Note that this procedure extracts the time-varying envelope amplitude of each band-pass-filtered signal. Next, the BLP signals were further filtered into this website slow (<0.1 Hz) fluctuations (two other frequency bands [0.1–1 Hz and >1 Hz] were also computed for comparison) using a second-order, zero-phase Butterworth band-pass filter. We calculated Pearson’s correlation coefficients between all possible pairs of ROIs (1) over the entire time course of the filtered BLP signals (“long epochs”) and (2) over the stable-eye epochs (“short epochs”; 135 ± 69 epochs per recording session; a total of 58 sessions). The significance of correlations was assessed using one-sample

t tests on Fisher Z-transformed coefficients. Coherence Analysis. find more We used multitaper methods

(three Slepian tapers, providing an effective taper smoothing of ± 4 Hz; Mitra and Pesaran, 1999) to calculate the coherence Cxy(f): Cxy(f)=|Sxy¯(f)|Sx¯(f)Sy¯(f),where Sx(f) and Sy(f) are the spectra of LFP time series, and Sxy(f) is the cross-spectrum. Coherence values range from zero to one, where zero coherence means that the LFPs are unrelated, and a coherence of one means that the LFPs have a constant phase relationship. We Fisher transformed coherence values and accounted for the different number of stable-eye epochs in each resting-state session according to: Cxy_t(f)=tanh−1(Cxy(f))−12m−2,where Cxy_t is the transformed coherence, and m is the product of K and the number of stable-eye epochs ( Bokil et al., 2007). We rejected the null hypothesis of no significant coherence between two ROIs only when the coherence was above zero (based on jackknife estimates of the variance) across a frequency range greater than the bandwidth (i.e., 8 Hz), to account

for multiple comparisons ( Bokil et al., 2007). Cross-Frequency Coupling. We measured cross-frequency coupling between low-frequency oscillations and gamma power using the SI ( Cohen, 2008). There were two reasons for using this measure: (1) the SI can be reliably computed on the short stable-eye Dichloromethane dehalogenase epochs examined in our study; and (2) the SI can capture dynamic changes in cross-frequency coupling. There were three processing steps to calculate the SI. First, we extracted gamma power time series for given frequency bands whose central frequency ranged from 30 to 100 Hz, stepped in 5 Hz increments, with a bandwidth of ± 5 Hz. Second, for each of the theta, alpha, low-beta (13–20 Hz), and high-beta (20–30 Hz) bands, we identified the low frequency with which the gamma power time series might synchronize. (The aim here was to identify the dominant frequency at which the gamma power time series oscillated.) Third, we identified the peak of the power of the gamma frequency envelope time series, extracted the phase time series from both the gamma- and low-frequency bands (low-frequency bandwidth ± 1.