Date
9-17-2026
Department
School of Behavioral Sciences
Degree
Doctor of Philosophy in Psychology (PhD)
Chair
Holly Walker
Keywords
AI, artificial intelligence, trust, curiosity, need for cognitive closure, need for closure, familiarity, ill-defined problems, ill-structured problems
Disciplines
Psychology | Social and Behavioral Sciences
Recommended Citation
Bouchard, Joel, "The Correlation of Information-Seeking Traits with Perceptions of Artificial Intelligence Output to Ill-Defined Problems" (2026). Doctoral Dissertations and Projects. 8868.
https://digitalcommons.liberty.edu/doctoral/8868
Abstract
Artificial intelligence (AI) has emerged as a new class of technological tool whose innovation exists in its ability to engage in reciprocal, socially fluent, undetermined response generation. How individuals come to trust such responses, especially as it regards classes of problems previously unavailable to programmed computational responses, is a question of importance. Prior research has addressed motivational dispositions towards epistemic decision-making, but work examining the relationship between these traits and perceptions of AI output in epistemically ambiguous circumstances is limited. The current study aimed to address this gap by exploring how trait curiosity and need for cognitive closure (NFCC) relate to perceptions of output from ChatGPT 5.2, and what their individual contributions may be. The proposal to facilitate this made use of a correlational, self-report online survey, employing reliable and valid instruments with adult participants. Exposing participants to three ChatGPT responses addressing Reflective Judgment Model problems and rating their trust in such responses allowed for bivariate correlations to be calculated for their trust assessments and their trait curiosity and NFCC scores, which revealed no relationships. A multiple regression analysis was performed to delineate the respective impact of information-seeking traits while controlling for other factors. Only AI familiarity was significantly related to trust in AI with ill-defined problems. Results may inform future theoretical research regarding epistemic orientations and emerging technologies, as well as practical educational, organizational, and consulting contexts.
