Blog Coordinator

Knobe's X-Phi Page

X-Phi Grad Programs

« Joshua Alexander's Experimental Philosophy | Main | A Profession-Wide Invitation »



Feed You can follow this conversation by subscribing to the comment feed for this post.

James Beebe

I’m unclear about the ‘expectation’ part of the experiment. Was each participant asked ‘What do you think Carl will/should do?’ or was each participant asked ‘What do you think Carl will do?’ and ‘What do you think Carl should do?’ Since descriptive and normative expectations are different things, it seems they cannot be combined into a single question. On the other hand, if participants were asked separate questions, I don’t understand how participant responses to this part of the experiment map onto the two graphs.

James Beebe

I also have a question of clarification about the 'attribution' part of the experiment. Each participant was given two forced choice questions? One about whether Carl intentionally helped other and another about whether he intentionally made his retirement more comfortable? Was the order between these questions varied and did it matter?

James Beebe

I see from the y axes that participants must have been given a Likert scale to use rather than a forced choice format.

James Beebe

Regarding the 'neither' condition: you suggest it might have something to do with the fact that this condition is structurally dissimilar from the other three. Here's another factor to think about: It struck me that the neither condition was similar to the both condition in a certain fashion. While neither norm in the neither condition is given your special salience treatment, the story set-up either implicitly or indirectly puts both norms on the table (because the relevant norms automatically come to mind when the two options are described) or at least makes the process of deliberation itself salient, since a choice between two reasonable options must be made. If the latter is correct, this will cue participants' intuitive grasp on dual-process theory directly, without needing to go through any thinking about norm violations. If the former is correct, then participants will be implicitly considering the norms associated with both options, and so the case is somewhat like the both condition.

Mark Alfano

James: Thanks for the questions. This is what comes of directly posting a script on the web....

1) On the expectations questions: participants were asked both "What do you think Carl will do?" and "What do you think Carl should do?" The possible answers were "invest" and "donate," so it was a binary forced choice. Almost everyone said he both will and should invest, and the answers to these questions were 100% correlated. I thought it was a bit optimistic of people to expect that Carl will do what he should, but maybe since the normative expectation was self-regarding, that's not too surprising.

2/3) For the attribution questions: we used a 7-point Likert scale, ranging from "strongly disagree" (1) to "strongly agree" (7).

4) Yes, we initially hypothesized that the "neither" and "both" conditions would generate the same pattern of responses, but participants were systematically more inclined to make attributions in the "neither" condition than the "both" condition, though only in what I called the "off" attributions, i.e., when the attribution has to do with investing after Carl donates or vice versa. For instance, in graph 1, participants were more likely to say that Carl intentionally helped others after he invested the money in the neither condition than the both condition. And in graph 2, they're more likely to say he intentioned made his retirement more comfortable after he donated the money in the neither condition than the both condition. The same goes for graphs 3 and 4. It's pretty weird.

The comments to this entry are closed.

3QD Prize 2012: Wesley Buckwalter