BKT adaptivity in a mixed-reality Intelligent Science Station improves children's post-test learning of scientific principles compared to random task selection¶
Claim · All claims
Evidence · 1 study · causalr2·q3·i2medium
Currency · newest study 2024 · checked against its sources 2026-10-28 (automated, 1 of 1 entry) · not reviewed by a person
Design limits · Yannier 2024: groups were assigned but students were analysed as independent (read from the study by two models, 2026-10-28)
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q3 i2 Children using BKT-adaptive task selection scored higher on the paper post-test (69% vs 54%) than children receiving random tasks. → Yannier 2024
Evidence¶
Yannier 2024¶
Yannier, N., Hudson, S. E., Chang, H., & Koedinger, K. R. (2024). AI Adaptivity in a Mixed-Reality System Improves Learning. International Journal of Artificial Intelligence in Education, 34, 1541–1558. https://doi.org/10.1007/s40593-023-00388-5
q3 · i2 · causal · r2
Randomized controlled experiment with 36 children (18 pairs) at a science center; ANCOVA on post-test with pre-test as covariate found "69% for the adaptive condition and 54% for the randomized condition", d = 0.66.
"The overall results indicated that the average scores on the full post-tests (prediction and explanation items) was 69% for the adaptive condition and 54% for the randomized condition, F(1,34) = 4.02, p = 0.05, d = 0.66 (Cohen's d)."
Discussion¶
Related Claims¶
- Adaptive condition shows only a non-significant trend for transfer to engineering outcomes (tower building) — related
- Log analysis shows adaptive condition concentrates practice on unmastered concepts while random condition over-practices easy ones — related
- Adaptivity does not sacrifice enjoyment — related
- AI-mediated feedback in hands-on exhibits improves learning and engagement — related
- Adaptive scaffolding policies (BKT and DRL) significantly improve posttest performance over a non-adaptive control in a logic ITS — a broader claim this one bears on
- Adaptive task difficulty in an ITS did not improve second-graders' mathematics posttest performance compared with an identical non-adaptive ITS — related
- All training policies narrowed the prior-knowledge achievement gap from pretest to posttest, with BKT achieving the largest reduction (77.1%) — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
bkt-adaptivity-mixed-reality-improves-science-learning |
title |
BKT adaptivity in a mixed-reality Intelligent Science Station improves children's post-test learning of scientific principles compared to random task selection |
description |
BKT adaptivity in a mixed-reality Intelligent Science Station improves children's post-test learning of scientific principles compared to random task selection |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2024 |
source_checked |
2026-10-28 |
generated |
by process:wiki-ingest on 2026-10-28 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — yannier-2024 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.