Nishida S, Nakamura M, Suwazono S, Honda M, Shibasaki H
Department of Communication and Computer Engineering, Fukuoka Institute of Technology, Japan.
Electroencephalogr Clin Neurophysiol. 1997 Sep;104(5):431-6. doi: 10.1016/s0168-5597(97)00048-8.
Among single sweep records of event-related potentials (ERPs), the peak latency of P300, which is one of the most prominent positive peaks in the ERP obtained in the oddball paradigm, may vary depending on the conditions of the subject. In the analysis of characteristics of the variability in the peak latency, it is important to know to what extent the variability of the measured peak latency (measured variability) is actually caused by physiological factors (physiological variability). In our previous study, a method was developed for judging whether the physiological variability really exists or not, and if it does exist, the developed method extracts the physiological variability from the measured variability based on a limited number of single sweep records. In the present study, based on the P300 waveforms which were detected by blinded visual inspection of the ERP data obtained by an auditory oddball paradigm from 12 healthy adults, the physiological variability was shown to exist with a confidence level of 95% for all subjects. Furthermore, its interval estimate was calculated by subtracting noise variability from the measured variability with a confidence level of 80%, and it was found to range from 17 to 57 ms for all subjects.
