The Heuristic in the Bias

In the art of annoying people with science, nothing is as effective as pointing out cognitive biases. Bringing out the Confirmation Bias in particular is unlikely to endear you to friends and colleges. But you usually get away with the point – there is almost always more research to be done – unless someone figures out that you already came equipped with the idea that your opponent would use the confirmation bias, and then choose the evidence that seemed to confirm that idea. The discussion that follows can take up a substantial part of the seminar, and effectively hide the fact that you haven’t done the required reading. Do try this at home.

To be caught exhibiting any kind of bias is usually held to be a bad thing, not only in science. But, as Kahneman, Tversky and Slovic (among others) points out: biases are heuristics. The are usually very useful indeed. It is in the nature of a bias/heuristics that it may lead us astray, but practically any epistemic strategy or habit is bound to lead astray in some cases. We usually solve this by having other strategies to keep the first in check. And so on. Peer-review is one such strategy, not fool-proof. Democracy might spring to mind, to.

The term ”confirmation bias” was coined, I believe, by the psychologist Peter Wason, but the notion is way older than that. My favorite wording comes from Laurence Sterne’s eternally ahead of its, and any, time novel ”Tristram Shandy”:

It is the nature of a hypothesis, when once a man has conceived it, that it assimilates everything to itself, as proper nourishment; and, from the first moment of your begetting it, it generally grows the stronger by everything you see, hear, read, or understand. This is of great use.

”This is of great use”. I used this paragraph as one of many mottoes for the second part of my dissertation. I certainly did pick the evidence that seemed to confirm my theory (that pleasure and value are very closely related indeed). But the point is that I did not conceive of a theory, derive the consequences and then do the researach. The project was much more preliminary than that (still is). I wanted to find out what, if anything, hedonism was true about. What are the facts, and can they be put together in a coherent, sense-making way to form a theory, still recognisable as ”hedonism”?  At this stage, at least, confirmation bias is a very useful strategy.