Linda you sly fox

Lately, LI has been a little baggy and disorganized. Unfortunately, I foresee this post straying into chaos too – I can feel its edges, even now, being pulled towards some strange attractors -- but I will try to be a bit more disciplined.

See, I want to write about two different things. I want to write about my web pal Paul’s post on the daemonic interpretation of the person – which rings some bells with me. And I also want to write about Gerd Gigerenzer’s essay in the new issue of Social Theory, I think, therefore I err. And at some point I wanted to use Schopenhauer’s image of the Veil of Maya to talk about traffic fatalities.

Uhh, right. Okay. Three things.

As fans of prospect theory know, Gigerenzer plays Moriarty to Kahneman and Tversky’s Holmes and Watson. Prospect theory – which takes the datum from psychological testing to understand patterns in how people make decisions according to their perspective of the probabilities involved in adopting a course of behavior – has busily revamped the way economists think of the rational agent. Kahneman and Tversky found that certain patterns of logical error occur across groups. For instance, given a constant probability of a course of action, one can manipulate responses to that course by framing it in terms of gain or loss. K and T developed what is called the Asian disease problem. Using students and professors as their pool of respondents, they posed this problem:

Imagine that the US is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. Two alternative programs to combat the disease have been proposed. Assume that the exact scientific estimates of the consequences of the programs are as follows:

If Program A is adopted, 200 people will be saved.

If Program B is adopted, there is 1/3 probability that 600 people will be saved, and 2/3 probability that no people will be saved.

72 percent chose A, 28 B. Then they proposed this problem:

Problem 2
If Program C is adopted 400 people will die.
If Program D is adopted there is 1/3 probability that nobody will die, and 2/3 probability that 600 people will die.
Which of the two programs would you favor?”
22 percent went for C, and 78 percent went for D.

The first question was framed in such a way that it brought out risk averseness: “the prospect of certainly saving 200 lives is more attractive than a risky prospect of equal expected value, that is, a one-in-three chance of saving 600 lives.” The second question brought out risk taking: “the certain death of 400 people is less acceptable than the two ­in­ three chance that 600 will die.”

Gigerenzer’s essay begins by showing that the paradigm within which Kahneman and Tversky are working is one that assumes that the brain is a logic machine. This goes back, according to G., to Piaget’s work. Piaget showed that children, as they get older, get better at answering questions that are, basically, about sets. For instance, children are shown a picture with ten flowers, of which five are daisies. They are asked if there are more daisies or more flowers in the picture. Eventually, by the age of eight or nine, they click to the fact that daisies are a subset of flowers, and thus, naturally, there are more flowers. But K. and T., those devils, upset this neat pattern by transposing the terms into probability terms with the famous Linda problem. Linda is 31 years old. Linda was a philosophy major. Linda is outspoken. Now, which one of the two is more probable? A. Linda is a bank teller. B. Linda is a bank teller and is active in the feminist movement? People match Linda’s characteristics to b., and so choose b, even though – by the laws of probability – the conjunction of the probability of two states is less than their separate probabilities.

(The conjunction fallacy, by the way, was rife during the early days of the Iraq occupation, as LI liked to point out at the time. But I’m not going off in that direction today.)

So: what? Tomorrow I will write about Gigerenzer’s problem with the Linda problem.