Knowledge elicitation as abstraction of purposive behaviour

  • Date:
  • Location: Sal IX, Universitetshuset, Biskopsgatan 3, Uppsala
  • Doctoral student: Axelsson, Anton
  • About the dissertation
  • Organiser: Avdelningen för visuell information och interaktion
  • Contact person: Axelsson, Anton
  • Disputation

Researchers use knowledge elicitation methods to document expert knowledge for the primary purpose of understanding cognitive processes and with this understanding, technical solutions to resolve human factors issues can be produced. This dissertation offers a novel perspective on knowledge elicitation as an abstraction process.

Such a theoretical framework has emerged by consolidating the ecological approach of Brunswikian psychology with the ideas of tacit and personal knowledge of Polanyian epistemology. Traditionally, knowledge elicitation has been considered an extraction process in which knowledge can be readily transferred from one individual to another. Here, this traditional position is rejected in favour of Polanyi’s premise that much of the knowledge individuals possess is tacit in nature, which implies that it cannot be documented easily, expressed in explicit form or explained. In this dissertation, knowledge is characterised as a personal process of knowing, highlighting context as a subjective knowledge structure of personal experiences that is formulated implicitly and indirectly over time through a dynamic interaction with the environment. Therefore, tacit knowledge cannot be articulated or shared; however, learners can be inspired by observing other individuals' purposive (i.e., goal-directed) behaviours and thus shape their own tacit knowledge once they practise the observed skills and develop conceptual understanding through reasoning about the learning process. Knowledge elicitation thereby makes use of observations, questions, or more structured process tracing methods in environments familiar to the observed individuals to elicit purposive behaviour from them. Accordingly, functional descriptions can be produced in this process that further conceptual understanding of a particular domain. Knowledge elicitation procedures are a powerful set of methods for reaching such functional descriptions. Moreover, by understanding the resulting knowledge elicitation data as an abstraction derived from multiple collection points in the same environment, the focus shifts from purely subjective mental constructs to the impact of environmental constraints.