Improving learning performance through AI-enabled task completing and network building: integrating task-technology fit and social-technology fit perspectives

Artificial intelligence-enabled learning applications (AILAs) are increasingly used to support personalized learning and peer interactions. However, research often overlooks how technological (i.e., task-technology fit, TTF) and social factors (i.e., social-technology fit, STF) jointly influence student performance. Drawing upon the extension of TTF model, we developed a nomological network model to explore how TTF and STF affect learning performance. A survey was conducted among 278 university students using AILAs. The data analysis indicated that both TTF and STF have significant positive main effects and interaction effects on learning performance. Further, AI-enabled pattern recognition, feedback generation, and content personalization positively influence TTF, while network size and network homophily positively influence STF. Unlike existing research that focuses solely on the technological aspects of AI technologies in education while neglecting their social aspects, our study expands the TTF model by incorporating the STF. Moreover, our study offers a contextualized typology of AI-enabled task completing and AI-enabled network building, providing valuable empirical evidence in the domain of AILAs for higher education. Practically, higher education institutions can acknowledge the technological and social advantages of AILAs, and AILA practitioners can enhance product design based on our findings.