Sie sind auf Seite 1von 6

Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)

Coupling functions between brain waves:


Significance of opened/closed eyes
Lal Hussain1, Wajid Aziz1,2, Sharjil Saeed1
1
QEC and Department of CS & IT, University of Azad Jammu and Kashmir, AJK, Pakistan
2
Department of CS & IT, University of Azad Jammu and Kashmir, AJK, Pakistan
Abstract

In dynamical systems the information flows converges or diverges in state space and is integrated or
communicated between different cells assemblies termed as CFC. This process allows different oscillatory systems to
communicate in accurate time, control and distribute the information flows in cell assemblies. The CF interactions
allow the oscillatory rhythms to communicate in accurate time, and reintegrate the separated information. The intrinsic
brain dynamics in Electroencephalography (EEG) with eye - closed (EC) and eye open (EO) during resting states have
been investigated to see the changes in brain complexity i.e. simple visual processing which are associated with increase
in global dimension complexity. In order to study these changes in EEG, we have computed the coupling to see the
inhibitory interneurons response and inter-regions functional connectivity differences between the eye conditions. We
have investigated the fluctuations in EEG activities in low (delta, theta) and high (alpha) frequency brain oscillations.
Coupling strength was estimated using Dynamic Bayesian inference approach which can effectively detect the phase
connectivity subject to the noise within a network of time varying coupled phase oscillators. Using this approach, we
have seen that delta-alpha and theta-alpha CFC are more dominant in resting state EEG and applicable to multivariate
network oscillator. It shows that alpha phase was dominated by low frequency oscillations i.e. delta and theta. These
different CFC help us to investigate complex neuronal brain dynamics at large scale networks. We observed the local
interactions at high frequencies and global interactions at low frequencies. The alpha oscillations are generated from
both posterior and anterior origins whereas the delta oscillations found at posterior regions.

Key Words: Electroencephalography (EEG) during resting state, Cross Frequency Coupling, Dynamic
Bayesian Inference, Wavelet Transform

--------------------------------------------------------------------***----------------------------------------------------------------------

1. Introduction was observed by [6] which disseminate the activity in cortical


Cross Frequency Coupling is used to investigate the area showing effective mechanism for communication during
intrinsic dynamics in EEG over different frequencies bands. cognitive process in human. The studies also reveal that cross
[2] reported different cross frequencies coupling relations frequency coupling where the frequency band of one
such as amplitude to amplitude, phase to phase and phase to oscillation modulates the activity in different frequency bands
amplitude at different frequencies. [3] observed that are more ample in animal than in human. [7, 8, 9] reported
oscillatory activates in human and animals modulated in that spike timing of single neuron and firing rate can be
various frequency bands and found that fast gamma modulated by theta rhythm and intracortical local field
oscillation (30-150 Hz) is modulated by the slow theta potential by the gamma power as reported by [4,7, 10].
oscillation (5-8 Hz). The intracranial electrical recordings Moreover [11, 12, 13] observed the changes in theta powers
measured in human such as EEG, EMG are reflected by the during task related activities in the human. [14] have detected
oscillatory electrophysiological signals. The theta and beta the cross frequency coupling at frequencies up to 40 Hz at
both bands are modulated during memory and perception scalp. [1] also reported that classical power spectral analysis
tasks. [4] reported that during neuronal delta phase (1-4 Hz) based on Fast Fourier Transform (FFT) or different time-
have modulated the theta amplitude (4-10 Hz) and theta phase frequency transforms (e.g. Hilbert, Wavelet, or Gabor
has modulated gamma (30-50 Hz) amplitude thereby transform) unable to identify the relationships among
controlling the baseline excitability through oscillatory different frequencies or frequency component as merely
hierarchy resulting stimulus related response in the neuronal because these techniques are based on amplitude modulation
ensemble. These studies have been reported in monkeys that with a set of defined frequencies across time whereas
theta phase and gamma power interactions are found in complex transformation in the complex signals like EEG,
auditory cortex during both stimulus driven activity and EMG containing different frequency components interacting
auditory cortex. A theta- gamma interaction in continuous with each other require a corresponding complex
word recognition memory task in human medical temporal transformation of the signal to provide information about the
lobe was observed by [5]. A robust high and low frequency phase changes. Thus cross-frequency coupling (CFC)
bands coupling in the human brain ongoing electrical activity technique is hypothesized between different frequency bands
which serves as local as well as global interaction among the
275
Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)

processes and related directly to the integrated distributed Using Hilbert transform, we obtained protophases
information. are transformed to true phases [37] which grows uniformly
Cross-frequency coupling (CFC) can be in time. We aimed to check the functional influence of delta-
investigated using different approaches i.e. using PSI for alpha, alpha-gamma and theta-gamma waves for each case
phase-amplitude coupling as described by [18] and i.e. which wave influence and dominate the other wave. The
correlation coefficients between different frequency delta waves are deep sleep in adult and waking in young
components i.e. phase-phase CFC (or n:m phase adults and gamma waves are associated with attention,
synchronization) as described by [15, 17, 20]. [20] used first memory and sensory processing. Keeping the property of
time the n:m phase synchronization technique based on phase phase-dynamics [21, 35] we can build the system of
synchronization of chaotic oscillator to evaluate the temporal stochastic differential equation.
evolution of the coordinated peripheral tremor activity 
affected directly by the time course of strength of 
synchronization in neural network involving cortical motor 
areas.
Where are the frequencies of oscillators, ,
To observe the evolution of natural processes in time
and space scientists have used the systematic observations to coupling functions, represents the corresponding
analyze the nature. Typically, every one aimed to generate angles and represents the stochastic part modelled as
models based on collected data over specified time interval white and Gaussian noise where the natural frequency is
and its dependency that how the faster processes be whether affected by the coupling influence of each corresponding
seconds or milliseconds. In order to analyze the data, different oscillator.
methods of data analysis be employed as suggested by [20, The differential equation 2) can be decomposed into Fourier
22, 23, 24, 32] depending on the characteristic of data may be components as [47].

for example Bayesian methods, particle filters, Kalan filters,
maximum likelihood estimators etc. The Bayesian methods    
are preferably used to detect the dynamics in the systems not
isolated but may be influenced by environment and other 2.2. DYNAMICAL BAYESIAN INFERENCE
weekly coupled processes [25, 25, 27]. Recently, [20, 28] The cross frequency coupling from phase oscillators
used the Bayesian method to identify the time-varying is to be inferred from Bayesian approach [20, 21]. Consider a
dynamics even in presence of noise because of its ability to time series  , where
detect the dynamics in presence of noise and follow the time m=1,2,3,..,M. from which phase dynamics are extracted
evolution of the parameters. The dynamical Bayesian (Luchinsky et a., 2008; 73]. Using Dynamic Bayesian
inference method can be applied to various types of inference we aimed to compute a set of model parameters

dynamical systems including coupled oscillatory dynamical where denote the coupling and
systems e.g. neuronal systems or cardio-respiratory [1, 19, denote the noise matrix. Thus using Bayes theorem, we can
29] which are time varying dynamical systems subject to infer the unknown model parameter M from X by calculating
external noise. The method can also be applied for detecting posterior density , given a prior density
causal interactions, directionality of influence [16, 30, 31], that encompass the previous knowledge of unknown
synchronization and coherence [15, 21, 33, 34] and coupling parameters based on observations and likelihood function
functions [23, 32, 35]. We here aimed to detect the cross i.e. the conditional probability density to observe X
frequency coupling and coherence between Eye-closed and given choice of unknown dynamical model M :
open subjects during resting states at 19 electrodes according 
to 10-20 international system. Dynamical Bayesian inference 
method) already used by [36] to investigate the 

cardiorespiratory coupling functions affected by aging. For high sampling frequency i.e. small sampling step h, the
phase dynamics in eq. 1can be well approximated using Euler
2. Methods midpoint discretization as:


2.1. COUPLING FUNCTION
To understand the interaction between two .

unidirectional coupled oscillators, consider an example of  
two coupled oscillators [35] with phases .  
  Where is the stochastic integral of the noise term
   over time i.e.

Where are the autonomous, natural frequencies of first
and second oscillators and is a coupling function added to 

the frequency of second oscillator. This coupling function Here H is the Cholesky decomposition of noise matrix N and
will influence the natural frequency of second oscillator is normally distributed random variable vector. The joint
depending on its magnitude values either accelerating or probability density of is employed to compute the joint
decelerating the second oscillator. The physics literature probability density of process in respect of
reveals this coupling as interaction between oscillators
whereas the biology literature describes it as correlation  by imposing , where
between them e.g. due to phase locking (Pikovsky et al.,
1997).
276
Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)


is the Jacobin term of transformed variables calculated are coupled between different brain areas. Likewise, same
from base function . location and cross frequency coupling can be served as
Thus the negative log-likelihood function potential mechanism in regulating communication in
 different spatio-temporal scales. The phase-phase CFC were
also investigated by [44, 45] that serves as physiological
mechanism to link the activities occurring significantly at

different rates e.g. in NREM sleep, the firing rate patterns

correlations are observed during learning with a rate of six to

seven times faster. [8] investigated that different frequencies
 interactions provide means to understand the complex neural
By considering prior probability of parameter M as dynamics in the frequency specific neural networks. The
multivariate normal distribution, and considering the processes in the brain could be efficiently integrated with
quadratic form of log-likelihood (6), thus posterior neuronal cell assemblies that are oscillating with different
probability will also be multivariate normal distribution. frequency synchronously. These phenomena can provide
Thus parameter c with particular distribution with mean and enhanced combinational opportunities to store the complex
covariance matrix , the stationary point S can temporal patterns and to optimize the synaptic weights used
be recursively computed using the below equations: in conjunction with relevant algorithms. [6] investigated that
cross - frequency coupling in the present research plays a vital
role in cognitive learning, neuronal computation and

communication.
We have computed phase-phase CFC in lower and


higher frequency bands for 19 electrodes. Figure 1 to 2 depict
 the CFC using lower frequency CFCs delta-alpha, delta-beta,

theta- alpha and higher CFCs theta-beta, theta-gamma and
alpha-gamma. The strongest CFC was observed between and

within delta and alpha, delta and theta frequencies using both



temporal and spatial scales. These different CFCs show that

  there are different neuronal interactions at work. In EC delta-
2.3. QUANTIFICATION OF COUPLING alpha the highest coupling was obtained at C4 electrode,
From Dynamic Bayesian inference, from inferred delta-alpha the strongest coupling was found in central
parameter c can be used to quantify certain characteristics of regions i.e. C3, C4 and Cz and right parietal P8. Similarly, in
coupling function either by computing quantitative measures delta-theta the strongest coupling was found at frontal Fp1
or comparing different coupling mechanisms. and Fp2 probes. Moreover, in theta-beta CFC during EC
Coupling strength coupling was found stronger in parietal P7 and Pz and left
For evaluating phase-phase cross frequency temporal T7. In theta-gamma coupling during EC it was
coupling, one need to determine the coupling strength found stronger in Fp1 and O2 whereas in alpha-gamma it was
inferred from parameter c estimated using Bayesian dynamics found stronger in occipital are O1 and O2 during EC.
inference. This coupling strength corresponds to the coupling Likewise, during EO delta-alpha coupling was
amplitude extracted from phase dynamics of the time series. found stronger in Cz, in delta-theta CFC, it was found
Here we aim to compute the cross frequency coupling of stronger in frontal probes such as F3, F4, Fz, F7 and F8 and
delta-alpha, alpha-gamma, theta-gamma frequency bands right temporal probe T8. In theta-alpha coupling was found
from 16 subjects of EEG with eye closed and eye open during higher in O1, while in theta-beta it was found stronger in Fz
resting states for selected frontal electrodes Fp1 and Fp2. and F8 whereas in theta-gamma this coupling was highest in
Mathematically, the coupling strength is computed using the T8 and O2 and in alpha-gamma the highest coupling was
Euclidean norm of inferred parameters that corresponds to found in T7, T8 and O2.
Fourier components of the coupling to oscillator from The significance map in Figure 1 to 2 denotes the
combination of oscillators i.e. statistically significance p-values computed using paired t-
test. The strongest significance was found in delta-alpha and

 theta-alpha CFC in most of the electrodes in each 19 chosen
electrodes. In delta-alpha CFC the very significant results are
For one oscillator, there are and its coupling obtained at electrode Cz, the electrodes F3, F7, Fp1, Fz, O1
strength is indexed into part out of , similarly and O2 also exhibit just significant results while other
two oscillators are composed of  and electrodes show no significant results. Thus left frontal and
occipital shows the significant results with EO>EC and
coupling strength is indexed into part of vector and
central Cz also shows the highest significance. In theta-alpha
accordingly same happen for more than two oscillators.
CFC the statistically very significant results are obtained at
3. Results electrodes F4 and T8 while the electrodes F3, F7, Fp1, O1,
The phase synchronization (phase-phase CFC or P4, Pz and P7 also showed only significant results (p<0.05).
amplitude amplitude CFC) plays vital role in the The electrodes C4, O2 and P8 show almost significant results.
functioning of brain in different ways. The regulations in the Thus the left frontal, occipital, parietal regions and right
inter-area communication was extensively studied by [39, 40, frontal F4, temporal T8 and parietal P4 regions shows the
41, 42, 43] where cross location and same frequency phases significant results in theta-alpha CFC.
277
Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)

In this study we computed different discussed in


below section:
Delta/theta-alpha CFC
Isleretal (2008) investigated the widespread CF
phase coupling of delta-theta and delta-alpha rhythms where
the delta oscillations phases are coupled to the localized theta
and alpha oscillations in different regions central, posterior,
right parietal regions etc. [1] in his study found that there
exists delta-alpha (phase-phase and phase-amplitude) CFC in
EEG resting state data. The coupling was seen within the
parieto-occipital and frontal regions and these regions are
connected via large scale couplings networks which show
that there is a mechanism of direct communication in these
regions in different cell assemblies. This coupling was found
asymmetric from posterior-occipital to frontal regions
especially during EC condition while during EO condition
during rest this delta-alpha coupling can be found inverse.
Theta/Alpha-Gamma CFC
There are many theories that are describing the
interplay between different frequency bands. They slow
oscillations such as theta or alpha due to low frequency are
considered to serve the network over long distances while the
fast oscillations such as gamma due to high frequency are
considered to synchronize the cell assemblies over short
spatial scales [3]. [6] studied the theta/alpha-gamma and
observed that phase-amplitude CFC vary across brain areas
in different manners such as task-relevant manner and
quickly changes in response to motor, sensory and cognitive
events and also linked by the learning tasks performance.
Figure 1: Delta-Alpha CFC EC, EO and Significance using paired Moreover, this CFC was observed as distinct in brain
test rhythms, however it varies across function of task demands
and cortical areas. For example, during auditory tasks, the
theta-gamma CFC was higher than alpha-gamma at the
anterior sites and equal across cortex, however, posterior
alpha-gamma was observed greater than the anterior alpha-
gamma CFC. Likewise, in visual tasks the alpha-gamma CFC
was observed at posterior electrodes and found greater than
theta-gamma CFC. Moreover, the theta-alpha phase
detection coupling was recently studied in the frequency
ranges (4-13 Hz) by [46, 47, 48].
At delta-alpha to alpha and theta-alpha to alpha
CFC, the Coupling in EC condition is stronger in most cases
than in EO condition and only few connections which are
found stronger in EO than in EC. In delta-alpha to alpha CFC,
stronger connection in EC are centro-occipital, temporo-
central few parietal and frontal; in theta-alpha few stronger
connection are found such as frontal, fronto-central, fronto-
occipital, fronto-parietal, occipito-central. [1] also
investigated CFC and found that in delta-alpha CFC showed
large scale connections going from anterior to posterior
where delta modulate alpha and using cross Bispectrum
(cBIS) an inverse CFC modulation alpha to delta was
observed. Where different delta-alpha or alpha-delta CFC
directions are dependent upon the location of generation of
delta and alpha generator. Previous studies [50, 51, 52] also
revealed that alpha oscillations have posterior and anterior
origins whereas delta oscillations have an anterior origin.
Figure 2: Theta-Alpha CFC EC, EO and Significance using paired Using both coupling directions and frequency generation
test origins, one might be able to investigate different types of
CFC at a larger scale particularly when alpha oscillations are
4. Discussion and Conclusion involved. Using coupling we also observed large scale
coupling with alpha oscillation dominant from posterior to
278
Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)

anterior direction which shows the influence of alpha [12] Sederberg, P. B., Kahana, M. J., Howard, M. W.,
frequency generator on other brain origins. This coupling was Donner, E. J., & Madsen, J. R. (2003). Theta and
found large during EC may be due to inhibition as reported in gamma oscillations during encoding predict
the literature [49]. subsequent recall. The Journal of Neuroscience,
23(34), 10809-10814.
Acknowledgement [13] Onton, J., Delorme, A., & Makeig, S. (2005). Frontal
I express my deepest gratitude to Dr. Aneta midline EEG dynamics during working
Stefanovska, Professor of Biomedical Physics, Physics memory.Neuroimage, 27(2), 341-356.
Department, Lancaster University, UK for her invaluable [14] Palva, J. M., Palva, S., & Kaila, K. (2005). Phase
assistance and discussions in her research group and Prof. synchrony among neuronal oscillations in the human
Peter McClintock for his general support in accomplishing cortex. The Journal of Neuroscience, 25(15), 3962-3972.
the work. I am very thankful to Dr. Tomislav Stankovski for [15] Tass, P., Rosenblum, M. G., Weule, J., Kurths, J.,
his discussions, ideas and technical support to use novel Pikovsky, A., Volkmann, J., & Freund, H. J.
techniques for EEG data analysis. A special thank goes to Ms. (1998). Detection of n: m phase locking from
noisy data: application to magnetoencephalography.
Valentina Ticcinelli, PhD Scholar, Lancaster University for Physical review letters, 81(15), 3291.
her pragmatic approach in assisting few robust methods in [16] Rosenblum, M., Pikovsky, A., Kurths, J.,
Matlab to evaluate really complex nature of EEG signals. Schfer, C., &Tass, P. A. (2001). Phase
REFERENCES synchronization: from theory to data analysis.
Handbook of biological physics, 4, 279-321.
[1] Jirsa, V., & Mller, V. (2013). Cross-frequency coupling [17] Witte, H., Putsche, P., Hemmelmann, C.,
in real and virtual brain networks. Frontiers in Schelenz, C., & Leistritz, L. (2008).Analysis and
computational neuroscience, 7. modeling of time-variant amplitude frequency
[2] Knyazev, G. G. (2011). Cross-frequency couplings of and between oscillations of EEG
coupling of brain oscillations: an impact of state bursts. Biological Cybernetics, 99(2), 139-157.
anxiety. International Journal of [18] Cohen, M. X. (2008). Assessing transient cross-
Psychophysiology, 80(3), 236-245. frequency coupling in EEG data. Journal of
[3] Jensen, O., & Colgin, L. L. (2007).Cross- neuroscience methods, 168(2), 494-499.
frequency coupling between neuronal oscillations. [19] Shiogai, Y., Stefanovska, A., & McClintock, P. V. E.
Trends in cognitive sciences, 11(7), 267-269. (2010).Nonlinear dynamics of cardiovascular
[4] Lakatos, P., Shah, A. S., Knuth, K. H., Ulbert, I., ageing. Physics reports, 488(2), 51-110.
Karmos, G., & Schroeder, C. E. (2005).An [20] Stankovski, T., Duggento, A., McClintock, P. V.,
oscillatory hierarchy controlling neuronal &Stefanovska, A. (2012). Inference of time-
excitability and stimulus processing in the evolving coupled dynamical systems in the presence of
auditory cortex.Journal of neurophysiology, 94(3), noise. Physical review letters, 109(2), 024101.
1904-1911. [21] Stankovski, T., Duggento, A., McClintock, P. V.,
[5] Mormann, F., Fell, J., Axmacher, N., Weber, B., &Stefanovska, A. (2014).A tutorial on time-
Lehnertz, K., Elger, C. E., & Fernndez, G. (2005). evolving dynamical Bayesian inference. The
Phase/amplitude reset and theta gamma European Physical Journal Special Topics,
interaction in the human medial temporal lobe 223(13), 2685-2703.
during a continuous word recognition memory [22] Smirnov, D. A., &Bezruchko, B. P. (2009).
task.Hippocampus, 15(7), 890-900. Detection of couplings in ensembles of stochastic
[6] Canolty, R. T., Edwards, E., Dalal, S. S., oscillators. Physical Review E, 79(4), 046204.
Soltani, M., Nagarajan, S. S., Kirsch, H. E., & [23] Kiss, I. Z., Zhai, Y., & Hudson, J. L. (2005).
Knight, R. T. (2006). High gamma power Predicting mutual entrainment of oscillators with
is phase-locked to theta oscillations in human experiment-based phase models. Physical review
neocortex. science, 313(5793), 1626-1628. letters, 94(24), 248301.
[7] Bragin, A., Engel, J., Wilson, C. L., Fried, I., & [24] Arulampalam, M. S., Maskell, S., Gordon, N., &
Mathern, G. W. (1999).Hippocampal and Clapp, T. (2002). A tutorial on particle filters for online
entorhinal cortex highfrequency oscillations (100 nonlinear/non-Gaussian Bayesian tracking. Signal
500 Hz) in human epileptic brain and in kainic Processing, IEEE Transactions on,50(2), 174-188.
acidtreated rats with chronic seizures.Epilepsia, 40(2), [25] Haken, H. (1975). Analogy between higher
127-137. instabilities in fluids and lasers. Physics Letters
[8] Buzski, G., Buhl, D. L., Harris, K. D., Csicsvari, J., A, 53(1),77-78.
Czeh, B., &Morozov, A. (2003). Hippocampal [26] Suprunenko, Y. F., Clemson, P. T., &Stefanovska,
network patterns of activity in the mouse. A. (2013).Chronotaxic systems: a new class of
Neuroscience, 116(1), 201-211. self-sustained nonautonomous oscillators. Physical
[9] Siapas, A. G., Lubenov, E. V., & Wilson, M. review letters, 111(2), 024101.
(2005).Prefrontal phase locking to hippocampal [27] Clemson, P. T., Suprunenko, Y. F., Stankovski, T.,
theta oscillations. Neuron, 46(1), 141-151. &Stefanovska, A. (2014).Inverse approach to
[10] Chrobak, J. J., & Buzski, G. (1998). Gamma chronotaxic systems for single-variable time series.
oscillations in the entorhinal cortex of the freely Physical Review E, 89(3), 032904.
behaving rat. The Journal of neuroscience, 18(1), 388- [28] Duggento, A., Stankovski, T., McClintock, P. V.,
398. &Stefanovska, A. (2012). Dynamical Bayesian
[11] Ekstrom, A. D., Caplan, J. B., Ho, E., Shattuck, K., inference of time-evolving interactions: From a pair of
Fried, I., &Kahana, M. J. (2005).Human coupled oscillators to networks of oscillators.
hippocampal theta activity during virtual navigation. Physical Review E, 86(6), 061126.
Hippocampus, 15(7), 881-889. [29] Stefanovska, A., Haken, H., McClintock, P. V.,
Hoi, M., Bajrovi, F., &Ribari, S.
279
Proceedings of The 21st World Multi-Conference on Systemics, Cybernetics and Informatics (WMSCI 2017)

(2000).Reversible transitions between synchronization memory sequences in prefrontal cortex during sleep.
states of the cardiorespiratory system. Physical Science, 318(5853), 1147-1150.
review letters, 85(22), 4831. [45] Johnson, L. A., Euston, D. R., Tatsuno, M., &
[30] Palu, M., &Stefanovska, A. (2003). Direction of McNaughton, B. L. (2010). Stored-trace
coupling from phases of interacting oscillators: an reactivation in rat prefrontal cortex is correlated with
information-theoretic approach. Physical Review down-to-up state fluctuation density. The Journal of
E, 67(5), 055201. Neuroscience, 30(7), 2650-2661.
[31] Jamek, J., Palu, M., &Stefanovska, A. [46] Busch, N. A., Dubois, J., & VanRullen, R. (2009).
(2010).Detecting couplings between interacting The phase of ongoing EEG oscillations predicts visual
oscillators with time-varying basic frequencies: perception. The Journal of Neuroscience, 29(24),
Instantaneous wavelet bispectrum and information 7869-7876.
theoretic approach. Physical Review E, 81(3), 036207. [47] Lakatos, P., Shah, A. S., Knuth, K. H., Ulbert, I.,
[32] Stankovski, T., McClintock, P. V., &Stefanovska, Karmos, G., & Schroeder, C. E. (2005).An
A. (2013). Coupling functions enable secure oscillatory hierarchy controlling neuronal
communications. Physical Review X, 4(1), 011026. excitability and stimulus processing in the
[33] Mormann, F., Lehnertz, K., David, P., &Elger, C. E. auditory cortex. Journal of neurophysiology, 94(3),
(2000). Mean phase coherence as a measure for 1904-1911.
phase synchronization and its application to the [48] Mathewson, K. E., Gratton, G., Fabiani, M., Beck,
EEG of epilepsy patients. Physica D: Nonlinear D. M., & Ro, T. (2009). To see or not to see:
Phenomena, 144(3), 358-369. prestimulus phase predicts visual awareness. The
[34] Rulkov, N. F., Sushchik, M. M., Tsimring, L. S., Journal of neuroscience, 29(9), 2725-2732.
&Abarbanel, H. D. (1995). Generalized [49] Klimesch, W., Sauseng, P., & Hanslmayr, S. (2007).
synchronization of chaos in directionally coupled EEG alpha oscillations: the inhibition timing
chaotic systems. Physical Review E, 51(2), 980. hypothesis. Brain research reviews, 53(1), 63-88.
[35] Kralemann, B., Frhwirth, M., Pikovsky, A., [50] Michel, C. M., Lehmann, D., Henggeler, B., &
Rosenblum, M., Kenner, T., Schaefer, J., & Moser, Brandeis, D. (1992). Localization of the sources
M. (2013). In vivo cardiac phase response curve of EEG delta, theta, alpha and beta frequency bands
elucidates human respiratory heart rate variability. using the FFT dipole approximation.
Nature communications, 4. Electroencephalography and clinical neurophysiology,
[36] Iatsenko, D., Bernjak, A., Stankovski, T., 82(1), 38-44.
Shiogai, Y., Owen-Lynch, P. J., Clarkson, P. B. M., [51] Tsuno, N., Shigeta, M., Hyoki, K., Kinoshita, T.,
&Stefanovska, A. (2013). Evolution of cardiorespiratory Ushijima, S., Faber, P. L., & Lehmann, D. (2002).
interactions with age. Philosophical Transactions Spatial organization of EEG activity from alertness to
of the Royal Society A: Mathematical, Physical sleep stage 2 in old and younger subjects. Journal
and EngineeringSciences,371(1997), 20110622. of sleep research, 11(1), 43-51.
[37] Kralemann, B., Cimponeriu, L., Rosenblum, M., [52] Canuet, L., Ishii, R., Pascual-Marqui, R. D., Iwase,
Pikovsky, A., & Mrowka, R. (2008). Phase M., Kurimoto, R., Aoki, Y., & Takeda, M. (2011).
dynamics of coupled oscillators reconstructed from Resting-state EEG source localization and functional
data. Physical Review E, 77(6), 066205. connectivity in schizophrenia-like psychosis of
[38] Duggento, A., Luchinsky, D. G., Smelyanskiy, V. N., epilepsy. PloS one, 6(11), e27863.
Khovanov, I., & McClintock, P. V. (2008).
Inferential framework for nonstationary dynamics. II.
Application to a model of physiological
signaling. Physical Review E, 77(6), 061106.
[39] Womelsdorf, T., & Fries, P. (2007). The role of
neuronal synchronization in selective attention.
Current opinion in neurobiology, 17(2), 154- 160.
[40] Womelsdorf, T., Schoffelen, J. M., Oostenveld, R.,
Singer, W., Desimone, R., Engel, A. K., & Fries,
P. (2007).Modulation of neuronal interactions
through neuronal synchronization. science, 316(5831),
1609-1612.
[41] van Elswijk, G., Maij, F., Schoffelen, J. M.,
Overeem, S., Stegeman, D. F., & Fries, P. (2010).
Corticospinal beta-band synchronization entails
rhythmic gain modulation. The Journal of
Neuroscience, 30(12), 4481-4488.
[42] Siegel, C. A., Marden, S. M., Persing, S. M., Larson,
R. J., & Sands, B. E. (2009). Risk of lymphoma
associated with combination anti tumor necrosis
factor and immunomodulator therapy for the treatment of
Crohn's disease: a meta-analysis. Clinical
Gastroenterology and Hepatology, 7(8), 874-881
[43] Gregoriou, G. G., Gotts, S. J., Zhou, H., &
Desimone, R. (2009). High-frequency, long- range
coupling between prefrontal and visual cortex during
attention. Science, 324(5931), 1207-1210.
[44] Euston, D. R., Tatsuno, M., & McNaughton, B. L.
(2007). Fast-forward playback of recent
280

Das könnte Ihnen auch gefallen