The photic feedback (PFB) enhances the alpha rhythm of brain waves using photic driving responses in which the brightness of photic signals is modulated by a subject's own alpha rhythm. While there have been some case reports in which depressive neurotic patients were treated successfully by PFB, the mechanism of effectiveness has not yet been sufficiently elucidated. The aim ot the present study was to investigate the effects of repetitive PFB sessions on electroencephalogram (EEG) and mood states of depressive as well as neurotic subjects compared with control subjects. Twenty subjects were selected from 116 healthy students in university or postgraduate school according to a neuroticism scale of the Maudsley Personality Inventory and the Center for Epidemiologic Studies Depression Scale. Ten subjects were randomly selected from students whose scores in both scales indicated more than 0.5 standard deviation over the average (High score group), and another 10 subjects were also randomly selected from the remaining students (Low score group). PFB was performed using PFB-1 (Pioneer Co, Ltd). All subjects experienced 8 PFB sessions, whose effects were compared betweem the 2 groups. EEG and mood states were measured in the 2nd and 8th sessions, and those sessions were composed of 7-minute pre-rest period, 15-minute photic stimulation period and 6-minute post-rest period. while the other sessions were composed of only pre-rest and photic stimulation periods. EEG was monitored from 6 electrodes including F3,F4,P3,P4,O1 and O2 of 10-20 EEG system. The data from pre-rest period (3-5 min), photic stimulation peniod (9-11 min, 14-16 min, 19-21 min) and post-rest period (24-26 min) were used for analysis, and the power and the average coherence were obtained in alpha 1 (8.0-10.0 Hz) and alpha 2 (10.0-13.0Hz) frequency bands based on the results of Fast Fourier Transformation. The Mood Inventory with five subscales including tension and excitement, refredhing mood, fatigue, depressive mood and anxious mood was administered at the beginning and end of each session. After excluding the data of three subjects spoiled by recording errors and thou of 2 subjects regarded as outliers by rejection analysis, those ot 15 subjects were analyzed by repeated-measures ANOVA. High score group showed higher alpha 1 power in O1 and higher alpha 1 coherence in F3-F4 and both groups showed higher alpha 1 coherence in O1-O2 in the 8th session than in the 2nd session. High score group also showed greater reduction of fatigue score in the 8th session than in the 2nd session. Then, there was a positive correlation between the increase of O1-O2 coherence and the increase of the amount of fatigue score reduction in High score group. It was suggested from these results that PFB increases alpha 1 coherence and that such EEG change in turn reduces fatigue in the depressive neurotic subjects.