Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Psy-Kosh comments on Categorizing Has Consequences - Less Wrong

25 Post author: Eliezer_Yudkowsky 19 February 2008 01:40AM

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

Comments (13)

Sort By: Old

You are viewing a single comment's thread.

Comment author: Psy-Kosh 19 February 2008 02:06:15AM 0 points [-]

*ponders* Just a thought for a possilbe start to a reasonable protocol for such an experiment:

Maybe something like having the subjects themselves create the categories instead of giving it to them? What I mean is this: You'd have, some set of plants (or whatever objects) and a group of test subjects. Then have each subject independantly produce a category. That is, tell them to select X plants at random and give a made up name to that group.

Several possible ways to deal with those that have been assigned to multiple categories. One way may be to remove each such entity. Other would simply be to give those a name somehow constructed from the names of the categories they're made of. Alternately, invent a new artificial category, like the "2 cat" category, the "3 cat" category, etc...

(This is just a bit if a fuzzy idea, not a full experimental protocol or anything, just a notion of a starting point for one)