Identifying indicators of expertise in an asset valuation setting

Knowledge engineers facilitate knowledge transfer by identifying experts through their performance, extracting knowledge from an expert, and then converting that knowledge into a transferable form. Knowledge engineering researchers have identified the extraction process as the bottleneck in knowledge engineering. The research reported in this dissertation is designed to determine if indicators other than performance might provide knowledge engineers better criteria with which to identify experts who can articulate their knowledge.; This dissertation consists of two experiments. Experiment One employs a 2 x 2 design in which practice and rehearsal are manipulated on two levels. Experiment One accomplishes two important objectives: First, it provides a re-test of prior work in the context of the task used in this dissertation, and second, it results in a trained set of subjects for participation in the second experiment.; Experiment Two is designed to determine if the alternative indicators of expertise developed in previous research differ between individuals whose performance is identical in Experiment One. Three methods used in previous studies are simultaneously applied to examine the participants while they perform the experimental task. The cognitive strategies that the participants use are assessed, the way they verbally explain how they performed the task is scrutinized and their level of cognitive effort while performing the task is monitored by observing fluctuations in pupil diameter.; Examination of the data gathered during Experiment Two reveals that all four individuals perform the task proficiently. Furthermore, differences exist in their cognitive strategies and verbal explanations, but not of the magnitude that the theories underlying cognitive search strategies and verbal behavior predict. Statistical tests indicate significant differences in the participants' levels of cognitive effort when they are prompted to articulate the steps required to perform the task. Participants with more experience used more cognitive effort to access their articulable knowledge of the required steps than less-experienced participants. These results demonstrate that triangulating experimental methods provides better insights about people's level of experience. The results of Experiment Two show that level of cognitive effort is the best indicator of the participants' experience level and their ability to articulate task-specific knowledge.