Sunday, March 05, 2006

David Buller: Adapting Minds

David Buller's 's Adapting Minds was pretty hard to get through, but the most important ideas seemed to have been covered in the first half. I only got about half-way through before skimming and skipping my way through the rest. I got the impression from the reading group's discussion that more people than usual hadn't finished the book.

Buller's goal is to show that the most well known findings of the Evolutionary Psychology movement (which intends to show that there are facets of human behavior that are evolved and universal) result from misinterpretation of the experimental results, or from improper controls in the experiments. In this, it resembles Judith Miller's The Nurture Assumption, which showed that most of the research intended to determine what traits are genetic based on comparisons of twins raised together or apart misunderstood the nature of the correlations they were looking for, and so used the wrong attributes as controls. Miller's book succeeded in demolishing those results, and put forward alternate explanations along with supporting evidence to show that her characterization worked better than the standard interpretation.

Buller succeeds in showing some flaws in the widely touted results, but doesn't present plausible alternative explanations. In the end, the reader is left with the feeling that EP's results have been nitpicked to death, but has no solid basis for guessing which way the experiments would come out if run again with better controls. On the other hand, since Buller started out by saying that he didn't want to attack the idea that there's an evolved basis for behavior, Buller wouldn't count this as a failure. All he wanted to do was shoot down the specific results of the EP proponents, and not the idea of heritable behavioral traits, or the program of research intended to establish the details.

One particular example that Buller beats to death is the Wason Test, the results of which have been widely interpreted to show that people have an innate cheater detection facility. Buller relies on the argument that the form of the two questions that are asked (if even then red, versus if drinking then must be of age) are different kinds of logical propositions. The most enlightening version of the explanation (he seems to explain at least a dozen times, with references to different experts and scholarly lines of discussion) is that the two cases that supposedly violate the given rules differ structurally. In one case, the consequent is negated ("the card shows red"), in the other, the rule's requirement is not complied with ("the patron is under age").

In the end, it seems clear that the EP crowd's explanation doesn't carry all the force it is commonly purported to. At the same time, it's also clear from introspecting on our difference in performance on the two tasks that we solve one by general reasoning, and the other using a built-in problem solver (it's fast, and the process is introspectively atomic). Even if it is the form of the question that affects the results rather than the subject matter of the question, there's still a difference in performance that matches the EPers' description of a behavioral facility that is inate, specific to humans, and universal among healthy individuals. It's an inherited behavioral trait. The boundaries aren't where the standard explanation puts them, put it's still there.

Filed in:

2 comments:

Peter McCluskey said...

You say:

Buller succeeds in showing some flaws in the widely touted results, but doesn't present plausible alternative explanations.

I disagree. If you had read the second half of the book more carefully, you would have seen clearly plausible alternatives. But the ways in which those alternative differ from the standard EP explanations aren't very exciting.

Also, the author of The Nurture Assumption is Harris, not Miller.

Nancy said...

Buller's goal is to show that the most well known findings of the Evolutionary Psychology movement (which intends to show that there are facets of human behavior that are evolved and universal) result from misinterpretation of the experimental results, or from improper controls in the experiments.

You don't describe Buller's intention accurately. He differentiates evolutionary psychology, which he thinks is a good idea, from Evolutionary Psychology, the highly flawed set of beliefs championed by Tooby, Cosmides, Pinker, Buss, Daly and Wilson.

He wrote a good article here:
http://www.skeptic.com/the_magazine/featured_articles/v12n01_sex_jealousy.html