Re: Introspection and Explanation

From: Harnad, Stevan (harnad@cogsci.soton.ac.uk)
Date: Wed Apr 30 1997 - 19:33:26 BST


> From: Taylor, Karl <krt196@soton.ac.uk>
>
> > (4) Introspection is not the way to explain the mind
> because:
> >
> > A. It is private
> > B. It does not exist
> > C. It doesn't explain anything
> > D. *A and C
> > E. B and C
>
> Why is something that is private no explanation? I thought
> it was just C.

Because science has to be based on things that can be tested and
confirmed publically. If a physicist say "F = ma" then if I
have a measure of F and of m, I should be able to calculate a
in time I feel I'm losing confidence in Newton's Law, and confirm
that it equals a.

But the things going on in your head only YOU can know; no one can
check. If you told me that you calculated Newton's law by picturing
a moving billiard ball with a speedometer, no one would have any way to
confirm whether or not that was true.

Nor would it explain anything (though if it delivered Newton's law,
it would correct.)

> If introspection did explain anything but was private (ie.
> I, but only I, COULD really observe what was going on in my
> mind) then surely that would be an explanation of how my
> mind works, at least (which would be more than we know now,
> right?).

I'm not quite sure what you're supposing, but let me interpret it this
way: Suppose that it was really true that the way we could tell whether
two 3D objects had the same shape was by mentally rotating one of them
to see whether we could make it fit exactly with the other one.

There's nothing wrong with first coming up with that hypothesis by
introspecting. (Actually, ALL hypotheses come from introspection -- even
Newton's Law, when he first discovered it.)

But what would make the hypothesis correct would not be the fact that
you had discovered it by introspection, but that it could be publically
confirmed. [It turns out that when people do this task, their reaction
times are proportional to how much the one shape needs to be rotated to
fit with the other. There is also evidence that there is activity in
the visual association areas of the brain while people perform this task.]

But although you would THINK that we'd get a lot of good ideas about how
our mind works from introspection, it turns out that we don't get many
good ideas that way at all. (For example, the mental rotation
happens much to fast for us to notice it; it's only the careful timing
of the responses in relation to the amount of rotation that allowed
Roger Shepard to INFER that an analog process like rotation must be
going on. It doesn't FEEL as if we are doing mental rotation, and mental
rotation was not discovered because it had first been noticed by
introspection.)

Make sure you understand this distinction: Something that you have
introspected is not data, because it's private. But if it gives you an
idea that CAN be tested with public data that anyone can collect and
confirm, then it can still be the source of correct explanations of the
mind -- or anything else.

It is paradoxical, though, that the method of introspection has worked
much better for non-mind sciences like chemistry and physics than for
the science of the mind itself, psychology. In physics (together, of
course, with the public data and whatever previous explanations already
exist), introspection seems to have given theorists quite a few correct
hunches; in psychology, so much closer to home, introspection has much
more often misled us, mostly by seeming to explain, while not really
explaining.

> I think A. would only be no explanation if 1) I couldn't
> tell you, or 2) it was different from person to person (not
> private really, but individual ie. I could tell you but you
> might not understand).
>
> If I could talk I'd tell you. And I would guess that most
> people's minds work pretty much the same way (at least,
> that's my experience so far).

Even if EVERYONE said "Yes, when I try to see whether two shapes are
the same I mentally rotate them to try to check whether they match,"
that would not be evidence for anything except that that's what people
tend to say (and, because we're not blinkered behaviourists, let's also
add that it's evidence that that's what people tend to FEEL is going
on). But the fact that it's REALLY going on can only be established
from public evidence (like reaction times or brain images).

> Has it got something to do with the Other Minds problem?

It certainly does: We have no way of knowing what's going on in anyone
else's mind; and even our introspections about what's going on in our
own mind are just evidence of how it FEELS, not about what is really
going on.

Note also that introspections typically generate homuncular
"explanations": "I rotate them in my mind, and when they match, I see
they're the same." But who does the rotating, and how? And who "sees
they're the same," and how? Whoever it is, we need to know what's
going on in THEIR mind!

[By the way, anon, your points were quite good; I have no idea why you
thought you needed a cloak of anonymity...]



This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:52 GMT