« Overcoming bias - what is it good for? | Main | "Inductive Bias" »

April 07, 2007


"One sign that science is not all bogus is that it enables us to do things, like go the moon."

I was wondering if engineers were less biased than other scientific types? They deal with the practical and concrete all day long, and they see their ideas either succeed or fail before their eyes--such as landing on the moon or exploding on the launch pad. Unlike social or psychological researchers who have the option of clinging to their theories through thick and thin, engineers are trained to identify and abandon incorrect ideas as quickly as possible.

I studied engineering as an undergrad, and I believe it taught a form of objectivism. Or perhaps it simply revealed it.

I was also a fighter pilot for a number of years. Clinging to incorrect assessments about one's abilities, strenghts, weaknesses, or about others' could get one killed pretty quickly even in peacetime. Or perhaps overconfidence is necessary even to begin such a dangerous career. I think it's a wonder I'm still alive after all I've read here on your blog.

Studies have shown chess to teach some forms of debiasing relating to looking for disconfirmation rather than confirmation.

As for public proof

Should be want a public proof? Would that not attract lots of people who are more interested in signaling than actually overcoming bias?

People should be aware of the advantages that de-biasing can bring, but we should let them know of it - quietly.

This is in response to Brian's comment, above.

Some years ago, I had the misfortune of being a member of a faculty senate, which gave me regular opportunities to hear highly intelligent people saying stupid things. At the university at which I then taught, some faculty senate members were elected on a university-wide basis, so one had to choose candidates from a group of people one didn't know and couldn't learn much about. One of my colleagues voted strictly according to department affiliation, using this system, which seemed good to me: engineers and business-school people, yes. Scientists, yes, except for physicists. Veterinarians, yes. Economists, yes. Everybody else, no. (We were lawyers, so we didn't have to make a decision about them as a group, as we knew the candidates, and the medical school wasn't part of the process). Looking back on the list, it does boil down roughly to a distinction between fields in which ability and lack of it lead to real-world consequences. I'm not at all sure how that would apply to academic lawyers. Most of us tend to be litigators (I'm not), and that's a profession much of like confidence tricksters.

I agree for the most part with Brian and Alan, but on the other hand Razib's Gene Expression post Nerds Are Nuts also rings true.

"One sign that science is not all bogus is that it enables us to do things, like go the moon."

Does this mean that "One sign that the humanities are at least partially bogus is that they don't really enable us to do things, like go the moon."?

Cynical prof, I don't think the humanities are meant to enable us to go to the moon. They're meant to output good literature. Expecting otherwise is like castigating physicists who can't play bongo the way Feynman did. So, the proper reason to label postmodernist poseurs as "bogus" is that they cannot output good literature, not that they are incapable of producing more "practical" benefits.

On public proof...

1) "Order of Bayescraft" not likely to be seen as anything other than a self-help cult, like Scientology or Landmark.

2) A single spectacularly huge public success will be unconvincing and considered just a normal scientific breakthrough or luck.

3) If there existed techniques that could be taught easily, would they not already have been discovered? And what about the more-harm-than-good landmines that exist everywhere?

To call a de-biasing program a success, one or more individuals would have to show repeated scientific breakthroughs and be able to document how in each case known biases got on the way of the breakthrough, and how the de-biaser(s) got around these biases and saw the truth more clearly. An dojo would probably help, but "systematically more successful" would not be a sufficient criteria to distinguish it from a self-help school. Systematic breakthrough would.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30