later.