Journal: The Journal of Neuroscience

Article Title: Genetic Contributions to Long-Range Temporal Correlations in Ongoing Oscillations

doi: 10.1523/JNEUROSCI.3083-07.2007

Figure Lengend Snippet: From broadband EEG to the DFA of narrow-band ongoing oscillations. A 10 s segment of EEG from occipital electrode O2 showing alpha oscillation bursts before (A) and after (B) bandpass filtering in the alpha-frequency range (8–13 Hz). We analyze the temporal structure of the amplitude (Ampl.) envelope of the oscillation, which is indicated with a thick line in B. Ongoing oscillations in the human EEG generally exhibit long-range temporal (auto-)correlations (LRTC), which are identified qualitatively as large variations in the duration and magnitude of the amplitude envelope over time as seen in the dizygotic twin siblings 1 and 2 (C, D). The temporal structure and correlations of the signal in C may be removed by randomly shuffling the signal in windows of 100 ms (E). The DFA exponent, α, provides a quantitative measure of LRTC, and the stronger correlations in dizygotic twin 1 (F, circles) compared with dizygotic twin 2 (F, squares) is reflected in a value of α closer to 1 (0.94 vs 0.68). The lack of temporal structure and correlations in E is reflected in the DFA exponent having the value of ∼0.5, which is characteristic of an uncorrelated random process (F, dots).

Article Snippet: Brain electric activity was recorded using Electro-cap electrodes placed according to the international 10–20 system ( n = 14 electrodes).

Techniques:

Journal: The Journal of Neuroscience

Article Title: Genetic Contributions to Long-Range Temporal Correlations in Ongoing Oscillations

doi: 10.1523/JNEUROSCI.3083-07.2007

Figure Lengend Snippet: Power-law decaying correlations in the amplitude fluctuations of alpha- and beta-frequency band oscillations. Three complementary autocorrelation analyses were performed: the autocorrelation function (Autocorr., A), the power spectral density (Power, B), and the DFA (C). Each analysis was performed on the amplitude envelope of alpha and beta oscillations and averaged across the 368 subjects and 14 electrodes (open circles; see Materials and Methods). The data have been fitted with a power-law function on time scales from 1 to 20 s (black lines). The autocorrelation function is plotted semilogarithmically, because the autocorrelation function can attain negative values; power spectral density and DFA are plotted in log-log coordinates. Each analysis points to a slow decay of correlation or LRTC. The dots indicate the analysis of computer-generated white noise that is filtered identically to the EEG data. The autocorrelation function is zero at all time lags, the power spectrum is flat, and the DFA exponent is close to 0.5 for both frequency bands, showing that the bandpass filters did not introduce autocorrelations in the amplitude time series on the long time scales investigated here.

Article Snippet: Brain electric activity was recorded using Electro-cap electrodes placed according to the international 10–20 system ( n = 14 electrodes).

Techniques: Generated, Introduce

Journal: The Journal of Neuroscience

Article Title: Genetic Contributions to Long-Range Temporal Correlations in Ongoing Oscillations

doi: 10.1523/JNEUROSCI.3083-07.2007

Figure Lengend Snippet: Genetic factors shape the power and the temporal correlation structure of ongoing oscillations. The scatter plots show the logarithmically transformed power (A) and the DFA exponents (B) of monozygotic (circles, n = 80 pairs) and dizygotic (pluses, n = 104 pairs) twin pairs for alpha and beta at occipital electrode O2. That both power and DFA of alpha and beta oscillations are heritable traits is indicated by the more than two times higher correlation between monozygotic twins than between dizygotic twins. Twin correlations, r, are indicated in the plots.

Article Snippet: Brain electric activity was recorded using Electro-cap electrodes placed according to the international 10–20 system ( n = 14 electrodes).

Techniques: Transformation Assay

Journal: The Journal of Neuroscience

Article Title: Genetic Contributions to Long-Range Temporal Correlations in Ongoing Oscillations

doi: 10.1523/JNEUROSCI.3083-07.2007

Figure Lengend Snippet: Genetic variances of DFA and power in ongoing oscillations are independent. There are no correlations between DFA exponents and power of ongoing alpha and beta oscillations at occipital electrode O2 (A), either before (upper plots) or after (lower plots) removing the subjects with a low SNR. B, At the central electrode C3, one may get the impression that DFA and oscillation power are correlated (upper plots). However, after removing 94 and 117 subjects with a low SNR in the alpha and beta band, respectively, we recover the nonsignificant zero correlation, which was observed also in occipital and parietal leads with a high SNR. The number of subjects included in each scatter plot is indicated with “n.” Significance levels of Pearson's coefficients of correlation: ns, p >0.05; **p < 0.001.

Article Snippet: Brain electric activity was recorded using Electro-cap electrodes placed according to the international 10–20 system ( n = 14 electrodes).

Techniques: