For accurate inverse solutions, eLORETA was modeled for the S-RS and M-RS separately, from which whole-brain source estimates of α (8–12 Hz) power were derived (Pascual-Marqui et al., 2011). For source-based α-frequency connectivity analysis, time series of regions of interest (ROIs) were derived from eLORETA (Samogin et al., 2019), which were then submitted to Granger causality analysis based on bivariate autoregressive (AR) modeling (Ding et al., 2006). A model order of 20 (80 ms in time for a sampling rate of 250 Hz) was chosen in a two-step process: (1) Akaike Information Criterion (AIC) and (2) comparing spectral estimates obtained by the Fourier-based AR model for data pooled across all subjects (Wang et al., 2016). Our focus on the α frequency was guided by a priori hypotheses stemming from (1) the purported role of α oscillations in the sensory disinhibition model of PTSD and (2) prior work by Clancy et al. (2017), demonstrating no surface-level effects of PTSD-related aberrations in neighboring θ or β frequencies (ps > 0.16).
