Agent-condition learning gains trended higher (marginal), with Agent LPK students gaining significantly more than Non-Agent HPK students¶
Claim · All claims
Evidence · 1 study (2 entries) · causalr1·q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person
Design limits · Man Su 2025 (2): the intervention came bundled with another change (read from the study by two models, 2026-10-27)
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.
q2 i? The main effect of condition on learning gains approached significance, indicating a trend towards higher gains in the Agent condition. → Man Su 2025
q2 i? The Agent LPK group achieved significantly higher learning gains than the Non-Agent HPK group in post-hoc comparisons. → Man Su 2025 (2)
Evidence¶
Man Su 2025¶
Man Su, Belle Dang, Andy Nguyen & Tomohiro Nagashima. (2025). Choice-making in an adaptive learning system with motivational pedagogical agents. npj Science of Learning. https://doi.org/10.1038/s41539-025-00366-7
q2 · i? · causal · r1
Two-way ANOVA on learning gains in the 49-student classroom study of the AlgeSPACE module. The condition main effect only "showed a trend towards higher gains" at p = 0.064; no significant interaction effect was found.
"The main effect of condition approached signi ficance (F(1, 45) = 3.62, p =0 . 0 6 4 ) , indicating that students in the Agent condition showed a trend towards higher gains."
Man Su 2025 (2)¶
Man Su, Belle Dang, Andy Nguyen & Tomohiro Nagashima. (2025). Choice-making in an adaptive learning system with motivational pedagogical agents. npj Science of Learning. https://doi.org/10.1038/s41539-025-00366-7
q2 · i? · causal · r1
Post-hoc pairwise (Tukey-adjusted) analyses from the same ANOVA on learning gains. Agent LPK significantly exceeded Non-Agent HPK (p = 0.002), a cross-condition, cross-prior-knowledge contrast reported alongside Table 1 descriptive gains.
"Post-hoc analyses indicated that the Agent LPK group achieved significantly higher learning gains compared to the Non-Agent HPK group (p = 0.002)."
Discussion¶
Related Claims¶
- Prior knowledge significantly predicted algebra posttest scores and learning gains, with high-prior-knowledge students scoring higher and low-prior-knowledge students gaining more — related
- Overall choice frequency did not differ by condition or prior knowledge, but the condition pattern reversed across prior-knowledge groups — related
- Lag sequential analysis revealed distinct choice-making trajectories: HPK students showed confidence-driven extra-task engagement and unre reflective repeated-failure sequences, while LPK students relied on preparatory worked examples — related
- Process mining showed LPK students began with preparatory tasks and achieved a higher proportion of correct decisions, while HPK students skipped preparation and increasingly took advanced tasks — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
agent-condition-learning-gains-trend-lpk |
title |
Agent-condition learning gains trended higher (marginal), with Agent LPK students gaining significantly more than Non-Agent HPK students |
description |
Agent-condition learning gains trended higher (marginal), with Agent LPK students gaining significantly more than Non-Agent HPK students |
status |
review |
evidence_strength |
weak |
newest_evidence |
2025 |
generated |
by process:wiki-ingest on 2026-10-27 |
trust |
unverified — no one has confirmed this page's content |
sources |
2 — man-su-2025, man-su-2025-2 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.