In English

Presentationsinformation     2012-12-03 (14:15)   •  The seminar room at Vi2

Talare Anders Jansson
Titel Collegial Verbalisation: A Method for Knowledge Elicitation in Cognitive Analysis
Sammanfattning It is well documented that levels of expertise affect the way humans interact with control systems. Consequently, design of support systems ought to be based on the fact that level of expertise is important. This is not the same thing as customizing systems to each individual. Rather, the cognitive strategies exhibited by the domain experts can be seen as independent from the individual, and instead be interpreted as consequences of the environmental constraints acting upon the expert in each specific situation. In situations and contexts where domain knowledge is shared between colleagues, we might find that they also share cognitive strategies. The question is: Can they also verbalise on the actions and strategies exhibited by a colleague? And what is the value of such a report? Those are the questions I will try to answer in this seminar. For the purpose of knowledge elicitation in situations where target operators and colleagues are highly familiar with the same tasks and systems, there are several methodologies available. One example is different sorts of verbalisation methods. Traditionally, verbal reporting is carried out in the form of concurrent or retrospective verbalisations. These methods have advantages and disadvantages, respectively. Recently, we have explored and developed a new method where we let independent observers verbalise the observable actions and non-observable thoughts of target operators. In this seminar, I will give a short summary of the history of using verbal protocols as a method for knowledge elicitation, the problems associated with the different methods, and in what way our method solves some of the problems. The usefulness and value of our method will be discussed in relation to criteria for assessing verbal reports. The new method has a number of interesting pay-offs, for example the possibility to compare different versions of verbalizations on the same task, making it possible to identify maladaptive mental models within the team of operators, and the possibility of using collegial reports for team-learning as part of proactive safety cultures.