Hi Everyone. I am linking here a new paper with a bunch of new studies about knowledge and practical interests. The main claims are that (1) folk attributions of knowledge are sensitive to stakes, (2) the data presented supports Interest Relative Invariantism (IRI), but (3) it is not so good for contextualism about knowledge. This paper might be of special interest since some recent work by Buckwalter, Feltz & Zarpentine, and May, Sinnott-Armstrong, Hull and Zimmerman have been interpreted by many to give evidence against (1). I discuss these other studies but, unfortunately, not in a lot of detail (the paper is not really about criticizing those studies). I also have not yet incorporated Keith DeRose's recent criticism of those papers but it looks like there is some overlap (also, see Jonathan Weinberg's blog post from two days ago).
One main idea behind the study is the introduction of "evidence seeking" experiments. This type of probe gets around some possible problems with previous experiments (there are no knowledge attributions inside the vignettes). Basically these new tests work like this: I set up a pair of vignettes where a student wants to proofread his class paper for typos. In one case there is a lot at stake if he has a single typo, and in another case there is little at stake. I then ask subjects to fill in the blank (with a number) in the following prompt:
"How many times do you think the student has to proofread his paper before he knows that there are no typos? ____ times."
It turns out that the subjects given the high stakes vignettes write higher numbers than those who are given the lower stakes vignette. (differences are very much stat. sig.). This sort of result supports the idea that folk attributions of knowledge are sensitive to stakes (practical interests). This is just (1)
Some other highlights: (i) results replicated with different vignettes, (ii) differences persist when the vignette concerns a student who is not aware that there is a lot at stake "ignorant high stakes", (iii) differences persist even when the cases are juxtaposed, (iv) No ordering effects were found on the various combinations in the juxtaposed cases. (v) Intelligence (or reflection) does not appear to affect answers and (vi) a coding of subjects' written explanations to the juxtaposed responses reveal an appeal to a principle connecting knowledge and rationality (in a way that dovetails well with IRI).
This paper is still in draft form. I hope that some of you may find this paper of some interest. I would definitely appreciate feedback. --Angel