Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
The Cost of Smarts (nytimes.com)
15 points by edw519 on May 8, 2008 | hide | past | favorite | 11 comments


Already debunked on news.yc here:

http://news.ycombinator.com/item?id=183339


So, a few people, speculating about what the paper might say, constitutes a debunking?

Scientists are not all idiots. Parsimony suggests that -- somewhere during the months of planning and thought that goes into the average publication -- they probably thought about the questions that one might raise, 30 seconds after reading a newspaper summary of the work.


A) There is no way to control for side effects of single trait selection, so it's impossible that they did.

B) If they really had solved one of the biggest problems in science then that would have been the headline, not that learning reduces fitness. In fact they probably would have been handed Nobels on the spot.

C) I read the scientific paper and they never even take the issue into consideration. It's literally not even mentioned in the methodology or the discussion, and there's no citations of prior research indicating that fruit flies aren't susceptible to this problem in the introduction.

The fact is that this "research" is a joke and the NYT got snowed plain and simple. If you don't believe me, just Google for Kawecki + quinine and read the original paper.

EDIT: The claim made by the original research article is that selecting for increased learning ability results in lower fitness. The claim being made by the NYT is that increasing learning ability results in lower fitness. Those are two completely different claims. The fact that people are still downmodding me only shows that news.yc is slowly assimilating mainstream culture, a culture that glorifies ignorance and scientific illiteracy.


Dude...people are down-modding you because you're being a jerk, not because of some misunderstood intricacy of your brilliant scientific argument.

First off, a quick search tells me that Kawecki has been publishing on this topic for years:

http://scholar.google.com/scholar?q=kawecki%20quinine%20lear...

He's had numerous publications in PNAS, Science, Evolution, and other top-tier journals, building this hypothesis step-by-step. So perhaps you should back off your rhetoric about the research being a "joke" just a tad, okay? (At the least, perhaps you should go read one of his other papers before you decide that the man is a fraud.)

Now, to directly address your argument: "intelligence" is not a single trait. It's a complex, emergent phenomena, presumably involving many genes, in many different genetic loci. It's therefore highly unlikely that selection for the abstract trait of "intelligence" is also going to select for a decreased lifespan (another complex trait).

Thus, even if you're right, and even if the whole result is attributable to single-trait selection for "intelligence", it's still an interesting result. You've done nothing do "debunk" the interesting premise of the article -- that selection for intelligence appears to decrease overall fitness in flies.

Your obsession with single-trait selection seems to rest on the notion that if the result could be attributable to linkage, then it doesn't count. That's silly. We can debate the mechanism of action, but there appears to be a very real trade-off between learning ability and survival in this particular species. Arguing that it might only be a consequence of selection is to completely miss the point.

(Aside: I think it's interesting that you've now latched on to a subtle difference in wording between the NYT article and the claims made by the researcher. The NYT may have (debatably) been exaggerating the claims of the researchers, but your posts have all jumped to the conclusion that the original research was a "joke". Bad form, man.)


Fair enough, I apologize for being a jerk. That said, if the NYT had really said that "selection for intelligence appears to decrease overall fitness in flies" then I would have had no issue with the article. However, what the article actually said was that, "Learning also turns out to have dangerous side effects that make its evolution even more puzzling." That's not a minor error or a small exaggeration, it's outright fraud. And what's more, the entirety of this new article is based off the false premise of the original. That's why I'm upset.

Anyways, my apologies to the researcher. My issue is mainly with the NYT.


don't back down or apologize because some jerk yelled at you "for being a jerk". :(

you made good points, which does not constitute being a jerk. he made invalid points about how the guy has a reputation and history that don't address whether the particular ideas at issue are true; plus, he literally called you a jerk, which is name calling.


No, he's right, I was being a jerk.

I still stand by my argument though.


Did you delete text from your post, or is it all visible now?


if you think carefully about all the possible objections, then in certain fields, like social "sciences", most studies are no good.

some people do this with integrity and some don't. only the ones who don't actually publish in those fields. so there is a major selection effect. it's not true that all scientists are idiots, but most published scientists in certain fields seem to be. if you read a selection of studies then you can confirm this.


"In certain fields, like social "sciences", most studies are no good."

Saying the average social science paper is no good is like saying the average blog post is no good. Even if it's true, it's a completely meaningless statement because there are enough really good social science papers to keep you busy for the rest of your life.


it's not meaningless b/c i'm talking about studies published in serious journals, of which they are a limited number, all of which are supposed to have quality standards, peer review, etc

but there are endemic problems, including for example the pitifully low 95% confidence target.

and i think there are hardly any good ones. want to link a few of the very best, in your opinion?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: