The Influence of AI-Assisted Problem Solving on Decision Quality: Examining the Moderating Role of Cognitive Reflection

This study examined the influence of AI-assisted problem solving on decision quality and the moderating role of cognitive reflection among university students in Pakistan. A quantitative, cross-sectional research design was employed, with a target sample of 335 undergraduate and postgraduate students from selected public and private universities, including the University of the Punjab, Bahauddin Zakariya University, Government College University Faisalabad, University of Karachi, University of Sindh, University of Peshawar, University of Malakand, and International Islamic University Islamabad. Primary data were planned to be collected through a structured, self-administered questionnaire distributed using Google Forms, while convenience sampling was adopted as the proposed sampling technique. The statistical analysis framework included Pearson correlation, simple linear regression, and moderation analysis using Hayes’ PROCESS Macro Model 1 in IBM SPSS Statistics. The preliminary analysis tables indicated a positive relationship between AI-assisted problem solving and decision quality (r = .582, p < .01), while regression analysis indicated that AI-assisted problem solving positively predicted decision quality (β = .582, R² = .339, p < .001). The moderation analysis further indicated a positive interaction between AI-assisted problem solving and cognitive reflection (B = .156, p = .001), suggesting that cognitive reflection may strengthen the relationship between AI-assisted problem solving and decision quality. These findings highlight the potential importance of combining AI-based support with analytical reasoning and reflective judgment in academic decision-making. The study addresses an important issue in higher education because the growing use of generative AI requires students to evaluate AI-generated information carefully rather than rely on automated recommendations without independent assessment. The research may contribute to understanding how individual cognitive characteristics influence the usefulness of AI tools and may help educators design learning activities that integrate AI literacy, critical thinking, and responsible technology use. The findings may also inform university policies and teaching practices aimed at encouraging effective AI adoption while preserving students’ independent reasoning and decision-making abilities. Future research using experimental or longitudinal designs may provide stronger evidence about the direction of these relationships and clarify how cognitive reflection influences decision quality across different academic contexts.