Abstract
This study evaluates the mental fatigue induced during VDT tasks using event-related potential (P 300). Since the waveform peak is not clear in the measurement of P 300, there exists a case when it is difficult to detect the amplitude and the latency. The removal of the noisy waveform based on the cross correlation between the grand average and each waveform was found to be effective for peak detection. Mental fatigue was evaluated by parameter extraction methods using a principal component analysis or temporal changes of cross correlation between the grand average and each waveform. As a result, the P 300 b component and the standard deviation of the time lag between the grand average and each waveform obtained using cross correlations were found to be effective indices that reflect mental fatigue (decrease in the cognitive information processing function).
