Friday, August 24, 2007

Peripheral vision and the limits of introspection

[A big welcome to everyone coming here from the philosopher's carnival. I invite everyone to check out the philosophy blogroll I'm building. Join up, and take the code and add it to your own site!]

A joke:
Philosopher's wife: You're incorrigible!
Philosopher: No, my knowledge of my internal mental states is incorrigible.
In philosophy, there's a common notion that our knowledge of our internal mental states is on a whole different level that our knowledge of the external world. As John Searle once said, there can't be a gap between appearance and reality because the appearance is the reality.

However, it seems like we can find examples where we are mistaken about our experiences. One such case is peripheral vision. Many people are inclined to assume that their visual inputs are quite rich, including fairly detailed representations of everything in their visual field. At least that's what I used to think. In fact, though, this is false: our most detailed information comes only in a fairly narrow range, something like 10 degrees or something out of a 90 degree visual field. How could anyone fail to realize this? The reason is that our eyes have a tendency to dart back and forth without our noticing it. What we may believe is a change in concentration is often really an eye-movement, taking some things out our central vision and putting others in. However, diligently forcing your eyes to stare at one spot while shifting your concentration generally results in finding that you can get a vague idea of what's there, but not make out exact shapes or letters (try it with this blog post!)

Is this a blow to our intuitions about our knowledge of internal states? I don't think so, not if the original thesis is properly understood. Among other things, it never claimed our memories are infallible, and this includes memories of previous conscious states. Since we must think and introspect in real time, memory errors can screw up attempts at purely logical reasoning and introspection. This doesn't threaten our knowledge of basic claims like "I am having a conscious experience" (which some people feel the need to try to doubt).

On the other hand, it should show us that there are some very definite limits to introspection. While I find myself agreeing with those who say that there is something mysterious going on with human consciousness, I cannot agree with people (from Sam Harris to philosophers with a solid reputation in academic circles) that refined introspection is the key to learning about it. Introspection can teach (indeed, has taught) us some things, but to get the sort of detailed understanding of consciousness that we have for other subjects ee need conventional, third-person science. As a closing note, I'm optimistic about our prospects here: there is very little reason to think consciousness exists in some other realm, inaccessible to third-person attempts to learn anything about it. After all, we can learn something about other people's conscious states via their own self-reports, suggesting there is no in principle barrier to more refined investigation.


Steven Carr said...

'In philosophy, there's a common notion that our knowledge of our internal mental states is on a whole different level that our knowledge of the external world.'

So philosophers never say 'Can you smell smoke?' , or 'I think I can hear something downstairs'

Tanasije Gjorgoski said...

Of course, the issue might not be about knowledge of our "internal mental states", but about the knowledge of our capacity to see things clearly when they are 10 degrees away from the direction in which we are looking.

We think we can see them clearly even without looking directly at them, but we can't. And we don't know as we never tried.