Tuesday, October 05, 2004

Jokes from the Amos Tversky Fan Club

A few weeks ago, I wrote a post about Lakoff, frames, heuristics (the emphasis was supposed to be on heuristics), and titled it "Karl Rove the Feminist Bankteller." I thought this was very clever at the time, but now I realize it was not only fairly unclever, but it was probably a bit too obscure. In my world, which is admittedly a very small one consisting of me and the other hard-core Amos Tversky fans, the reference, as the title of a post about cognitive heuristics, would be easy to recognize. Still, I don't know how many of you are hard-core Amos Tversky fans, or into behavioral economics in general. So, I decided that a quick explanation might be in order.

In a 1972 paper1, Tversky and Daniel Kahneman conducted an experiment using the following story:

Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in antinuclear demonstrations.

Participants were then asked to rank sentences by their probability. Among the sentences were the following:
(1) Linda is a bank teller.

(2) Linda is a bank teller and is active in the feminist movement.

As anyone familiar with probability can tell you, sentence (1)is more probably than sentence two, because the second involves a conjunction of (1) and another statement. However, participants almost always (86%) ranked (2) over (1), indicating that they thought it was more probable. This is the so-called "conjunction fallacy," and the standard explanation is that given her description, Linda is more representative of feminist banktellers than banktellers in general. When people refer to the conjunction fallacy, or the Kahneman and Tversky experiment, they often call the results the Linda the Feminist Bankteller problem/experiment.

So, I thought it would be clever to call Karl Rove a feminist bankteller in a post on cognitive heuristics (which is what the representative bias is), and how Republicans take advantage of them. I'm sure most of you caught that, but for those who don't get my twisted and unwitty sense of humor, now you see where the title came from.

1 Kahneman, D. and Tversky, A. (1972). Subjective probability: A judgment of representativeness. Cognitive Psychology, 3:430–454. See also Tversky, A. and Kahneman, D. (1983). Extension versus intuititve reasoning: The conjunction fallacy in probability judgment. Psychological Review, 90, 293-315 for a review of the various approaches to the Linda problem, and this paper for a shorter look at the research.

No comments: