Rate of change in working memory performance correlates with fluid intelligence independently of final accuracy¶

Claim · All claims
Evidence · 1 study (2 entries) · associational r2 · q3 · i2 medium
Currency · newest study 2021 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i2 Rate of change on the WM task correlated with Gf (r(85) = −0.30), with steeper trajectories (change in less time) associated with higher Gf. → Aaron Cochrane and C. Shawn Green 2021 q3 i? The rate-of-change link remained reliable when controlling for final accuracy in a regression model (b = −0.111, ΔR2 = 0.046). → Aaron Cochrane and C. Shawn Green 2021 (2)

Evidence¶

Aaron Cochrane and C. Shawn Green 2021¶

Aaron Cochrane and C. Shawn Green. (2021). Trajectories of performance change indicate multiple dissociable links between working memory and fluid intelligence. npj Science of Learning. https://doi.org/10.1038/s41539-021-00111-w

q3 · i2 · associational · r2

Correlational analysis of time-sensitive WM modeling in 87 participants. Rate of change related to Gf with a printed correlation of r = −0.30 (CI = [−0.49, −0.08], BFlog3 = 2.05); the same pattern held using Spearman ρ.

"both rate of change on the WM task ( r(85) = −0.30, CI = [−0.49, −0.08], BFlog3 = 2.05) and final accuracy ( r(85) = 0.33, CI = [0.11, 0.52], BF log3 = 2.99) showed evidence of relations to Gf"

Aaron Cochrane and C. Shawn Green 2021 (2)¶

Aaron Cochrane and C. Shawn Green. (2021). Trajectories of performance change indicate multiple dissociable links between working memory and fluid intelligence. npj Science of Learning. https://doi.org/10.1038/s41539-021-00111-w

q3 · i? · associational · r2

Regression analysis of the same 87-participant experiment confirmed the two links' independence: "each remained reliable when controlling for the other", with printed b and ΔR2 values. Results persisted under robust regression and with starting accuracy as covariate.

"a regression model was run containing main effects for both predictors (rate of change and final accuracy), and each remained reliable when controlling for the other (rate of change b = −0.111, CI = [−0.213, −0.008], ΔR2 = 0.046; final accuracy b = 0.428, CI = [0.112,.745], ΔR2 = 0.073"

Discussion¶

Learner Variables¶

• General Cognitive Ability — predictor: learners who differ on it differ in outcomes

Related Claims¶

• Estimated final WM accuracy is a statistical near-isomorphism of the standard overall-accuracy measure — related
• Final estimated working memory accuracy correlates positively with fluid intelligence — related
• Initial working memory accuracy shows no reliable relation with fluid intelligence — related
• Fluid abilities (Gf) significantly predicted the rate of learning, whereas crystallized abilities (Gc) significantly predicted final learning performance in a one-week instructional process — related