Just a quick progress report here – My previous graphs of relative power for each frequency band (delta, theta, alpha, beta, gamma) used only the mean value of all four sensors. Now I have been plotting spectrograms of /muse/eeg values. I am not seeing dropped samples, so I simply use all eeg values at 220 Hz. The attached graphs are for a four-minute meditation session with eyes open–first half is while gazing at a single spot on the floor; latter half is with eyes moving around.

I expected relative power of alpha to decrease, but I was surprised to see such a dramatic increase in delta.

Examining the spectrogram, I noticed that the right rear electrode gave the strongest signal, so I turned to a graph of relative power for that electrode alone, which shows a rather dramatic effect from the earlier to the later part of the session.

So I recommend spectrograms for picking up on these kinds of signals.



Delta high-band is not normal, meditation process, this band should not be too high, otherwise it is in a drowsy and distracted, after practice, should be invigorating, thought peace, stability.


Not sure if averaging 4 channels is a good idea.