One major issue in cross-frequency coupling analysis is the presence of non-sinusoidal sawtooth-like oscillations (Jensen et al., 2016; Cole et al., 2017), which can result in spurious estimates of PAC (Lozano-Soldevilla et al., 2016). This property of oscillations can be quantified by calculating the time taken from trough to peak (rise-time), peak to trough (decay-time), and the ratio between these values (Dvorak and Fenton, 2014; Cole and Voytek, 2017). We therefore calculated this ratio for the visual V1 data from 7 to 13 Hz, and performed a t-test to check for differences in non-sinusoidal oscillations between grating and baseline periods (script: 6_check_non_sinusoidal.m).
