Figure 3
From: A Memory Computational Basis for the Other-Race Effect
Both d’ and Area Under the Curve measures find ORE in Mnemonic Discrimination (MD) but not Match-to-Sample (MTS) tasks. (A) An Analysis of Variance finds significant main effects of task (p < 0.0001), stimulus race, (p = 0.0345) and an interaction of the two (p < 0.0089) on performance. A post hoc multiple comparisons test finds a significant difference in performance for SR and OR faces only during mnemonic discrimination (p < 0.0001, labeled on figure A). (B) Comparing area under the curves (AUCs) for SR and OR faces in both mnemonic discrimination and match-to-sample tasks reveals a strong ORE only in the mnemonic discrimination task. The larger the significant difference in SR and OR AUCs, the greater the ORE. Analysis revealed significant main effects of both task type (p < 0.0001) and stimulus race (p < 0.05) on performance, as well as a significant interaction of the two (p < 0.005). A post hoc test found significant AUC differences for SR and OR input-output transfer functions for the mnemonic discrimination but not match-to-sample tasks (p < 0.0001, labeled on figure B).
