This article has outlined various steps involved in the detection and validation of PAC in a visual MEG dataset (data shared at: https://doi.org/10.6084/m9.figshare.c.3819106.v1), utilizing the open-source Fieldtrip toolbox (Oostenveld et al., 2010) and customized Matlab scripts (all scripts shared at: https://github.com/neurofractal/sensory_PAC). We first confirmed that presentation of the visual grating was accompanied by decreases in alpha power (8–13 Hz) and increases in gamma power (>40 Hz) within visual area V1. Although this may seem redundant given the wealth of evidence for alpha and gamma oscillations in visual processing (Hoogenboom et al., 2006; Bonnefond and Jensen, 2015; Michalareas et al., 2016), it is crucial to establish clear increases/decreases in the power spectrum at two distinct frequencies as a first step in MEG-PAC analysis (Aru et al., 2015; Hyafil et al., 2015). Using four PAC algorithms, we showed that visual responses obtained from area V1 displayed a general increase in alpha-gamma PAC as expected (Voytek et al., 2010; Spaak et al., 2012; Bonnefond and Jensen, 2015). However, it is important to note that specific patterns of coupling depended on the algorithm selected.
