Psychology

[selenite0]

celticdragonfly thinks I'm being silly by keeping duplicate books in our library. Well, here's my scientific justification for keeping them. Now, if somebody else wants to trade for them, and promises to give them a good home, we can talk.

Comments

[joyeuse13.livejournal.com]

classical economic theory of people behaving entirely rationally where money is concerned.

Who on earth would expect people to behave entirely rationally?? Especially where money is concerned.

[jeriendhal.livejournal.com]

What I find unbelievable is the economists' previous assumption that "everyone reacts to money rationally." You'd think a quick trip to Vegas or a store selling lottery tickets would disabuse that notion!

[selenite.livejournal.com]

Economists who want the equations to work out neatly so they can get tenure. (http://en.wikipedia.org/wiki/Homo_economicus) There's some upstarts pushing them toward looking at what people actually do, but it's an uphill fight for them.

[joyeuse13.livejournal.com]

"looking at what people actually do"

What a novel concept.

[(Anonymous)]

It's the same trick we use to understand things in Physics. If faced with a really complex problem, then simplify it to one you can work. See if that suggests anything that will help with the solution to the real ( and much more complex ) problem.

The economists have no hope of dealing whit real live irrational human behavior, so a simplified model will have to do.

[selenite.livejournal.com]

I tend to prefer "ad hoc approximation that roughly fits the real world" to "elegant model that precisely describes an abstract situation." But as a working engineer I have no need to get tenure.

[tmc4242.livejournal.com]

( That was me above. Logged out by mistake. )

I've resorted to engineering too. Being able to buy food and such is a nice thing after all...

But Physics is still Phun. :-)

[selenite.livejournal.com]

Ah. I suspected it came from an academic coming up on his tenure review, hence the snark.

[tmc4242.livejournal.com]

Nope. One degree short of needing to worry about tenure.

Tempting to go back for the Ph.D. once in a great while, but not tempting enough.

Physics solutions are almost always more fun than engineering solutions. Physicists don't usually worry about little things like marketability, cost, practicality. All those annoying little engineering constraints... ;-)

[selenite.livejournal.com]

I so do not want to go for a PhD. Too much work, too little reward, and most of the PhDs I've dealt with have been seriously handicapped in getting real work done.

It's not just physicists who ignore constraints. I was digging into the academic literature to find some help on a trajectory problem and got very fed up with all their simplifying assumptions. "Set thrust/drag = N" If I could do that I wouldn't be in the @#$%ing technical library . . .

[ernunnos.livejournal.com]

What's that joke with the punchline "Assume a perfectly spherical cow..."?

I don't think economists are in quite that much of a bind. Human behavior may be irrational, but that doesn't necessarily equate to unpredictable, or even complex. Sometimes irrationality is so predictable it's downright boring. All you really have to do to fix economics is change the assumption that humans act in their own best interest to account for interests more diverse than making money.

[tmc4242.livejournal.com]

>>What's that joke with the punchline "Assume a perfectly spherical cow..."?

Joke with closely related punchline:

What's a physicist's answer to "Why did the chicken cross the road ?"

Answer begins "First, we assume a spherical chicken..."


As far as human behaviour - I think it's a matter of scale. Large groups and you have a shot. It's called "herd behaviour" and it's kinda scary sometimes. Statistics work at that level. With individuals there are so many possible competing interests involved that prediction is very very difficult. Hence us physics types pick a different problem with fewer independent variables.

Owwww. That caused a painful thought. Eigenvectors for human behaviour.

Run away.

[ernunnos.livejournal.com]

Eigenvectors for human behaviour.

This is the second time I've seen that idea today. And that brought up a political test which is already applying it.

[tmc4242.livejournal.com]

Very very scary. I think the real problem with the idea is that there will prove to be WAY more than 60 basis vectors involved.

And it would be even more scary to find that 60 was enough.

Thanks for the link.

Rational

[p-o-u-n-c-e-r.livejournal.com]

There is also the problem that one's evaluation of each instance of a possession -- one book, for instance -- is colored by the evalution of it in context. Say you have 80 random issues of Spiderman -- and somebody offers to trade you two copies of Batman (which you haven't read, and have Neal Adams' art!) for one Spiderman. Well fine. But then suppose instead you had a serially sequential run of all Spiderman issues from say #114 to #194 and the offer -- if taken -- would break up the "set". No deal, right? Is that rational? I don't see exactly how. But real behavior, I'm fairly sure.

[selenite.livejournal.com]

Fascinating. Thanks for posting the link. I've been looking at various ways of graphing political opinions for a while now. That's a good addition to the collection.

I've got an intuitive feeling that 10-20 axes might be enough to describe just about any variation, but that's probably not useful for anything except academic research.