Hardly Rational

One would assume that individuals would endeavor to be rational and logical when making decisions, especially when the lives and fates of others lie in the balance. However, surprise surprise, this is not the case. Now you might be saying, "obviously Komma, this is nothing new, emotion tinges our decision making process!" and you would be right -- but there is more.

I am in the process of reading this article by Dan Simon and I just had to post. The idea is this: when making a decision humans do not (necessarily) follow the unidirectional flow of logical processes, from premises to conclusions etc; instead the process is *bidirectional* "premises and facts both determine conclusions and are affected by them in return".

Lets look at an example. In one experiment individuals were presented with 12 vignettes and a statement or two that could be inferred with the vignette. After reading the vignettes, the participants were asked to rate how well they agree with the 12 inferences.

Next, the participants were asked to play the role of a judge in a court case. They were then presented with 12 arguments, 6 from the prosecution and 6 from the defense -- the rub is, the vignettes and the arguments were _virtually_ identical. The facts of the case are complex, and strong arguments could be made for both the prosecution and the defense. After deciding a verdict, the participants were asked to rate how well they agreed with the arguments from each side.

Now logically, you would expect that the individuals would rate the arguments the same way they rated the vignettes -- however they didn't! There was a statistically significant difference in how a participant rated the same vignette/story from before/after the decision making process! For instance, if a participant judged for the prosecution, then the arguments for the prosecution would be rated higher than arguments for the defense, even though both vignettes would be rated about the same before hand.

Even more fun -- when participants were asked to recall their original ratings they incorrectly recalled ratings that matched their post-decision ratings. In all cases, deciding for the prosecution or the defense, the participants were highly confident of their decisions.

To make a long story short, Simon argues that for decision to be made "effectively and comfortably", individual must possess a "coherent" mental model -- one which there is strong support for the decision and weak support for the opposite decision. And when this isn't in the case, us humans go about and make it so :).

Now this paper is focused on legal decision making, but its clear to me that these types of effects seem to appear in all areas of society. I've always wondered why nearly every issue quickly becomes polarized (democrats vs. republicans, pro vs. anti abortion, pro vs. anti war; etc) and perhaps a coherence based theory of decision making can account for some of this tendency. As we make decisions, we start reducing our belief in the opposite's arguments, naturally leading us to a polarized state.

I don't know how to stop this, although Simon claims to have some ideas in mind for rectifying this situation in the context of juries, and perhaps we can't. Maybe we will just have to survive with a polarized society.