Nonlinearity in normal human EEG: cycles, temporal asymmetry, nonstationarity and randomness, not chaos

Abstract

Two-hour vigilance and sleep electroencephalogram (EEG) recordings from five healthy volunteers were analyzed using a method for identifying nonlinearity and chaos which combines the redundancy-linear redundancy approach with the surrogate data technique. A nonlinear component in the EEG was detected, however, inconsistent with the hypothesis of low-dimensional chaos. A possibility that a temporally asymmetric process may underlie or influence the EEG dynamics was indicated. A process that merges nonstationary nonlinear deterministic oscillations with randomness is proposed for an explanation of observed properties of the analyzed EEG signals. Taking these results into consideration, the use of dimensional and related chaos-based algorithms in quantitative EEG analysis is critically discussed.

Frequency and dependence of long range temporal correlations in human hippocampal energy fluctuations

“…The data were analyzed in four forms: unfiltered (other than by the analog filter), low-frequency (0.5-30 Hz), high-frequency (30 -100 Hz), and the standard clinical EEG bands, ␦ (0.5-4 Hz), (4)(5)(6)(7)(8), ␣ (8 -12 Hz), ␤ (12-30), and ␥ (30 -70 Hz). The offline filtering was performed with a wavelet filtering scheme.…”

“…Hippocampal EEG activity, as measured by depth electrode recordings, is generated by local populations of synchronously firing neurons and when examined over sufficiently long time scales (approximately minutes) exhibits a wide range of amplitude and temporal variation that have received little attention [5]. Unfortunately, most quantitative methods for studying the temporal dynamics of the EEG, such as spectral analysis and nonlinear dynamics [6,7] require signal stationarity. Although the energy of human hippocampal EEG may remain nearly constant for minutes at a time, the mean energy and variance can also fluctuates widely [8], limiting the usefulness of standard spectral and nonlinear dynamics methods for investigating correlations over long time scales.…”

The role of nonlinearity in computing graph-theoretical properties of resting-state functional magnetic resonance imaging brain networks

“…13 The equiquantal mutual information estimator has been successfully used in detection and quantification of nonlinearity in the dynamics of various complex systems, ranging from the Earth's atmosphere 14 to the human heart 15 and brain. [16][17][18][19] In neuroscience, functional magnetic resonance imaging (fMRI) is one of the prominent methods for the study of large-scale brain dynamics. Following the increase of interest in the study of complex network properties, a wealth of graph-theoretical studies utilizing resting-state fMRI data has been published in the last 5 yr. 20 Importantly, linear connectivity measures such as Pearson or partial linear correlation of the local activity time series are used to derive the connectivity matrix that is then transformed to the network representation.…”

Channel Division Based Multiple Classifiers Fusion for Emotion Recognition Using EEG signals

“…However, EEG signals have low signal-to-noise ratio (SNR) and often mixed with much noise when collected. The more challenge problem is that, unlike image or speech signals, EEG signals are temporal asymmetry and nonstationary because the human brain can be seen as a complicated nonlinear dynamic system [11]. So these EEG-based emotion recognition methods using a simple feature integration or fusion are difficult to obtain stable recognition performance and even bring some negative effects.…”