The primary statistical analyses evaluated group differences in P3 amplitude at midline sites (Fz, Cz, Pz, and Oz) to novel, target, and nontarget stimuli. Preliminary analyses did not reveal evidence that group differences interacted with gender and therefore gender was not considered in subsequent analyses. Amplitudes in the P3 window (220–470 ms) were submitted to repeated-measures ANOVA with Group (depressed, control) as a between-subjects factor and two within-subjects factors: Stimulus (novel, target, nontarget) and Site (frontal, central, parietal, occipital). Given the focus of this study on the novelty P3 and target P3b, analyses of simple effects examined group differences in P3 amplitude separately for each stimulus condition. Analyses also examined group differences in the earlier N1 or N2 windows. These ANOVA used the same Group, Stimulus, and Site factors as in the above P3 analyses. A secondary analysis examining laterality differences in P3 amplitude included an additional within-subject factor of Hemisphere using medial–lateral electrode sites (F3/4; C3/4; P3/4; O1/2). An ANOVA of P3 latency at Pz included Group (depressed, control) as a between-subjects factor and stimulus (novel, target) as a
