Tuesday, August 18, 2009

Modeling and Rationality

This is kind of a complicated story, but it is a good lesson in how mathematical or economic modeling can help you realize that people are not really crazy.

I read this article, and my initial reaction was that the people surveyed were insane:

"Unknown to the participants, everyone was offered a fictitious candidate partner who had been tailored to match their interests exactly. The photograph of "Mr Right" was the same for all women participants, as was that of the ideal women presented to the men. Half the participants were told their ideal mate was single, and the other half that he or she was already in a romantic relationship.

"Everything was the same across all participants, except whether their ideal mate was already attached or not," says Burkley.

The most striking result was in the responses of single women. Offered a single man, 59 per cent were interested in pursuing a relationship. But when he was attached, 90 per cent said they were up for the chase.

Men were keenest on pursuing new mates, but weren't bothered whether their target was already attached or not. Attached women showed least interest and were slightly more drawn to single men."

My thoughts were as follows:  How can you possibly expect that anything good will come of getting involved with someone you have convinced to leave a relationship?  That person will have proven him or herself to be unreliable and untrustworthy.  Why would you want someone who will most likely leave you whenever someone better comes along?

But the people were assuming that the people in a relationship were of higher quality.  This makes sense, if you assume that other people want the same thing as you and that all of the singles have been tested by someone else and found wanting.

Ideally, you would want someone who is both honorable and of high quality, but those people are assumed to be unattainable.  So they had a choice of high-quality people with no honor, or low-quality people with unknown honor.  I actually drew out several graphs, with different assumptions about statistical distributions of honor and quality, before I came to the conclusion that it was almost always rational, under those assumptions, to try to poach from the taken people rather than take a chance on the singles.

Now, I will never use this strategy, for several reasons.  First, I care a lot more about honor, both in me and other people, than the average person.  Second, I believe that relationship happiness depends mainly on how well you match each other's personal references, rather than an absolute scale of quality.  And third, I believe that many people are single not because they have been repeatedly rejected, but because they have not felt the need to chase a relationship and/or have not yet met the right person. 

2 comments:

e said...

AS you say, honorableness is not necessarily consistently sought for by everyone even within one culture. What is considered 'honorable' is not always consistent, either. Genetic and social quality is more likely the common factor. (p.s., where is it assumed such people with both are 'unattainable'?)

If perceived 'Quality' is or has been the biggest factor in choosing mates, do you think choosing in favor of honor can also be a rational action (in same terms of genetic success)?

Alleged Wisdom said...

The assumption was that the taken people were all above a certain threshold of quality, and that all the single people were below it. Anyone who was both honorable and above that quality threshold would be taken, and unwilling to leave their existing mate.

With any animal where the children require a lot of care, it is very important to choose someone who will stay around and help care for the kids until they are grown up. A mate who has less 'genetic fitness' but who stays around will give you more surviving offspring than one who leaves you to take care of the kids by yourself, even if those kids have great genes.