3.2.1. Group level analyses
We first aimed to replicate that the mean N250 amplitude to target trials increases from the first to the second part of the experiment as reported by Tanaka et al. (2006) and Sommer et al. (2021). This was indeed the case: N250 amplitude became significantly more negative from the first half of the session, M = 0.81, SD = 1.16 µV, to the second half, M = 0.38, SD = 0.98 µV, as confirmed by a one-sided Wilcoxon signed-rank test for H1 alternative (p = 0.0007). ERP waveforms were obtained by averaging over all electrodes within each ROI, per condition and each half of the experiment (see Fig. 2).
For the N250 components to the non-target faces, we asked whether these responses changed across the experiment and how this depends on face distinctiveness. A repeated measures ANOVA indicated a significant amplitude change towards more negative (less positive) values from the first to the second half of the experiment, M = 1.03 µV,SD = 1.03) vs. M = 0.77 µV, SD = 0.71,F (1,38) = 7.52, p < .05, η2 = .17 . In contrast to the P200, there was no main effect of distinctiveness in the N250, F (1,38) = 1.70, p> .05, η2 = .043, nor did it interact with experiment half, F (1,38) = .774, p > .05, η2 = .020.