The Bayesian yoking procedure closely approximates individualized ideal learning paths, with no posttest or transfer difference between perfect and slightly mismatched children¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2–r3 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Yoked responses matched an individualized ideal model 94.6% of the time on average (range: 79–100%), with 43% of children receiving a perfect match. → Brod 2025 q2 i? Children with a perfect match and those with a slight mismatch did not differ significantly on posttest (79% vs. 83%, t(44) = −0.48, p = 0.632) or transfer test (92% vs. 88%, t(44) = 0.47, p = 0.638). → Brod 2025 (2)

Evidence¶

Brod 2025¶

Brod, G., Holstein, E., Weindorf, L., Colantonio, J., Bonawitz, E. & Theobald, M. (2025). Do it yourself: discerning the effects of self-directed activity on conceptual learning. npj Science of Learning. https://doi.org/10.1038/s41539-025-00364-9

q2 · i? · causal · r2

Procedure validation: "yoked responses matched an individualizedideal model 94.6% of the time (range: 79–100%), with 43% of children (20/46) receiving a perfect match." Pilot testing and simulations indicated the binning approach captured children's priors with high precision.

"yoked responses matched an individualizedideal model 94.6% of the time (range: 79–100%), with 43% of children (20/46) receiving a perfect match"

Brod 2025 (2)¶

Brod, G., Holstein, E., Weindorf, L., Colantonio, J., Bonawitz, E. & Theobald, M. (2025). Do it yourself: discerning the effects of self-directed activity on conceptual learning. npj Science of Learning. https://doi.org/10.1038/s41539-025-00364-9

q2 · i? · causal · r3

Null comparison between perfect-match and slight-mismatch children: no significant posttest difference (79% vs. 83%, p = 0.632) or transfer difference (92% vs. 88%, p = 0.638), "with no observable impact from slight mismatches."

"children with a perfectmatch and children with a slight mismatch did not differ in performance in the posttest (79% vs. 83%, t(44) = −0.48, p = 0.632) or transfer test (92% vs. 88%, t(44) = 0.47, p = 0.638)"

Discussion¶

Related Claims¶

• The Excel sheet's knower-level inference closely matches fully Bayesian inference, with no discrepancies in posterior maxima — related
• Children's active-condition predictions closely align with an optimal Bayesian learner, supporting rational Bayesian inference as a model of children's belief revision — related
• DIY prediction advantage on transfer is limited to receptive transfer and does not extend significantly to productive or explicit conceptual transfer — related