Agentic education · Learning science

Actionable companion · Retention & transfer

Learning with agency: Using a coloring activity to identify children’s development of visual–motor integration: an application of artificial intelligence

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): Early childhood · preschool, Children · stage unspecified. Verify the participant population and task before transferring the findings.

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

Abstract Aim Visual–motor integration (VMI) is an important indicator in children with learning disabilities. We aimed to use performance in a coloring activity to identify children’s VMI developmental status. Methods A sample of 505 preschool children (mean = 57.64, SD = 11.10) were recruited. Among them, data from 404 and 101 children were used as the training and testing data, respectively. The Beery–Buktenica Developmental Test of Visual–motor Integration, fourth Edition, (VMI-4) was used as an indicator for the model of artificial intelligence (AI). The total scores of the VMI-4 were calculated, and then based on the children’s age, the total scores were transferred into standard scores and the developmental status of visual–motor integration. The AI model comprised a regression model and classification model to predict the developmental status rated by the VMI-4. Results In the training data, we found that the AI model comprising the support vector machine (SVM) regression model and eXtreme Gradient Boostin (XGBoost) classification model exhibited the best performance (accuracy: 86.2%; sensitivity: 84.7%; and specificity, 85.4%). The results of the trained AI model on the testing data indicated good performance, with accuracy, sensitivity, and specificity of 80.20%, 73.68%, and 81.71%, respectively. Conclusions Combining the coloring activity with the AI technique has great potential as a screening tool to identify children’s VMI developmental status.

What kind of evidence is this?

Supporting AI learning science · Design requires full-text appraisal

Stage: Early childhood · preschool; Children · stage unspecified. 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

Preschool children were recruited for this study. The inclusion criteria were (1) age of 3 to 6 years, (2) ability to participate in group activities, and (3) informed consent provided by caregivers. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and was approved by the Institutional Review Board of E-DA hospital in Kaohsiung, Taiwan [EMRP46107N]. This study was conducted during 2020 to 2021.

Limitations and uncertainty · source passage

Our study had two limitations. The first is that group activities rather than individual activities were used to collect children’s coloring products. Children’s performance may be influenced by the assessment environment (e.g. peers’ movements and noise may distract them). Thus, children’s performance in group activities may differ from that in individual activities. More studies are needed to collect children’s activity products in individual activities to compare the differences between the two data collection methods. Second, the children were recruited from kindergartens; therefore, most of the children in this study had typical development. Future studies should recruit children in rehabilitation clinics to increase the sample variation and further increase the generalizability of the AI model.

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.

  • Include a delayed assessment and an unfamiliar transfer problem without AI. Immediate supported performance cannot establish durable learning.
  • Co-design accommodations with learners and specialists. Check access and error burden across groups without inferring subgroup effectiveness from small or unreported samples.

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. Use a brief adult-mediated story, turn-taking or language activity. An educator controls every interaction; the agent never becomes an unsupervised companion. Preserve hands-on play and human conversation. 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. Measure independent naming, explanation or turn-taking in a human-led activity after the tool is removed. Record frustration, overstimulation and the amount of adult help. 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

Tzu-Yun Huang; Kuan-Lin Chen; Gong-Hong Lin; Chien-Yu Huang. Using a coloring activity to identify children’s development of visual–motor integration: an application of artificial intelligence. Annals of Medicine (2025-11-03). https://doi.org/10.1080/07853890.2025.2578725

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.