g comments on Human Evil and Muddled Thinking - Less Wrong

40 Post author: Eliezer_Yudkowsky 13 September 2007 11:43PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (138)

Sort By: Old

You are viewing a single comment's thread.

Comment author: g 17 September 2007 09:26:00AM 0 points [-]

Tom and Douglas, I'd take "bias" to mean something like "common feature of human thinking that gives us predictably wrong answers". Some biases might (for any given person) not be worth correcting, if correcting them requires sustained effort and similar effort could be more effective if deployed to other ends. I don't see why whether something's a "bias" should depend on whether it's the result of a heuristic operating out of the regime it was designed for, or on whether we can correct it.

I have the *feeling* that the term "bias" gets overused here (possible explanations if I'm right: mission creep plus rationalization, and the satisfaction of labelling others as "biased"), but I don't actually have concrete examples in mind. Perhaps I'm suffering from overidentification of overidentification of bias bias bias.

(Remark: this construction generalizes to produce noun phrases ending with an arbitrary number of instances of "bias". Similar but nicer construction, not due to me: "oysters oysters oysters split split split" is an English sentence, at least notionally, for all values of 3.)