All statistical analyses were performed using SPSS (SPSS Inc., Chicago, IL). The measures submitted to statistical analysis were (1) evoked theta power (log-transformed) averaged over electrodes Pz and CPz and (2) induced theta activity (theta ERS) averaged over electrodes FCz and Cz. These electrodes were those within which each of the measures was found to be maximal, both in the current study and in previous reports (e.g., Jones et al. 2006; Andrew and Fein 2010a,b; Gilmore and Fein 2012), and thus were considered those best characterizing each of the measures. The evoked and induced theta measures, and ITC (also averaged over FCz and Fz), were submitted, separately, to univariate analyses of variance (ANOVAs) with between-subjects factor group (NAC, LTAA, and STAA). Given our a priori hypotheses (that for evoked theta, power would be reduced to the same degree in both STAA and LTAA compared with NAC; for induced theta, the magnitude of the theta ERS would be greater in LTAA vs. NAC, and greater in STAA compared with both LTAA and NAC), pairwise comparisons between each group within the evoked
