When AI Learns With Students: The Impact of Cognitive Co-Pilots on Self-Regulated Learning in Digital Education
Year: 2025
Type: article
Study design: Observational Study
Abstract: Recent studies emphasize a notable transition in e-learning platforms from features driven by automation to tailored learning experiences facilitated by artificial intelligence [1].Although this shift has enhanced content flexibility and student involvement, its educational consequences are still insufficiently examined.Specifically, little focus has been placed on the impact of AI-driven personalization on students' self-regulated learning activities, such as goal-setting, monitoring, and the use of cognitive strategies.As AI systems like ChatGPT evolve into interactive learning partners instead of merely passive tools, an important question arises: does this customization promote learner independence or encourage reliance?Generative AI has been envisioned as a "cognitive co-pilot" in language acquisition, effectively aiding in idea formulation, writing progression, and linguistic precision [2].Likewise, in math education, AI-powered tools assist in problem-solving via guided engagement and incremental assistance [3].Additionally, AI has been described as a "metacognitive mirror," enabling learners to monitor and repair comprehension breakdowns during reading activities [4].Though these advancements showcase AI's ability to transform various fields, they also highlight worries about excessive dependence, decreased cognitive involvement, and fewer chances for constructive challenges.This research presents a conceptual and empirical examination of artificial intelligence as a cognitive co-pilot in online learning environments, with a particular focus on its influence on self-regulated learning.Employing a quantitative research design based on survey data, the study examines how varying levels of AI support shape learner autonomy, metacognitive awareness, and knowledge retention.Data collected from students engaged in distance education are analyzed using SPSS for descriptive statistics and reliability testing, and SmartPLS for structural equation modeling to assess relationships between key constructs.The study contributes to the field by bridging personalization, co-pilot theory, and self-regulated learning, offering a novel framework for understanding human-AI collaboration in education.It further provides practical implications for designing AI-integrated learning environments that balance support with the preservation of independent thinking.
Authors: Mirna Safi, Farid Abdallah
Institution: Australian University, Kuwait
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: bronze
OpenAlex ID: W7222000551