AI cognitive differences and the evolution of learning behaviors from an interdiscipli-nary perspective: a data-driven study based on a survey of AI literacy among college students

Abstract

This study analyzes the current situation and challenges of the integration of AI and education, and finds that existing studies have problems such as insufficient theoretical depth and lack of methodological innovation. Through a data-driven approach, the study parsed the differences in AI cognition among college student groups in an interdisciplinary context, and constructed a model for the evolution of learning behaviors by combining behavioral data. The study suggests that future research should strengthen dynamic analysis and interdisciplinary integration, promote the transformation of AI literacy assessment from static indicators to multi-dimensional cognitive landscape, and construct a “three-dimensional integrated” AI literacy cultivation system, so as to ultimately realize the transformation of the AI education ecosystem from instrumental application to cognitively enhanced ecological construction.