The Inside/Outside Distinction and Epistemic Intuitions

Psychologists sometimes distinguish between two ways of getting evidence about a particular object. One way is to actually be in some kind of contact with the object itself, either by perceiving it directly or by observing something that has been caused by it (inside view). A second way is to learn certain facts about a more general category of which this object is a member (outside view).

Clearly, these two ways of getting evidence could turn out to be similar with regard to their reliability and with regard to the extent to which they allow a person to rule out alternative possibilities. Still, it seems that people have the intuition that these two kinds of evidence are deeply different from an epistemic point of view. That is, people feel that actually perceiving or examining an object is deeply different somehow from just making inferences about the object from more general information about a category.

Psychologist Ori Friedman and philosopher John Turri have a new paper in which they apply this distinction to questions about people’s epistemic intuitions and arrive at some very intriguing results.


For example, in one study, participants were randomly assigned to receive one of the following two vignettes:

[Outside view] Luigi purchased a plot of land in Springville. He has an electronic device that lists the likelihood that soil is terragenic. The device says there is only a 1-in-10-million chance that soil in Springfield is terragenic. So Luigi concludes that his soil is non-terragenic. And he is right: it is non-terragenic.

[Inside view] Luigi purchased a plot of land in Springville. He has an electronic device that tests the likelihood that soil is terragenic. The device says there is only a 1-in-10-million chance that Luigi’s soil is terragenic. So Luigi concludes that his soil is non-terragenic. And he is right: it is non-terragenic.

Though the probability of error is exactly the same in the two cases, the difference between outside and inside views led to different judgments as to whether the epistemic subject had knowledge. Specifically, a substantially greater percentage of participants said that Luigi knew the soil was non-terragenic in the inside view (48%) than in the outside view (27%).

This finding seems like it has a great deal of potential to illuminate questions about the lottery paradox and perhaps about a number of other important issues in epistemology. I would definitely encourage you to read the complete paper, but please feel free to write in with any thoughts even if you haven’t read it. I would love to hear any suggestions people might have about how to make sense of these results.


Comments

The Inside/Outside Distinction and Epistemic Intuitions — 1 Comment

  1. Thanks for this concise and extremely informative post reviewing our work on this, Josh!

    I wanted to mention quickly that Ori and I had done some earlier work on knowledge judgments in lottery cases, part of which covered an intriguing hypothesis by Gilbert Harman (“Knowledge, inference, and explanation,” American Philosophical Quarterly 5, 164-73). The hypothesis pertained to why people deny knowledge in a basic lottery cases where the person reasons, “There’s only a one-in-a-million chance that this ticket is a winner, so it’s a loser.” Harman proposed that we don’t attribute knowledge to this person because our “natural non-philosophical” view of knowledge involves an explanatory connection, but “no explanation is involved” in this person’s reasoning. (At the time, we called this an “unanchored statistical inference.”)

    Ori and I tested this view with a case that we thought involved an explanatory connection: the mafia rigged a lottery so that they were overwhelmingly likely to win (Experiment 3 of “Winners and losers in the folk epistemology of lotteries,” in J. Beebe (Ed.), Advances in experimental epistemology). More specifically, we thought that the mafia’s rigging was obviously a big part of the explanation of why the agent’s ticket lost. But people didn’t attribute knowledge in this case, even when the agent knew that the odds were long because of the mafia’s rigging. We took this as evidence that Harman’s account was not psychologically correct. Still, this surprised us somewhat because Harman’s account seemed to cohere nicely with subsequent findings in cognitive psychology (including findings by Kahneman and Tversky and Icek Ajzen).

    However, the findings you reviewed here were the result of later experiments that probably better operationalize the relevant factor implicated in the psychological research. Accordingly, it seems like a distinct possibility that Harman’s view captures an important part of the psychological story, but our earlier attempt to detect this failed. Perhaps the distinction between “unanchored/anchored” or “outside/inside” statistical judgments does help explain the intuitive pattern of knowledge attributions in lottery cases.

    If so, then it’s a testament to the acuity of Harman’s reflections on this. He proposed something in the ballpark before any of the psychological work (at least that I’m aware of) had been done!

Leave a Reply

Your email address will not be published. Required fields are marked *