Skip to main content
Fig. 3 | Cognitive Research: Principles and Implications

Fig. 3

From: ROC curve analyses of eyewitness identification decisions: An analysis of the recent debate

Fig. 3

Left panel: Diagnosticity (hit rate/false alarm rate) as a function of true area under the receiver operating characteristic curve (AUC = 0.6, 0.7, 0.8, or 0.9). For each AUC level, the false alarm rate varies from 0.01 (more conservative) to 0.99 (more liberal); given each false alarm rate and AUC, the hit rate is determined by H = Φ[√2 z(AUC) + z(F)]. Right panel: For the same AUC, hit rates, and false alarm rates, diagnosticity is plotted as a function of response bias [measured with c = −0.5(zH + zF)]

Back to article page