HOW AGE-RELATED CHANGES IN THE CNS OF THE NORTHERN PUPILS ARE REFLECTED IN THE EEG TEMPORAL DYNAMICS OF INTEGRAL FEATURES (LONGITUDINAL STUDY)
Trifonov M.I., Rozhkov V.P.
Sechenov Institute of Evolutionary Physiology and Biochemistry of the Russian Academy of Sciences Russia, Saint-Petersburg; firstname.lastname@example.org
The functional state of an organism living in the northern regions strongly depends on various seasonal factors. It can substantially increase the variability of the EEG data masking age-related EEG changes for pupils. For this reason we study a possibility to describe age-related changes in the EEG activity of child’s brain using several parameters characterizing this activity on the whole and taking into account seasonal features. We examined a group of pupils from Arkhangelsk region from 2005 till 2012. The EEG data were recorded from every pupil 8-10 times in the different seasons. Multichannel analysis of the EEG data (16 electrodes, 20-s epochs) was based on the calculations of the normalized temporal structure function of the first order (NTSF) for EEG activity on the whole, defined as
where Sj(ti) is time series sampling the EEG process from electrode jat time ti, kt – time lag. The minimal value of bk(1) equals to 1 independently of k corresponds to the case when time series Sj(ti) from each electrodes are white noise processes. So, the analysis of bk(1) behavioral dependence on k can be used for the estimation of temporal correlation scales of the EEG activity on the whole. In the cases under our study we found that outside the zone of influence that is defined by k<11, the function bk(1) looks like as narrow-band random process where clearly expressed periodic components can be revealed and the mean value bmean(1) is practically unchanged. It was found that for all pupils bmean(1) value tends to decrease as the child becomes older. This result can be interpreted as an increasing a random (chaotic) component of the EEG activity, that to some degree is necessary for an adaptation of child’s organism in constantly changing environment. The estimation of the spectral power density of bk(1) taken for k11, showed that the dominant frequency of the EEG activity on the whole tends to increase while the child becomes older. It was shown that including the physical factors of the environment (in addition to the age of a child) in the linear regression model predicting the EEG activity in terms of the mean value of NTSF allows noticeably improving the quality of this model.
This work was supported by Russian Foundation for humanities grant № 13-06-00494а