Table 2.

Subjects more accurately identify repeat and foil stimuli compared to lures across all standard experiments.

Experiment	Accuracy: repeat	Accuracy: lure	Accuracy: foil
E1	0.86 (0.18)	0.40 (0.32)	0.77 (0.30)
E2	0.87 (0.09)	0.41 (0.29)	0.78 (0.18)
E3	0.84 (0.15)	0.54 (0.23)	0.80 (0.13)
E4	0.84 (0.18)	0.44 (0.22)	0.82 (0.13)
E5	0.81 (0.14)	0.52 (0.24)	0.88 (0.13)
E6	0.92 (0.12)	0.31 (0.27)	0.86 (0.16)
E7a	0.87 (0.14)	0.64 (0.18)	0.87 (0.11)
E7b	0.34 (0.27)	0.29 (0.12)	0.71 (0.20)

We show median (IQR) accuracy for each response type. Across most experiments, subjects are more accurate when identifying repeat and foil stimuli vs. lure stimuli (Wilcoxon Rank Sign Test comparing repeat vs lure accuracy and foil vs. lure accuracy, E1−E7a: p < 0.001). This is not the case for experiment 7b, delayed test, where subjects were most accurate at identifying foil stimuli (though see Table 4: they were also most often making Foil responses).