Digital 19-channel EEG was recorded using a Nihon-Kohden (Inc., Tokyo, Japan) system with the electrodes positioned according to the International 10–20 system (i.e., Fp1, Fp2, F3, F4, C3, C4, P3, P4, O1, O2, F7, F8, T7, T8, P7, P8, Fz, Cz, Pz). The EEG activity was acquired using a linked ears reference, sampled at 500 Hz, and filtered offline between 1 and 30 Hz. Impedance was kept below 5 kΩ. Vigilance-controlled recordings were made according to usual clinical standards, including a 10-min resting eyes-closed state, 3 min of alternate 10 s-eyes open/eyes closed conditions, 3 min of hyperventilation, 1 min of recovery (post-hyperventilation), and photic stimulation. Analysis was circumscribed to the resting, awake, eyes-closed state. Brain Electrical Source Analysis (BESA) software (www.besa.de) was used for visual inspection of the EEG recordings and manual selection of samples. For each subject, 15 non-overlapping, 2-s artifact-and spike-free segments were randomly selected. We carefully avoided, particularly epochs containing ocular movements (present in up to 40% of the 10-min eyes-closed resting state recording, especially in many patients with psychosis), muscle or cardiac contamination, drowsiness signs
