In Experiment 1, the Confirmatory-Only condition learned the target category faster than the Complete condition after the category switch.¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q2 · i2–i3
Currency · newest study 2015 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i2 Across blocks 6–10 the Confirmatory-Only condition outperformed the Complete condition, ηp2 = .096. → Ramsburg 2015 q2 i3 The rate of learning was faster for the Confirmatory-Only condition, interaction ηp2 = .146. → Ramsburg 2015 (2)

Evidence¶

Ramsburg 2015¶

Ramsburg, J. T., & Ohlsson, S. (2015). Category Change in the Absence of Cognitive Conflict. https://www.google.com/search?q=%22Category+Change+in+the+Absence+of+Cognitive+Conflict%22

q2 · i2 · causal · r?

Repeated-measures ANOVA with Greenhouse-Geisser correction on percentage correct for blocks 6–10 in Experiment 1. The condition main effect showed "the Confirmatory-Only condition performed better than the Complete condition, F(1, 72) = 7.60, p < .01, ηp2 = .096".

"The analysis revealed a main effect for blocks showing that participants improved with training, F(2.646, 190.477) = 75.01, p < .001, ηp2 = .671, and a main effect for condition showing that the Confirmatory-Only condition performed better than the Complete condition, F(1, 72) = 7.60, p < .01, ηp2 = .096."

Ramsburg 2015 (2)¶

Ramsburg, J. T., & Ohlsson, S. (2015). Category Change in the Absence of Cognitive Conflict. https://www.google.com/search?q=%22Category+Change+in+the+Absence+of+Cognitive+Conflict%22

q2 · i3 · causal · r2

The condition × blocks interaction in the same Experiment 1 repeated-measures ANOVA was significant: "The rate of learning was faster for the Confirmatory-Only condition than the Complete condition, F(2.646, 190.477) = 5.21, p < .01, ηp2 = .146".

"The rate of learning was faster for the Confirmatory-Only condition than the Complete condition, F(2.646, 190.477) = 5.21, p < .01, ηp2 = .146."

Discussion¶

Related Claims¶

• Experiment 2 replicated that direct falsification of the initial category is not necessary for learning the new category, with a Confirmatory-Only advantage on block 6 only. — related
• After successful re-categorization, learners' feature ratings show the prior conception is retained in memory alongside the newly learned conception. — related
• In Experiment 1, the extent of informative interleaving did not correlate with category assignment performance, which the authors interpret as a utilization deficiency — related