Friday, November 16, 2007

On not knowing thyself

In my philosophy of mind class, the very first article in our course reader was not a philosophy article at all, but rather a 30 year old piece by a couple of psychologists. * It discussed just how wrong many people are when it comes to understanding their mental processes.

In one sample experiment, they presented people with four identical white shirts, and asked which one they liked the best. People invariably picked the one on the right. When asked why, they gave every answer but the fact that it was on the right, and acted as if the researchers were crazy for suggesting that perhaps that was the reason.

They give several other examples, but that one alone is enough to make me stop and go "WTF?"--and bug the shit out of me. Bugs me bad enough that I tend to babble about it in ordinary conversations.

You see, we're told from a young age that it is important to "know thyself," and the standard thinking on that issue seems to be that once you've imbibed the wise maxim, it's just a matter of applying it by being mildly more conscientious than you used to be. Not so simple it turns out.

Or: a few weeks ago, I mentioned becoming somewhat obsessive about honesty. Honesty, it would seem, involves telling people why you're really doing things, not the reasons that will make them think well of you. And of course, that's absolutely mandatory, because you know why you do things. Except you don't.

And it gets even worse, as the folks at Overcoming Bias tell me that there's all kinds of fakeness people people engage in, starting with fake morality (an obvious enough threat I can deal with looking out for), but also fake optimization criteria, explanation, justification, causality, and even fake selfishness. The last one has got to be the buggiest of them all, because it seems like a fairly safe way of avoiding self-deception is to think of what selfish motives you might have for doing something, and take seriously the idea that they might be your real motives. However, the arguments in the fake selfishness post suggest it's actually easy to attribute selfish motives to yourself incorrectly.

How does one live with all this? I think I need to start by telling myself that I needn't go overboard with all this skepticism about myself, given that it's based on the idea that I do know a few things about myself. However, knowing that's of little comfort if I don't know nearly as much as I used to think. To answer questions about motives with an insistence that one doesn't know risks a kind of self-deception in itself if one does, in fact, know. A compromise strategy would be present any possibly risky claims about why one did something as only your best guess. Sure to annoy people you talk to, but may be the most you can do, until you've done far more towards obeying the old slogan than most people ever dreamed was necessary.

*Richard E. Nisbett and Timothy DeCamp Wilson. "Telling More Than We Can Know: Verbal Reports on Mental Processes." Psychological Review May 1977, p. 231

4 comments:

ADHR said...

I've read that paper (not recently, though, so forgive me if I garble details). I didn't find it that convincing. It seems to assume, without argument, that the explanations people offer are mere rationalizations and not genuinely explanatory. Why is "it's on the right" supposed to be the "correct" explanation? There's some theoretical assumptions under the surface that I find disturbing.

I was told of a similar experiment involving cereal boxes. The punchline is the same: people supposedly concocted various elaborate stories about which cereal box design they liked better, when their choices could also be accounted for by the explanation "they picked the one on the right" (or possibly left; it's been a while since I heard the story). Again, though, why assume that the non-intentional explanation is the correct one?

Presumably, the implied argument is something like picking the one on the right is a simpler explanation of the choice. But I'm not convinced that it is. What sort of simplicity is at stake here? Is it ontologically simpler? No, because not all spontaneous explanations of our own behaviour are going to be confabulations. Is it epistemically simpler? No, because there's (as yet) no systematic way to tell when proffered reasons are confabulations and when they are not. So, in what sense could "they picked the one on right" be a simpler, and thus superior, explanation?

Hallq said...

I think I made this clear enough in the post: "it's on the right" is the "correct" explanation because that's what people consistently do, there is no other clear explanation, and the explanations people do give are demonstrably wrong: it can't be the fabric, for example, because all the shirts are made out of identical fabric.

Just Al said...

I suspect the biggest stumbling block in the subject is that there is such a large separation between conscious, willful thought ("reason") and un- or sub-conscious thought ("instinct" and/or "reaction"). We want to believe we are creatures of reason and control, but a significant number of our actions stem from an entirely different part of our brains.

The significance of this? Not much, once you recognize it for what it is - it perhaps matters more to evolutionary biology and social research, but a key point of the experiment was that the decisions were over trivialities. In other words, "See? We don't make reasonable decisions in situations where reason isn't needed." I'm not sure what impact this really has.

I also think it would be relatively simple to change the approach to see what affect that has. Instead of saying, "Pick the best shirt," simply say, "Take any one you like." The demands on the test subject have now changed - they are no longer being tested, but essentially, they're being asked to make the exact same decision. Also, what happens if you repeat the tests or run them in a series? Do the decisions alternate, or is this simply a matter of right-handedness favoring one side (like starting off on your right foot)?

I think that "knowing thyself" is a useful pursuit insofar as important decisions, but not to the point that one obsesses over it ;-). We cannot possibly break down everything that influences our thought processes, and trying probably isn't healthy nor effective - you can easily influence your decisions in the effort not to ever do this. And then you realize it and get even weirder ;-)

We are allowed to be creatures of instinct, emotion, and capriciousness. We're also allowed to be wrong. Trying to control that is not a useful pursuit, but simply being aware of it should be sufficient.

ADHR said...

Hallq,

There's a big lacuna in that argument, though, namely that everyone was capable of recognizing that there really was no difference between all the shirts on offer. This is what I mean about "big theoretical assumptions". If someone tells me that they like shirt x, and yet I know shirt x is physically indistinguishable from shirts a through z, why on Earth am I entitled to infer that that their reasons for not liking shirt x are no good? It's blatant projection: I know shirt x is indistinguishable, therefore they, too, should know shirt x is indistinguishable. But that in no way follows.