Article #16
1998
 
 
 
Newsletter list 

Newsletter 

Previous Article 

Next Article 

 
 
 
 

Lens model research at Buffalo

Ann Bisantz
Amherst, New York

In the past year, I have been involved with several judgement and decision-making research projects. While at Georgia Tech, I worked with Alex Kirlik on research which investigated how performance on several different judgment tasks might be adaptive to the nature of the uncertainty in the environment when people are exposed to that uncertainty through explicit experience acting in the environment, or by task formats which allow people to tap into their experiential knowledge of uncertainty. This work was based on prior research (e.g., Cosmides and Tooby, 1996; Kirby, 1994; Klayman and Ha, 1987) which suggested that performance on tasks such as evidential reasoning, and rule verification or hypothesis testing might reflect adaptation to environmental probabilities rather than non-normative biases, or might approach normative solutions when task formats mimic people's natural experience with uncertainty. The research showed some adaptivity to environmental probabilities in a categorization task and an evidential reasoning task. Other research with Kirlik, Neff Walker, Dan Fisk, Donita Phipps, and Paul Gay at Georgia Tech used the Lens Model to model judgments in a complex Naval Command and Control Task. This research was interesting because, due to the dynamic nature of the environment, and the degree to which participants could choose when they made judgments (in this case, identifying aircraft), and what information they had available to make a judgment, the cues and cue values for a given judgment were not consistent across judges. This necessitated the use of a different environmental model, and thus different values of Re, for each judge. Through the Lens Model analysis we found that differences in performance between good and poor performers were attributable to differences in consistency rather than differences in knowledge; this was consistent with results from an analysis of individual errors.

Finally, at SUNY Buffalo, I am working with two graduate students who are utilizing Lens Models in their research. Gordon Gattie has begun working in collaboration with faculty at the SUNY at Buffalo Dental School to develop a computer based training tool for dental students to help them learn to diagnose different oral diseases based on clinical photos, and plans to incorporate and test aspects of cognitive feedback in this application. In the context of a project through the Center for Multi-source Information Fusion at SUNY Buffalo, Younho Seong is interested in modeling human trust in automated systems that may be degraded or sabotaged. We are exploring the use of the Lens Model to capture aspects of calibration between the extent to which a human operator relies on or trusts information provided by an automated decision aid and the degree to which that system is in fact trustworthy.

References

Cosmides, L. & Tooby, J. (1996). Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under uncertainty. Cognition, 58, 1-73.

Kirby, K. N. (1994). Probabilities and utilities of fictional outcomes in Wason's four-card selection task. Cognition, 51, 1-28.

Klayman, J. & Ha, Y.-W. (1987). Confirmation, disconfirmation, and information in hypothesis testing. Psychological Review, 94(2), 211-228.

Contact Ann Bisantz

Previous Article                Next Article

Newsletter list