Agentic education · Learning science

Actionable companion · Self-regulation & metacognition

Learning with agency: Choice-making in an adaptive learning system with motivational pedagogical agents

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:

Adaptive learning systems increasingly employ pedagogical agents (PAs) to enhance students’ engagement and learning outcomes, yet little is known about how motivational PAs influence students’ conceptual understanding and strategic choice-making. This study compared 49 school students (at 9th and 10th grade) using an adaptive algebra learning environment with either motivational PAs or instructional prompts (non-PAs). Results revealed that while all students demonstrated learning gains, prior knowledge significantly moderated outcomes. Lower-knowledge students achieved the greatest gains through reflective engagement with foundational tasks, whereas higher-knowledge students often adopted intuitive but error-prone strategies. Notably, process mining and lag sequential analysis revealed distinct choice-making trajectories, uncovering how motivational PAs influenced self-regulation patterns over time. This study advances the field by operationalizing choice-making as a measurable self-regulated learning construct and reframing strategic disengagement as an adaptive, agentic behavior. Findings underscore the importance of designing adaptive systems that support both content mastery and strategic choice-making.

What kind of evidence is this?

Conversational & pedagogical agents · Design requires full-text appraisal

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

The study examines AlgeSPACE, an adaptive learning system designed for early algebra, co-developed with a school teacher over 13 months to support students’ “learning-by-doing” through stepwise problem-solving63. AlgeSPACE includes four independent modules focused on solving linear systems of equations, emphasizing distinct strategies such as substitution and elimination methods through real-world examples as well as flexible problem solving using the most suitable strategy. Students can engage with the modules in any order, as they operate independently without predefined sequences.

Limitations and uncertainty · source passage

Despite the integration of PAs into adaptive learning systems, a critical gap remains in understanding how PAs influence students’ choice-making behaviors, particularly in relation to their prior knowledge. While systems such as Betty’s Brain and MetaTutor provide motivational and instructional scaffolds, their impact on decision-making processes is often examined through aggregated metrics such as task completion rates rather than granular process data like problem-solving pathways, hint request frequency, and persistence after errors. This limitation prevents a nuanced understanding of how learners dynamically interact with PAs, highlighting the need for research that captures the fine-grained mechanisms underlying choice-making within adaptive learning environments.

Design the system before the pilot

Define whether the agent asks questions, gives hints, explains or assesses. Test hallucinated feedback, excessive praise, misleading confidence and responses when the learner is confused. A conversational interface alone does not establish autonomous agency.

  • Use plan–attempt–check–reflect prompts and compare confidence with actual performance. Keep learner choice and reflection visible.
  • Measure engagement separately from achievement. Check whether novelty-driven participation persists and whether it translates to independent performance.

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

Man Su; Belle Dang; Andy Nguyen; Tomohiro Nagashima. Choice-making in an adaptive learning system with motivational pedagogical agents. NPJ Science of Learning (2025-11-18). https://doi.org/10.1038/s41539-025-00366-7

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.