Limits...
Coupling among electroencephalogram gamma signals on a short time scale.

McAssey MP, Hsieh F, Smith AC - Comput Intell Neurosci (2010)

Bottom Line: We first apply our variable-window method to simulated signals and compare its performance to a fixed-window approach.We then focus on gamma signals recorded in two regions of the rat hippocampus.Our results indicate that this may be a useful method for mapping coupling patterns among signals in EEG datasets.

View Article: PubMed Central - PubMed

Affiliation: Department of Statistics, University of California Davis, MSB 4118 One Shields Avenue, Davis, CA 95616, USA.

ABSTRACT
An important goal in neuroscience is to identify instances when EEG signals are coupled. We employ a method to measure the coupling strength between gamma signals (40-100 Hz) on a short time scale as the maximum cross-correlation over a range of time lags within a sliding variable-width window. Instances of coupling states among several signals are also identified, using a mixed multivariate beta distribution to model coupling strength across multiple gamma signals with reference to a common base signal. We first apply our variable-window method to simulated signals and compare its performance to a fixed-window approach. We then focus on gamma signals recorded in two regions of the rat hippocampus. Our results indicate that this may be a useful method for mapping coupling patterns among signals in EEG datasets.

Show MeSH

Related in: MedlinePlus

Progression of BIC as p increases, for EEG application.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2926578&req=5

fig8: Progression of BIC as p increases, for EEG application.

Mentions: Our goal is to reduce this eight-dimensional time series to a single dimension by identifying distinct IC states when different subsets of signals are coupled with the base. This single dimension—the IC state—will take one of the values in {1,…, p*} in each measurement window, where p* is defined in (12). The histograms in Figure 7 may be considered superpositions of the histograms of the IC estimates corresponding to these distinct states, each of which is modeled by an MVB distribution. We implement the EM algorithm as described in Section 2.2 to determine the parameters of these distributions and to determine the state membership of each IC estimate, using the R statistical package. To illustrate our method, we consider only a ten-second block from the full recorded time series, corresponding to 746 consecutive and overlapping three-cycle windows of the base signal. Moreover, we work with only four dimensions rather than all eight, by choosing representatives from each location in the brain in which multiple EEG signals are recorded. Using the IC estimates between the base signal at Tetrode 1 and the signals at Tetrodes 6, 11, 25, and 26, computed during an interval when the rat is in its cage, we proceed with the EM algorithm. In Figure 8 we note that the BIC decreases as p increments from 2 to 4, is relatively constant for p in the range of 4 to 6 states, and then increases thereafter. We conclude that the instantaneous coupling between the base signal recorded in the MEC and the signals from the four other selected locations transition among four to six distinct IC states during the chosen ten-second block. Since the simplest model is preferred, we adopt a model consisting of four IC states.


Coupling among electroencephalogram gamma signals on a short time scale.

McAssey MP, Hsieh F, Smith AC - Comput Intell Neurosci (2010)

Progression of BIC as p increases, for EEG application.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2926578&req=5

fig8: Progression of BIC as p increases, for EEG application.
Mentions: Our goal is to reduce this eight-dimensional time series to a single dimension by identifying distinct IC states when different subsets of signals are coupled with the base. This single dimension—the IC state—will take one of the values in {1,…, p*} in each measurement window, where p* is defined in (12). The histograms in Figure 7 may be considered superpositions of the histograms of the IC estimates corresponding to these distinct states, each of which is modeled by an MVB distribution. We implement the EM algorithm as described in Section 2.2 to determine the parameters of these distributions and to determine the state membership of each IC estimate, using the R statistical package. To illustrate our method, we consider only a ten-second block from the full recorded time series, corresponding to 746 consecutive and overlapping three-cycle windows of the base signal. Moreover, we work with only four dimensions rather than all eight, by choosing representatives from each location in the brain in which multiple EEG signals are recorded. Using the IC estimates between the base signal at Tetrode 1 and the signals at Tetrodes 6, 11, 25, and 26, computed during an interval when the rat is in its cage, we proceed with the EM algorithm. In Figure 8 we note that the BIC decreases as p increments from 2 to 4, is relatively constant for p in the range of 4 to 6 states, and then increases thereafter. We conclude that the instantaneous coupling between the base signal recorded in the MEC and the signals from the four other selected locations transition among four to six distinct IC states during the chosen ten-second block. Since the simplest model is preferred, we adopt a model consisting of four IC states.

Bottom Line: We first apply our variable-window method to simulated signals and compare its performance to a fixed-window approach.We then focus on gamma signals recorded in two regions of the rat hippocampus.Our results indicate that this may be a useful method for mapping coupling patterns among signals in EEG datasets.

View Article: PubMed Central - PubMed

Affiliation: Department of Statistics, University of California Davis, MSB 4118 One Shields Avenue, Davis, CA 95616, USA.

ABSTRACT
An important goal in neuroscience is to identify instances when EEG signals are coupled. We employ a method to measure the coupling strength between gamma signals (40-100 Hz) on a short time scale as the maximum cross-correlation over a range of time lags within a sliding variable-width window. Instances of coupling states among several signals are also identified, using a mixed multivariate beta distribution to model coupling strength across multiple gamma signals with reference to a common base signal. We first apply our variable-window method to simulated signals and compare its performance to a fixed-window approach. We then focus on gamma signals recorded in two regions of the rat hippocampus. Our results indicate that this may be a useful method for mapping coupling patterns among signals in EEG datasets.

Show MeSH
Related in: MedlinePlus