Frontal delta effects in time-frequency data

Hello Mike,

I have extracted a time-frequency representation of my EEG data (fs=1kHz) using Morlet wavelet convolutions. For each channel, I analyzed the 2s-long epochs and 50 frequencies in the range of 1 to 40Hz. The number of kernel cycles was variable increasing from 3 to 12 in a logarithmic manner.

I contrasted 2 experimental conditions by applying a cluster-based permutation testing in its typical form: neighboring in space, in time and/or in frequency pixels were assigned in the same clusters. The null distribution consists of the t-scores resulted from the “maximum cluster” of each iteration. Maximum cluster is the cluster that corresponds to the max (or min) average t-score.

This testing revealed, among other clusters, a significant cluster consisted of 4 frontal channels and broad (300-1000ms) delta effects (1-3.5Hz).

I was wondering if this cluster is attributed to “real delta alterations” or to noisy properties of frontal channels (eyes are closed in both conditions), or even to convolution properties in the delta range.

Thanks in advance,
A.

Hi A. Without more knowledge, it’s impossible to say. Did you do ICA cleaning to remove EOG artifacts? Is the topographical distribution central-anterior that looks like a blink? Are there residual blink artifacts in the data?

Maybe you can include some topographical maps and details about the data cleaning.

Something you might also consider is running ICA a second time, after removing the blink artifacts, to see if there are residual artifacts to remove.

Hello Mike and thanks for your response.

I did not perform ICA cleaning because subjects had their eyes closed during the 1s-long period of interest. Howerer, I rejected the epochs showing blinks (> ±80μV) during the selected pre-stimulus baseline (-300 to -100 ms). The topographical map of grand average ERPs does not look like a constant blink (a broad central-anterior distribution appears only in 300-500ms).

In any case, I will run ICA to confirm (or not) the absence of strong blinks. Do you have any suggestions about the cause of frontal delta effects?

Thank you in advance,
Anastasios

Well, if it’s not caused by an artifact then I guess it’s caused by the brain!

1 Like