This study recruited 20 patients with MDD (11 women), 18 patients with schizophrenia (9 women), and 16 healthy controls (8 women). All participants were native Koreans. Inclusion criteria of all participants were as follows: (1) age ranged 19 to 65 years; (2) in case of patients, met the requirements of the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-4); (3) normal vision or hearing. Participants and patients with (1) vision or hearing problems, (2) drug and/or alcohol abuse, (3) traumatic brain injury, and (3) a lifetime history of neurological disorders were excluded. Furthermore, healthy subjects with a lifetime history of psychiatric disorders were excluded. Patients and healthy individuals were diagnosed based on the Structured Clinical Interview using the MINI International Neuropsychiatric Interview in the DSM-4. The MINI, a clinician-administered structured interview, was designed to measure anxiety, mood, eating, substance use, and psychotic disorders. According to DSM-4 criteria, and patients with MDD and schizophrenia were diagnosed. Clinical symptoms were evaluated by a trained psychiatrist. Hamilton Depression and Anxiety [31, 32] rating scales were evaluated in patients with MDD. Positive and Negative Syndrome Scales were evaluated in patients with schizophrenia. Healthy participants were recruited through public advertising in Seoul, Korea. The mean (± SD) age of all participants was 37.63 ± 11.38 years (range, 19–59 years). The present study was conducted in compliance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea (approval number KC14DDSE0479). All participants provided written informed consent. All experimental procedures followed relevant institutional guidelines and regulations.
Electrophysiological Measurement And Analysis
Participants were seated in a comfortable chair in a sound-attenuated room. EEG data were recorded using an amplifier (NeuroScan SynAmps Compumedics USA, El Paso, TX, USA) with a headcap equipped with AgCl electrodes according to the international 10–20 system. We used a EEG device that records from 62 scalp positions—15 standard channels (FP1, FP2, F7, F3, FZ, F4, F8, C3, CZ, C4, P3, PZ, P4, O1, and O2) and 47 extended channels (FPZ, AF3, AF4, F5, F1, F2, F6, FT7, FC5, FC3, FC1, FCZ, FC2, FC4, FC6, FT8, T7, C5, C1, C2, C6, T8, TP7, CP5, CP3, CP1, CPZ, CP2, CP4, CP6, TP8, P7, P5, P1, P2, P6, P8, PO7, PO5, PO3, POZ, PO4, PO6, PO8, CB1, OZ, and CB2). Additional electrodes were placed above and below the left eye for vertical electro-oculography (VEO) and at the outer canthus of each eye for horizontal electro-oculography. EEG data were recorded using a 0.1–100 Hz bandpass filter at a sampling rate of 1,000 Hz. The signals were referenced to both mastoids, and the ground electrode was placed on the forehead. The impedance between the electrodes and the scalp was maintained below 5 kΩ during the entire recording session. Subsequently, the EEG data were preprocessed using Scan 4.5 software, Curry 7.0 (Compumedics USA, El Paso, TX, USA). Gross artifacts were rejected through visual inspection of the recording by a trained individual who had no previous information regarding the data origin.
Resting State Eeg Paradigm And Alpha Asymmetry Calculation
Resting EEG was recorded with eyes open and closed for 5 min each. Eye blinking artifacts can have an undesirable effect on EEG band power, and therefore were corrected using established mathematical procedures [34, 35]. Additionally, based on VEO, positive and negative components exceeding 300 µV from before and after a maximum peak of blinking interval (-100 ms to 300 ms) in the frontal regions were considered covariant. Data were re-analyzed using Matlab 2016 software (Mathworks, Inc, Natick, MA, USA), including a fast Fourier transform with a 1–50 Hz bandpass filter to calculate the absolute power: delta (1.0 to 4.0 Hz), theta (4.0 to 8.0 Hz), alpha (8.0 to 12.0 Hz), beta (12.0 to 30.0 Hz), and gamma (30.0 to 50.0 Hz) signals. The power values were displayed as averaged points in the frequency range. Artifacts exceeding ± 100 µV were rejected at all electrode sites. For each participant, 30 randomized artifact-free epochs (epoch length 2.048 s) were used in the analysis. The F4 and F3 electrodes covered the middle-frontal scalp region, while the F8 and F7 electrodes covered the lateral-frontal scalp areas, both of which are associated with frontal alpha asymmetry for depressive disorder (Fig. 1-a) . To normalize the FAA data, a common log transformation was applied to the power values of selected electrodes . FAA has been defined as hemispheric differences , which were calculated as the difference between selected electrodes, right frontal alpha power, and left frontal alpha power.
FAA = “log10 F4 - log10 F3” and “log10 F8 - log10 F7”
Demographic statistics with age and sex between participant groups were tested using analysis of variance (ANOVA) and chi-squared tests. A comparison of alpha asymmetry was performed using multivariate analysis of covariance. Within-subject factors included alpha asymmetry values (log-transformed F4–F3 and F8–F7) with eyes open and closed. The groups constituted the between-subject factors. Age and sex were considered as covariates. Partial correlations between alpha asymmetry and clinical symptoms were analyzed to account age and sex. Bootstrapping tests were performed in the correlation analysis, and the sampling number was 10,000, which has been accepted in previous studies [38–40]. Alpha asymmetry between men and women was compared and analyzed using ANOVA. p-values were corrected using the Bonferroni method, which is applied to multiple comparisons of several experimental conditions and variables [41, 42].