Edwin Hutchins's position on self-confrontation (a research methology that asks participants to explain their activity based on traces of their interactions, e.g. a video), taken from a discussion with the cognitive anthropology research group:
« My position with respect to verbal data goes back to the earliest work I told you on Monday [Hutchins, 1980], in which I was working at discourse and looking for a structure which seemed to be repeated, a schema, which seemed to account for memorization of the discourse that does not account for particular facts as mentioned in the discourse (…). The underlying hypothesis is that, in terms of this cognitive system, in order to communicate effectively from one person to the other, we have no choice but to use these structures that we share, that are legitimate ways of assembling ideas in our culture (…). If I violate the terms of these kinds of structures, then people say that does not make any sense, that those things don't go together that way. » (...) the problem for me precisely is : I observe some people engaged in a task, and who are producing verbal behaviour, the pilots for example, and I look at this action and communication and see if I can discover recurrent structures which tell me what pilots expect. If this happened and then that happens, we also expect that to happen. And if this did not happen, that will not happen. This is a structure of belief that pilots have about how events happen together, how events cause each other or precludes each other. And we would like to find out what that is. We can get data about that by watching them actually doing the task, or we can ask them questions about it, we can interview them. What we look for when they talk to us is not the truth or the falseness of the assertion they make. We are looking at what is the structure of the schema by which this is a sensible thing to say. Just as if we're going to the used-car salesman, we have to study the car salesman : we don't believe what he says but we ask why it is a reasonable thing for him to say that this automobile was owned by an old lady. So, if we come to the question of what do we do with the data of self-confrontation, the question is : what kind of interpretants are involved ? We present the subject with a task which is : generate for me a culturally meaningful account of your own behaviour while we will remind you what happened by showing you this videotape. At that point, the question is : do we take the content of what the subject says as to actually be information about what happened in that very event, or do we look at the structure of the account to see what it is that pilots believe is a meaningful way to construct a story about what happened in the event ? (…) The point is: shall we take the linguistic behaviour that is produced here as something which is true, or is this another source of data about the structure that subjects believe ? »
Why do I blog this? this statement is important in terms of how we can use self-confrontation (as we do in CatchBob!): participants should be placed in situations where their account should not just be culturally possible as stated by Jacques Theureau in his next book.