Bob Thatcher is a leading expert on this subject. He is the researcher / inventor / developer who produced the Neuroguide suite of QEEG applications. He also was a key figure in the initial creation of QEEG databases with E. Roy John of NYU.
Here is the key definition in that document (previous pdf) on page 14, section 6: "When the phase difference between two signals is constant then coherence = 1, when the phase difference between signals is random then coherence = 0."
The 'Synchrony' measure is only maximum when phase difference between the two signals is ZERO. That is, they are exactly in phase synchrony. Coherence indicates the cortical relationship, even when phase is different from zero. This is important, because with functional networks in the brain, the networks are frequently connected and correlated, even though phase difference is non-zero.
Pete's analogy is useful, though not as precise as what Thatcher is saying here. Pete is implying that synchrony is measured at the same moment in time. Whereas in coherence, the relationship can be spread out in time.
The plan was to sort out the phase coherence code for a presentation last month. Of course, I run out of time, so only managed to use FZ's theta median with a neuro-feedback threshold: when this median was above threshold a synthesised chakapa sound faded in and circulated around the audience.
My premature code for phase coherence only managed to calculate amplitude differences between theta medians. As I had four speakers around the audience, I ended up mapping the amplitude difference between P3-P4 and F3-F4 to panning parameters of sound textures. These sound textures used live sound as their source, went through granular synthesis and delay effects and then were spatialised with the theta amplitude differences.
Good news is that I have been receiving help on the SuperCollider forum and now very likely have a code that I can use for phase coherence - still testing, but it feels good!
Comments
Bob Thatcher is a leading expert on this subject. He is the researcher / inventor / developer who produced the Neuroguide suite of QEEG applications. He also was a key figure in the initial creation of QEEG databases with E. Roy John of NYU.
https://pdfs.semanticscholar.org/2bb5/53438cd587bc8b0a8d53f6aa493dcb8b7f35.pdf
Hand Calculator Calculations of EEG Coherence, Phase Delays and Brain Connectivity
Here is the key definition in that document (previous pdf) on page 14, section 6: "When the phase difference between two signals is constant then coherence = 1, when the phase difference between signals is random then coherence = 0."
The 'Synchrony' measure is only maximum when phase difference between the two signals is ZERO. That is, they are exactly in phase synchrony. Coherence indicates the cortical relationship, even when phase is different from zero. This is important, because with functional networks in the brain, the networks are frequently connected and correlated, even though phase difference is non-zero.
Pete's analogy is useful, though not as precise as what Thatcher is saying here. Pete is implying that synchrony is measured at the same moment in time. Whereas in coherence, the relationship can be spread out in time.
William
quick update:
The plan was to sort out the phase coherence code for a presentation last month. Of course, I run out of time, so only managed to use FZ's theta median with a neuro-feedback threshold: when this median was above threshold a synthesised chakapa sound faded in and circulated around the audience.
My premature code for phase coherence only managed to calculate amplitude differences between theta medians. As I had four speakers around the audience, I ended up mapping the amplitude difference between P3-P4 and F3-F4 to panning parameters of sound textures. These sound textures used live sound as their source, went through granular synthesis and delay effects and then were spatialised with the theta amplitude differences.
Video of the performance part:
Here a survey where the audio is tested (without EEG):
https://www.surveymonkey.co.uk/r/shamanic-obcif
// please consider participating!
Good news is that I have been receiving help on the SuperCollider forum and now very likely have a code that I can use for phase coherence - still testing, but it feels good!
thanks for all your support! k
// some notes to add to the thread:
there is some interesting info on coherence used in nf here:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6193115/pdf/fnins-12-00729.pdf
and I also liked this 'old' TM video:
... in case anyone is interested in binaural beats, please consider participating in this (no EEG) study:
https://bcmi.khofstadter.info/aru-talking-science-series-2020/
// you can skip the video where I am talking, the introduction to the survey explains the study in a nutshell (it should be ~35 minutes long).
thanks! k