Agentic education · Learning science

Actionable companion · Feedback & scaffolding

Learning with agency: The impact of artificial intelligence on students’ metacognition: a systematic review

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Age / stage not established. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Abstract Given the critical role of metacognition in academic achievement and the rapid integration of Artificial Intelligence (AI) in educational settings, understanding the interplay between the two has become essential. In recent years, a growing number of studies have examined their relationship, creating a pressing need for a systematic review to synthesize the accumulating evidence. This review adhered to the PRISMA guidelines, involving the search and screening of papers from three major databases (Web of Science, Scopus, EBSCOhost). A total of 22 empirical studies were included in the final synthesis. Data extraction and thematic analysis were employed to identify key impacts and underlying interaction patterns. This review distinguishes between Learning-Oriented AI (LAI) and General-Purpose AI (GAI) to organize the evidence. The findings suggest that LAI was more often associated with Guided Interaction, in which AI provides guiding feedback without directly supplying answers, whereas GAI was more commonly associated with Direct Feedback Interaction, in which AI provides answers or explicitly corrects errors. Across both categories, positive findings were reported for several metacognitive outcomes, including regulation, awareness, and shared metacognition. However, some GAI studies also reported that students relied excessively on AI, particularly when they accepted direct answers from AI without sufficient reflection. Overall, the findings highlight the importance of interaction design and teacher guidance in shaping how AI supports students’ metacognitive development. Keywords: metacognition, artificial intelligence, interactive pattern, Learning-Oriented AI, General-Purpose AI

What kind of evidence is this?

Supporting AI learning science · Evidence synthesis

Stage: Age / stage not established. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

This study employed the Systematic Literature Review (SLR) method, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This study was not registered in PROSPERO, which mainly serves health-related reviews, and this aligns with some systematic reviews in education domain (e.g., Hu and Xiao, 2025; Qian et al., 2026). In addition, the completed PRISMA 2020 checklist is provided as Supplementary Material. In accordance with methodological standards recommending the use of three or more databases as sources (Page et al., 2021), searches were conducted in Web of Science, Scopus, and the EBSCOhost platform (see Table 1). The EBSCOhost search was conducted through the institutional platform’s integrated search interface, which simultaneously searched across multiple subscribed databases, including Academic Search Premier, APA PsycINFO, APA PsycArticles, and ERIC. These sources were selected because they provide broad coverage of the interdisciplinary literature relevant to this review, including education, educational psychology, educational technology, and artificial intelligence. The search cutoff date was set to June 6, 2025, with no restrictions on the start date, educational level, or other criteria. The initial search yielded 379 papers (175 from Web of Science, 148 from EBSCO, and 56 from Scopus). After removing 20 duplicates, 359 papers proceeded to the screening stage (see Figure 1).

Limitations and uncertainty · source passage

Based on a systematic review of 22 empirical studies, this study thoroughly investigates the impact of AI on students’ metacognition. The main conclusions can be summarized as follows:

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

  • Give feedback on the learner’s reasoning, then fade hints. Record whether the learner can complete the next task without that scaffold.
  • Use plan–attempt–check–reflect prompts and compare confidence with actual performance. Keep learner choice and reflection visible.

A 90-day district pilot

  1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
  2. Days 16–30 · Design. Treat the paper as background evidence until its population is verified in the full text. Do not assign an age-specific intervention or extend adult results to children. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
  4. Days 61–90 · Decide. Extract the participant ages or education stage, recruitment, task and comparator before selecting a local pilot. If these cannot be established, retain the paper for design discussion rather than age-band efficacy claims. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Chunyu Ji; Qi Tan. The impact of artificial intelligence on students’ metacognition: a systematic review. Frontiers in Psychology (2026-10-08). https://doi.org/10.3389/fpsyg.2026.1947198

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.