A second and related issue concerns the potential confound of muscle artifacts in high frequency abnormalities observed in our clinical populations. While muscle and ocular artifacts have been shown to bias high frequency activity (e.g., Whitham et al. 2007; Yuval-Greenberg et al. 2008; Shackman et al. 2009), this explanation seems unlikely in our data for several reasons. For one, ICA has been shown to effectively eliminate ocular artifacts that may have been overlooked in the traditional threshold-based artifact rejection procedure especially in the low and high frequencies (Keren et al. 2010; Shackman et al. 2010). This step would thus minimize residual effects of artifacts on our EEG variables. Further, if artifacts were the cause of high frequency abnormalities, they would be more likely to manifest in SZ since this clinical population has been associated with higher levels of ocular artifacts (Chan and Chen 2004; Chan et al. 2010). However, we observed greater beta power in BP instead. Collectively, that muscle artifacts alone led to the high frequency abnormalities observed in our sample appears to be an unlikely explanation.
