Statistical analysis of EEG microstate sequences

1. Data input

Enter your microstate sequence here. Click one of the buttons to either copy and paste your sequence in a text box, or to upload the data from the file.

Choose your input option:

Sample rate (Hz):

2. Parse input

Input sequence:

Parsed symbols:

In case your symbols are not numeric, e.g. A, B, C, D, they will be mapped to integer values.

Mapped symbols:

Mapped sequence:

3. Analysis

If you hit this button, a first-order Markov surrogate will be analyzed,
NOT the sequence you provided as input.

History length k = samples.

Distribution:

Transition matrix:

Markov-0 test:

Markov-1 test:

Shannon entropy:

Entropy rate:

Active information storage (AIS):

Partial autoinformation function (PAIF):

Autoinformation function (AIF):

Max. AIF time lag: samples.

Python user?

If you want to include this analysis into a Python script, visit the GitHub repo for the Python implementation

The Python code has been used in these publications:
[1] F. von Wegner, S. Bauer, F. Rosenow, J. Triesch, H. Laufs, “EEG microstate periodicity explained by rotating phase patterns of resting-state alpha oscillations.”, NeuroImage, 224:117372, 2021.
[2] F. von Wegner, H. Laufs, “Information-theoretical analysis of EEG microstate sequences in Python.”, Front Neuroinform, 12:30, 2018.
[3] F. von Wegner, P. Knaut, H. Laufs, “EEG microstate sequences from different clustering algorithms are information-theoretically invariant.”, Front Comp Neurosci, 12:70, 2018.
[4] F. von Wegner, E. Tagliazucchi, H. Laufs, “Information theoretical analysis of resting state EEG microstate sequences - non-Markovianity, non-stationarity and periodicities.”, NeuroImage, 158:99-111, 2017.