Evaluation of Context Awareness in Similarity Measurement: Total-Set Versus Classic Pairwise Comparison

An Evaluation of Context Awareness in Similarity Measurement: Total-Set Versus Classic Pairwise Comparison poster

Research Authorship:

Kayleigh Aubin & Dr. John D. Kulpa

Faculty Mentor:

Dr. John Kulpa | College of Arts and Sciences | Department of Psychology

Abstract:

How someone perceives similarity gives insight into how they learn and process information. Pairwise comparison is a useful tool in determining a person’s perception of similarity. In classic pairwise comparison, a participant is shown two items of a set at a time, repeating the process until the entirety of the set has been evaluated. Total-set pairwise comparison shows the participant the entire set while highlighting items in the set to evaluate. It has been assumed that, in the classic method, the participant’s judgments across trials are made with increasing awareness of the total-set context, and that the total-set method fixes this awareness from the start. Our study will systematically evaluate changes in awareness across trials in each procedure and explore whether there is a bias toward basic-level representation of the (unknown) total set in the classic method. The program created will evaluate and probe participants on three levels of superordinate, basic, and subordinate in both classic and total-set pairwise comparison. Thus far, the program measuring this study has been created and is currently in the stages of data collection. The results of our study will help researchers to more intelligently choose between classic and total-set pairwise comparison when measuring subjective similarity.

css.php