Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Overcoming bias guy meme | quickmeme

4 saliency 13 March 2013 04:56PM
Comment author: Douglas_Knight 03 November 2012 01:55:41PM *  5 points [-]

Why are you in favor of breast-feeding at all?
It is correlated with socio-economic status, and thus with all good things. But every correlation I have seen corrected for parental SES goes away, including IQ. Decades of randomized controlled studies demonstrate no effects. On the positive side, I must admit one cluster-randomized (n=31) study of IQ in Belarus, but I am not impressed.

Comment author: saliency 03 November 2012 08:02:10PM 1 point [-]

Because when in doubt go with convention.

I think there is a lot we don't understand.

Now if my wife found it bothersome perhaps we would not follow convention, but so far it she likes doing it. From a fathers perspective it is vastly superior due to the ability to leave a bottle out of the refrigerator for 6 hours instead of 1.

Less Wrong Parents

11 saliency 03 November 2012 04:59AM

Less Wrong Parents
https://groups.google.com/forum/?hl=en&fromgroups=#!forum/less-wrong-parents

Recently the NYC LW/OB community had two babies and is expecting a third.
I created a google group as a way of sharing information, primarily thinking of the NYC community.

I posted my pre-baby purchase list and William Eden posted an extensive list of books on early parenting.

William suggested opening up the group so as to get insight from the larger LW community on parenting.
I think this is *probably* a good idea.  Google groups are simple to set up but have limits.
For this reason I request that if you are going to have an extensive debate on a subject you create a new thread (aka: get a room)

The primary objective is to lower the cost of obtaining information on parenting.
I believe this overall goal to be more important then any particular "truth".

My hope is that this will primarily serve as a place for people to ask parenting question and post guides.
Perhaps if enough guides are posted they can eventually be consolidated into a wiki.

Comment author: saliency 03 November 2012 01:28:44AM 13 points [-]

If you told me I had a 35 percent chance of winning a million dollars tomorrow, I’d try to sell you my chance for 349 thousand dollars.

Comment author: shminux 02 November 2012 10:18:01PM *  12 points [-]

Google says:

breastfeeding was found to raise intelligence an average of nearly 7 IQ points if the children had a particular version of a gene called FADS2.

Later:

Ninety percent of the children in the two study groups had at least one copy of the "C" version of FADS2, which yielded higher IQ if they were breast-fed. The other 10 percent, with only the "G" versions of the gene, showed no IQ advantage or disadvantage from breastfeeding.
The gene was singled out for the researchers' attention because it produces an enzyme that helps convert dietary fatty acids into the polyunsaturated fatty acids DHA (docosahexaenoic acid) and AA (arachidonic acid) that have been shown to accumulate in the human brain during the first months after birth.
Since the first findings about breastfeeding and IQ appeared a decade ago, many formula makers have added DHA and AA fatty acids to their products.

Basically, it's one of those cases where doing it is probably better than not, but not at all costs. If you are constantly stressed because your baby prefers the bottle, or you get repeated breast infections, or you don't produce enough milk, it's time to direct your efforts elsewhere.

In response to comment by shminux on Less Wrong Parents
Comment author: saliency 03 November 2012 01:12:49AM 0 points [-]

Very interesting, thanks.

In response to comment by shminux on Less Wrong Parents
Comment author: Vaniver 02 November 2012 08:28:59PM 5 points [-]

It is generally well known that the eager new parents tend to overthink everything and overdo everything they can think of (stroller type? formula vs breast milk? daycare amenities? nursery wall color?).

This list strikes me as a good example of why a LW listserv / shared research about this is useful. Yeah, stroller type won't make much of a difference, and nursery wall color won't either. About half of important life outcomes are heredity, and you've missed the opportunity to change that.

But breastfeeding? We're talking ~8 points of IQ, here. Breastfeed your children, and do it for as long as you can stand.

In response to comment by Vaniver on Less Wrong Parents
Comment author: saliency 02 November 2012 09:00:24PM 5 points [-]

I am pro-breast feeding but skeptical of the IQ claim. Can you link the study?

My guess is they compare the avg IQ to the average of a selection of people who breast feed.
I would expect this to be subject to selection bias.

In response to comment by shminux on Less Wrong Parents
Comment author: DaFranker 02 November 2012 06:42:48PM *  7 points [-]

Is this from real data?

I would think that the behavior of parents has a massive impact on the way the children grow up (but indeed, not the material stuff that so many parents fuss so much over), considering how strong of a correlation there is between the parents' belief systems and behaviors and their children's.

I'm not much compared to even a small survey, but from my small sample I've noticed a possible strong correlation between the way parents respond to questions / handle "problematic" behavior / do anything to "educate" their children and the intelligence, rational behavior and open-mindedness of the children later in life.

The most salient example (but not the most statistically significant) is that everyone I talked to about this who were on the higher end of the intelligence scale had clear memory of their parents responding "I don't know, let's find out" to their curiosity when they were a child, while everyone else I talked to had no such memory.

I think looking into actual pedagogical research results and how to best behave towards children would probably be very high expected utility / value of information if maximizing your child's chances of not being stupid is something you care about.

At the very least, a parent can affect the environmental factors that the book mentioned in the parent post mentions (I haven't read the book, only the abstract) by carefully selecting a good initial environment with these things in mind in the first place.

Obviously also worth looking into is alternative forms of education. Public schools are far from optimal both for social and intellectual development.

If any of this is of interest, I can try to help with some research on it.

Comment author: saliency 02 November 2012 06:56:53PM 5 points [-]

The school of thought shminux represents, though not popular in the main stream, is one I ascribe to. Bryan Caplan and a few others have books on the subject.

Shminux, this though is exactly why I see this group to be of value. I don't want to spend a lot of time doing research. I want to examine three peoples strategies and trust that I can blindly go with the suggestions, or at least have a strong starting point.

Comment author: saliency 12 June 2012 07:19:39PM 0 points [-]

Interesting read, thanks for the post.

Comment author: bogus 12 June 2012 01:43:18PM *  0 points [-]

Do you think most people pay more taxes then they need to because they are signaling?

No; I think they are trading off compliance costs vs. the risk of paying more taxes than they owe. But it's not clear that the price discrimination story is applicable here.

To the paper being convincing. Be specific.

Basically, the problem with Gabaix--Laibson is that its "myopic" consumers are persistently biased in addition to having bounded rationality. They persistently expect to be charged less e.g. for the hotel stay than they actually are. A boundedly rational consumer would expect to be overcharged for some addons, even if she dosn't know in advance what the marked-up addons will be or whether she can avoid the surplus charge (unlike Gabaix--Laibson's sophisticated consumers). This may or may not change her response to efforts at more "transparent" pricing.

That people who care "sophisticated" will prefer systems in which they can obtain an advantage.

Yes, this is fairly obvious. But this also implies that naïve folks will avoid these same systems. In general, it will pay for sophisticated folks to credibly refrain from using such systems, unless the system provides further benefits (say, effective price discrimination).

Does your car mechanic or IT guy tell you the exact truth or do they pad things just a little. Do teammates working on a project ever slack but make it look like they are doing work?

This is an asymmetrical information problem. People expect that a car mechanic will pad costs if she can get away with it; so they try to establish norms under which more info is provided, or else the practice is deterred directly.

Comment author: saliency 12 June 2012 02:25:10PM 0 points [-]

"compliance costs vs. the risk of paying more taxes" -- This is why I use health savings accounts and commuter plans as an example.

"myopic" consumers -- There really are no individual consumers there are transactions. Myopic transactions perhaps would be a better description. On aggregate we have lots of myopic transactions. (bounded rationality) To answer you question -- I agree with you second part on myopic's but don't see how it is a problem for G&L. Sophisticateds are the ones driving the evolution of the system.

Comment author: bogus 12 June 2012 01:01:25AM *  1 point [-]

Part of the problem is that while non-transparent pricing may or may not be self-sustaining (Gabaix--Laibson is not wholly convincing on this point), price discrimination is unambiguously so because it serves a useful economic function, namely defraying intra-marginal costs (including fixed costs and normal profit) in the most effective and least burdensome way. Cutting coupons is a comparatively efficient way of signaling that one is a highly price-sensitive customer and should not bear a significant share of these intra-marginal costs. The JC Penney strategy was affected by a similar issue in that they got rid of most of their discount sales, which also attracted highly price-sensitive folks.

Comment author: saliency 12 June 2012 04:42:38AM 0 points [-]

First thank you for the thoughtful response. This is more what I was hoping for when I posted... I don't agree with you signaling story but it is something I would not have considered.

"price discrimination" I don't think this is at all a story of signaling. I think it is a story of information/time costs.

My stories: If my wife picks up the circular at the store entrence and tells me that if I rip out this page an hand it to the cashier I will save a buck I do so. Most people don't do their health savings accounts or mail into NYC to have their metro cards mailed to them so they can save a few bucks by deducting the cost. Do you think most people pay more taxes then they need to because they are signaling? Tell me your signaling story.

To the paper being convincing. Be specific. I bet that your story will involve agents who can not defect or some external structure which alters incentives. My story is very close to that of the paper. That people who care "sophisticated" will prefer systems in which they can obtain an advantage.

I'm expanding the strict definition of price discrimination by including taxes ect but believe they are the same. By doing so I think it can be seen that price discrimination is a supply and demand side. In addition I would argue that because I am talking about systems that it is an emergent phenomena. Agents within a system shroud. Does your car mechanic or IT guy tell you the exact truth or do they pad things just a little. Do teammates working on a project ever slack but make it look like they are doing work?

View more: Next