Table 2.

Mean d’ for intact images used in the testing sessions

Experiment and training schedule	Testing image	Pre-test	Post-test 1	Post-test 2	Post-test 3	Final test	Pre-test vs. final test
1A (N = 12) 30-image	Trained	−0.04	0.67	0.64	0.95	1.58	P < .001
	Untrained	−0.10	0.40	0.22	0.68	0.88	P < .002
1B (N = 12) 60-image	Trained	0.30	0.76	0.96	0.86	0.88	P < .001
	Untrained	−0.03	0.47	0.88	0.48	1.01	P < .007
2 (N = 12) 60-image	Trained	0.18	0.94	0.49	1.01	0.68	P = .16a
	Untrained	−0.15	0.80	0.54	0.70	0.66	P < .005
3 (N = 24) 30-image	Trained	−0.11	Not applicable			1.14	P < .001
	Untrained	0.08	Not applicable			0.74	P < .001

P values show results from t tests comparing pre-test with the final post-test

^aIn experiment 2, although the pre-test did not differ significantly from the final post-test, it was significantly worse than the average of all four post-tests, p < .05