抄録
EEG related to time dependent varying tasks are signals of non-stational processes, and ordinal information criteria for which stational processes are assumed can not be applied directly. In this paper, we propose an analysis method for EEG signals related to time dependent varying tasks using an information criterion which has been published by one of the authors. By using the method, time dependent varying information amount of the time dependent varying EEG can be evaluated related to a time dependent varying task. A task performed as an example was a color object image in which the color varies continuously depending on the time axis starting from a natural color to a strange color on the object. The total time of the variation was 20 seconds. The time dependent varying color image was displayed on a CRT display. The subjects were five graduate students. For measurements of EEG at observations of the image, an electrode cap with 19 electrodes has been mounted on the subjects. For each electrode, information amount analysis was performed on the time axis. The averaged result of 19 electrodes show that the information amount of alpha band decreases monotonously, and theta band increases monotonously with progress of time. It is supposed that the variation of the information amount in the example indicates activated interpretations related to the strange color on the object with progress of time. Though, EEG is taken up as an example in this paper, the method can be applied also for other signals observed from human body. [J Physiol Sci. 2007;57 Suppl:S71]
