Re: Increasing coherence?

February 3, 2006

and listmates~

I'll take a stab but please correct me if I'm off base...

But my understanding is, (with the exception of the T3 T4) that because you are training both sides of the head in those 2 channel trainings that you would also be training different frequencies on each side and so it wouldn't train the coherence up. (Beta on the left and lobeta/SMR or alpha on the right.) Training different frequencies on different sides can actually break up high coherences. I believe ....that training with an interhemispheric protocol like F3F4G or C4C3 G can increase coherence between the 2 sites.

~

Increasing coherence?

Dear Pete & Listmates,

I was wondering if when we're training montage like F3/A1/g/F4/A2, C3/A1/g/C4/A2 and even T3/A1/g/T4/A2 with linked ears, rewarding frequences as LoBeta or even Beta in some cases, aren't we increasing the coherence between the hemispheres? Because if is this true, the client may will present some problems related with LoBeta and/or Beta high coherence. ,C.H.BrainTrainer @ Training for LifeMiami,FL

- Helps protect you from nasty viruses.

February 3, 2006

et al. -

What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected.

Bi-polar recording looks at the difference between the two recording sites. If you train this difference up, you're pushing dissimilarity up betweent he two sites. If training it down, you're pushing site similarity up. Similarity between the two sites can be a matter of the amplitudes, frequencies and phase relationship of their brain waves. Any one or two of these, or all three.

There are different ways to compute coherence; but all I believe primarily measure the degree of similarity between the frequency spectra of the two signals and also take amplitude similarity into some account. However, coherence is not sensitive to phase differences. It is unlikely but not theoretically impossible, that bi-polar training could primarily affect only the phase relationships. Nevertheless, bi-polar training has been referred to at times as the poor man's coherence training, particularly since it does not involve two-channel recording.

I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.

Coherence is a complex matter; and I welcome any correction of my inexpert comments here.

Larry , PhD155 East 38th Street (corner Third Avenue)New York NY 10016212-697-5990llewis@...

Increasing coherence?

Dear Pete & Listmates,

I was wondering if when we're training montage like F3/A1/g/F4/A2, C3/A1/g/C4/A2 and even T3/A1/g/T4/A2 with linked ears, rewarding frequences as LoBeta or even Beta in some cases, aren't we increasing the coherence between the hemispheres? Because if is this true, the client may will present some problems related with LoBeta and/or Beta high coherence. ,C.H.BrainTrainer @ Training for LifeMiami,FL

- Helps protect you from nasty viruses.

February 3, 2006

Larry,

I'm far from being an expert in this area myself, but a few comments:

> What said seemed basically correct to me; but I would add that

inter-hemispheric training, or for that matter any form of bi-polar recording,

is likely to affect coherence - but also may not even when the similarity

between two sites is affected.

Actually what the bipolar training, interhemispheric or not, is most likely to

affect is PHASE, not coherence. When two signals are in phase, subtracting one

from the other will result in a very low value; when they are 180 degrees out of

phase, it will result in a large value. So, in addition to training down the

difference in raw amplitude, it is highly likely that downtraining bipolar

signals will move the signals IN phase and uptraining them will move them OUT of

phase. Since coherence is independent of phase, it will not necessarily be

affected.

> I think one factor that favors the use of bi-polar, one channel recording is

that it by-passes the contaminating effect of the 2 reference recording sites

that are used in 2 channel coherence training, one for each of the 2 " active "

recording sites. . These effects can be large. One particularly huge effect

that should be totally avoided is using a linking connector to provide the same

reference to each channel. This greatly elevates coherence values because the

linkage of the references automatically makes the two signals more similar.

Here I would disagree and say that when comparing two sites against a common

reference actually INCREASES the quality of the reading, since we are truly

comparing what is happening at the two sites, independent of what might be

happening differently at the two reference sites.

Pete

February 3, 2006

Yes , I totally agree with you, that was the reason of my question, because somtimes we need to train e.g. T3-T4 12-15Hz as a stabilization interhemispheric protocol, basicly we're training same frequence at both sides, no?! and when I referred to 2C designs I was thinking in temporal training first. Thank you for your feedback, JR Duncan <karenduncan@...> wrote: and listmates~ I'll take a stab but please correct me if I'm off base... But my understanding is, (with the exception of the T3 T4) that because you are training both sides of the

head in those 2 channel trainings that you would also be training different frequencies on each side and so it wouldn't train the coherence up. (Beta on the left and lobeta/SMR or alpha on the right.) Training different frequencies on different sides can actually break up high coherences. I believe ....that training with an interhemispheric protocol like F3F4G or C4C3 G can increase coherence between the 2 sites. ~ Increasing coherence? Dear Pete & Listmates, I was wondering if when we're training montage like F3/A1/g/F4/A2, C3/A1/g/C4/A2 and even T3/A1/g/T4/A2 with linked ears, rewarding frequences as LoBeta or even Beta in some cases, aren't we increasing the coherence between the hemispheres? Because if is this true, the client may will present some problems related with LoBeta and/or Beta high coherence. ,C.H.BrainTrainer @ Training for LifeMiami,FL - Helps protect you from nasty viruses.

Relax. virus scanning helps detect nasty viruses!

February 4, 2006

I'm confused by this discussion and speculation regarding coherence and bipolar training. Is there any evidence beyond what seems to be anecdotal that the assertions connecting bipolar training and coherence are true? Is it too much to ask that some controlled clinical trials be referred to when these assertions are made? Absent trials, wouldn't it be possible for those of you who are professional NF trainers -- ie. have a lot of data --to review those data (eg. C3/C4 single channel SMR up, theta down) to see if coherence moves in the directions suggested? As a home trainer, I'm a sympathetic and friendly user of NF -- I think it works (better still, I don't see why it wouldn't work!), but I think a little more hard data would serve us all well. nick

mammanoLarry <llewis@...> wrote: et al. - What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. Bi-polar recording looks at the difference between the two recording sites. If you train this difference up, you're pushing dissimilarity up betweent he two sites. If training it down, you're

pushing site similarity up. Similarity between the two sites can be a matter of the amplitudes, frequencies and phase relationship of their brain waves. Any one or two of these, or all three. There are different ways to compute coherence; but all I believe primarily measure the degree of similarity between the frequency spectra of the two signals and also take amplitude similarity into some account. However, coherence is not sensitive to phase differences. It is unlikely but not theoretically impossible, that bi-polar training could primarily affect only the phase relationships. Nevertheless, bi-polar training has been referred to at times as the poor man's coherence training, particularly since it does not involve two-channel recording. I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar. Coherence is a complex matter; and I welcome any correction of my inexpert comments here. Larry , PhD155 East 38th Street (corner Third Avenue)New York

NY 10016212-697-5990llewis@... Increasing coherence? Dear Pete & Listmates, I was wondering if when we're training montage like F3/A1/g/F4/A2, C3/A1/g/C4/A2 and even T3/A1/g/T4/A2 with linked ears, rewarding frequences as LoBeta or even Beta in some cases, aren't we increasing the coherence between the hemispheres? Because if is this true, the client may will present some problems related with LoBeta and/or Beta high coherence. ,C.H.BrainTrainer @ Training for LifeMiami,FL - Helps protect you from nasty viruses.

February 4, 2006

Pete,

I said that bi-polar training may not affect coherence; you argue that it is likely not to.

Pete: > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected

This seems correct except, perhaps, for the following considerations.

It has always seemed to me that if a parameter of the EEG is trainable, e.g. amplitude; it must be because the brain self-responds to such changes on its own and then correlates information from its own response (or self-monitoring) with the neurofeedback information. Its as if the brain says to itself: "Hey! What's going on here, I notice that whenever I register change X, something sounds out there in the environment! This never happened before.!" And this draws the brain's attention to the neurofeedback signal; and the brain tries to understand this novelty better (attention is always drawn to novelty) by playing with it.

Assuming that something like this is the case, it is likely that the brain responds more to some parameters than others. The fact that phase is very difficult to train suggests that the brain may not do much self-monitoring of this particular parameter; and as a result what you argue may not be the case. The effect of bi-polar training on phase similarity might be quite weak. I think this is something that needs empirical testing. The brain will not relate to the protocol ideas we have about it if those ideas are at variance with how the brain actually functions.

I would disagree with your second point.

Pete::> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.

Whether using a single or common references, one is recording both reference site activity as well as at that of the so-called "active" sites. In no way is the recording "independent" of the reference(s). There appears to be some logic to the idea that subtracting the same reference activity from each "active" site purifies, as it were, the referential contamination by taking it out; but you do not end up "truly; comparing what is happening at the two sites" but rather comparing what is happening with the signals at each site after they have been modified in the same way by the subtraction. I can see advantages in this for amplitude recordings; but major disadvantages for coherence recording.

The modification of the signals from the active sites will result in a much larger coherence value than that obtained if separate references are used. True, separate references will introduce two separate modifications; but because these will be different from one another, they will affect the situation in different, unrelated ways. This should produce a smaller effect on the coherence value than the huge increase in similarity that is caused by modifying each signal in the same way with a common reference.

As far as I know, no way exists of recording single site brain wave activity independently of any other sites. I've wondered, though, if it might be possible to use circuitry of the noise canceling sort with two separate reference electrodes traingulating the "active" electrode to derive an "independent" signal.

Larry , PhD155 East 38th Street (corner Third Avenue)New York NY 10016212-697-5990llewis@...

Re: Re: Increasing coherence?

> Larry,> > I'm far from being an expert in this area myself, but a few comments:> >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.

February 4, 2006

Sorry, got to jump in,Phase is taken into account in comodulation training developed by Kaiser and Barry Sterman, as I understand it. Com-mod is trainable and clinical reports suggest fast responses.

Best,

Sheila

PS Did virus check on my whole computer and found nothing, and no comments from others I email either...go figure???

Re: Re: Increasing coherence?

> Larry,> > I'm far from being an expert in this area myself, but a few comments:> >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.

February 5, 2006

I bought into the argument that 2C training was preferred to single channel training – that the latter was the “poor man’s method” or whatever, or that you get more bang for the buck with 2 channel training, etc. But Pete’s recent reference to amplitudes and phase differences makes me wonder if simpler single channel bipolar training might be the way to go in many cases. Pete’s point: (really fundamental physics) it is the phase difference between the signals from two sites

in bipolar training, eg. C3C4g, that is being trained up or down. Well phase differences have to do with the TIMING between the two signals. And this reminded me of a very challenging (and persuasive) article by Othmer (“Quantitative EEG and Neurofeedback” , & Arbabanel, Eds, pp262--)in which he argues that what he calls “time binding” may be THE MOST IMPORTANT consideration in EEG feedback training. He quotes “what fires together, wires together” where “wiring” can refer to short transient or

“functional” connections as well as the more permanent hard wired ones. “Firing together” is another way of saying the signals (that is the EEG “brainwaves”) are in phase. Presumably there are signals (theta, hibeta?) that are best left out of phase by training down. Thus is it correct, in light of the above, to infer that the bipolar protocols (or montages) are in fact FUNDAMENTALLY different from the referential ones? nickLarry <llewis@...> wrote: Pete, I said that bi-polar training may not affect coherence; you argue that it is likely not to. Pete: > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition

to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected This seems correct except, perhaps, for the following considerations. It has always seemed to me that if a parameter of the EEG is trainable, e.g. amplitude; it must be because the brain self-responds to such changes on its own and then correlates information from its own response (or self-monitoring) with the neurofeedback information. Its as if the brain says to itself: "Hey! What's going on here, I notice that whenever I register change X, something sounds out there in the environment! This never happened before.!" And this draws the brain's

attention to the neurofeedback signal; and the brain tries to understand this novelty better (attention is always drawn to novelty) by playing with it. Assuming that something like this is the case, it is likely that the brain responds more to some parameters than others. The fact that phase is very difficult to train suggests that the brain may not do much self-monitoring of this particular parameter; and as a result what you argue may not be the case. The effect of bi-polar training on phase similarity might be quite weak. I think this is something that needs empirical testing. The brain will not relate to the protocol ideas we have about it if those ideas are at variance with how the brain actually functions. I would disagree with your second point. Pete::> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites. Whether using a single or common references, one is recording both reference site activity as well as at that of the so-called "active" sites. In no way is the recording "independent" of the reference(s). There appears to be some logic to the idea that subtracting the same reference activity from each "active" site purifies, as it were, the referential contamination by taking it out; but you do not end up "truly; comparing what is happening at the two sites" but rather comparing what is happening with the signals at each

site after they have been modified in the same way by the subtraction. I can see advantages in this for amplitude recordings; but major disadvantages for coherence recording. The modification of the signals from the active sites will result in a much larger coherence value than that obtained if separate references are used. True, separate references will introduce two separate modifications; but because these will be different from one another, they will affect the situation in different, unrelated ways. This should produce a smaller effect on the coherence value than the huge increase in similarity that is caused by modifying each signal in the same way with a common reference. As far as I know, no way exists of recording single site brain wave activity independently of any other sites. I've wondered, though, if it might be possible to use

circuitry of the noise canceling sort with two separate reference electrodes traingulating the "active" electrode to derive an "independent" signal. Larry , PhD155 East 38th Street (corner Third Avenue)New York NY 10016212-697-5990llewis@... Re: Re: Increasing coherence? > Larry,> > I'm far from being an expert in this area myself, but a few comments:> >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining

bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two

sites, independent of what might be happening differently at the two reference sites.

February 5, 2006

Nick,

I think it's safe to say that monopolar and bipolar montages are fundamentally

different in what they measure and do. However, if I said that bipolar inhibit

montages must be training phase, then I did not mean to say that. That's ONE of

the options. It could also simply be training one amplitude down relative to

the other. Sometimes a cigar is just a cigar.

That's what many find discomfiting about bipolar montages: they could be doing

a variety of things, but there simply is no way to KNOW.

Pete

>

> From: NICK MAMMANO <nickmammano@...>

> Date: 2006/02/05 Sun PM 01:24:53 EST

>

> Subject: Re: Re: Increasing coherence?

>

> I bought into the argument that 2C training was preferred to single channel

training – that the latter was the “poor man’s method” or whatever, or that you

get more bang for the buck with 2 channel training, etc. But Pete’s recent

reference to amplitudes and phase differences makes me wonder if simpler single

channel bipolar training might be the way to go in many cases.

>

> Pete’s point: (really fundamental physics) it is the phase difference

between the signals from two sites in bipolar training, eg. C3C4g, that is being

trained up or down.

>

> Well phase differences have to do with the TIMING between the two signals.

And this reminded me of a very challenging (and persuasive) article by Othmer

(“Quantitative EEG and Neurofeedback” , & Arbabanel, Eds, pp262--)in which

he argues that what he calls “time binding” may be THE MOST IMPORTANT

consideration in EEG feedback training.

>

> He quotes “what fires together, wires together” where “wiring” can refer to

short transient or “functional” connections as well as the more permanent hard

wired ones.

>

> “Firing together” is another way of saying the signals (that is the EEG

“brainwaves”) are in phase. Presumably there are signals (theta, hibeta?) that

are best left out of phase by training down.

>

> Thus is it correct, in light of the above, to infer that the bipolar

protocols (or montages) are in fact FUNDAMENTALLY different from the referential

ones?

>

> nick

>

> Larry <llewis@...> wrote: Pete,

>

> I said that bi-polar training may not affect coherence; you argue that it

is likely not to.

>

> Pete: > Actually what the bipolar training, interhemispheric or not, is most

likely to affect is PHASE, not coherence. When two signals are in phase,

subtracting one from the other will result in a very low value; when they are

180 degrees out of phase, it will result in a large value. So, in addition to

training down the difference in raw amplitude, it is highly likely that

downtraining bipolar signals will move the signals IN phase and uptraining them

will move them OUT of phase. Since coherence is independent of phase, it will

not necessarily be affected

>

> This seems correct except, perhaps, for the following considerations.

>

> It has always seemed to me that if a parameter of the EEG is trainable, e.g.

amplitude; it must be because the brain self-responds to such changes on its own

and then correlates information from its own response (or self-monitoring) with

the neurofeedback information. Its as if the brain says to itself: " Hey! What's

going on here, I notice that whenever I register change X, something sounds out

there in the environment! This never happened before.! " And this draws the

brain's attention to the neurofeedback signal; and the brain tries to understand

this novelty better (attention is always drawn to novelty) by playing with it.

>

> Assuming that something like this is the case, it is likely that the brain

responds more to some parameters than others. The fact that phase is very

difficult to train suggests that the brain may not do much self-monitoring of

this particular parameter; and as a result what you argue may not be the case.

The effect of bi-polar training on phase similarity might be quite weak. I think

this is something that needs empirical testing. The brain will not relate to the

protocol ideas we have about it if those ideas are at variance with how the

brain actually functions.

>

> I would disagree with your second point.

>

> Pete::> Here I would disagree and say that when comparing two sites against

a common reference actually INCREASES the quality of the reading, since we are

truly comparing what is happening at the two sites, independent of what might be

happening differently at the two reference sites.

>

> Whether using a single or common references, one is recording both

reference site activity as well as at that of the so-called " active " sites. In

no way is the recording " independent " of the reference(s). There appears to be

some logic to the idea that subtracting the same reference activity from each

" active " site purifies, as it were, the referential contamination by taking it

out; but you do not end up " truly; comparing what is happening at the two

sites " but rather comparing what is happening with the signals at each site

after they have been modified in the same way by the subtraction. I can see

advantages in this for amplitude recordings; but major disadvantages for

coherence recording.

>

> The modification of the signals from the active sites will result in a much

larger coherence value than that obtained if separate references are used.

True, separate references will introduce two separate modifications; but because

these will be different from one another, they will affect the situation in

different, unrelated ways. This should produce a smaller effect on the coherence

value than the huge increase in similarity that is caused by modifying each

signal in the same way with a common reference.

>

> As far as I know, no way exists of recording single site brain wave

activity independently of any other sites. I've wondered, though, if it might be

possible to use circuitry of the noise canceling sort with two separate

reference electrodes traingulating the " active " electrode to derive an

" independent " signal.

>

> Larry , PhD

> 155 East 38th Street (corner Third Avenue)

> New York NY 10016

> 212-697-5990

> llewis@...

> Re: Re: Increasing coherence?

>

> > Larry,

> >

> > I'm far from being an expert in this area myself, but a few comments:

> >

> >> What said seemed basically correct to me; but I would add that

inter-hemispheric training, or for that matter any form of bi-polar recording,

is likely to affect coherence - but also may not even when the similarity

between two sites is affected.

> >

> > Actually what the bipolar training, interhemispheric or not, is most likely

to affect is PHASE, not coherence. When two signals are in phase, subtracting

one from the other will result in a very low value; when they are 180 degrees

out of phase, it will result in a large value. So, in addition to training down

the difference in raw amplitude, it is highly likely that downtraining bipolar

signals will move the signals IN phase and uptraining them will move them OUT of

phase. Since coherence is independent of phase, it will not necessarily be

affected.

> >

> >> I think one factor that favors the use of bi-polar, one channel recording

is that it by-passes the contaminating effect of the 2 reference recording sites

that are used in 2 channel coherence training, one for each of the 2 " active "

recording sites. . These effects can be large. One particularly huge effect

that should be totally avoided is using a linking connector to provide the same

reference to each channel. This greatly elevates coherence values because the

linkage of the references automatically makes the two signals more similar.

> >

> > Pete> Here I would disagree and say that when comparing two sites against a

common reference actually INCREASES the quality of the reading, since we are

truly comparing what is happening at the two sites, independent of what might be

happening differently at the two reference sites.

>

February 6, 2006

Hi Nick

My pet theory is that bipolar montages particularrly left referenced to right are training the left hemisphere to be dominant.

Mark

Re: Re: Increasing coherence?> > > > > Larry,> > > > I'm far from being an expert in this area myself, but a few comments:> > > >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> > > >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > > > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.> > > >

February 6, 2006

Pete: You're right of course. I should not have implied that bipolar montages MUST be training phase (or that the amplitude differences observed MUST be a result of phase changes). But even if phase shifts are only one of the options, the possible connection to "time binding", brain rhythms and to the compelling arguments of Othmer lit a bulb. I don't any longer view bipolar montages as a "poor man's method", or something "less" than the "ear clip" referential montages. The proof is in the pudding I guess. There must be a huge amount of "data" among all the professional trainers who are members of this list. Based on their experience, and yours, I wonder what the general view is on the relative benefits of each method. nick mammano Van Deusen <pvdtlc@...>

wrote: Nick,I think it's safe to say that monopolar and bipolar montages are fundamentally different in what they measure and do. However, if I said that bipolar inhibit montages must be training phase, then I did not mean to say that. That's ONE of the options. It could also simply be training one amplitude down relative to the other. Sometimes a cigar is just a cigar.That's what many find discomfiting about bipolar montages: they could be doing a variety of things, but there simply is no way to KNOW.Pete> > From: NICK MAMMANO <nickmammano@...>> Date: 2006/02/05 Sun PM 01:24:53 EST> > Subject: Re: Re: Increasing coherence?> > I bought into the argument that 2C training was preferred to single channel training

– that the latter was the “poor man’s method” or whatever, or that you get more bang for the buck with 2 channel training, etc. But Pete’s recent reference to amplitudes and phase differences makes me wonder if simpler single channel bipolar training might be the way to go in many cases. > > Pete’s point: (really fundamental physics) it is the phase difference between the signals from two sites in bipolar training, eg. C3C4g, that is being trained up or down.> > Well phase differences have to do with the TIMING between the two signals. And this reminded me of a very challenging (and persuasive) article by Othmer (“Quantitative EEG and Neurofeedback” , & Arbabanel, Eds, pp262--)in which he argues that what he calls “time binding” may be THE MOST IMPORTANT consideration in EEG feedback training. > > He quotes “what fires

together, wires together” where “wiring” can refer to short transient or “functional” connections as well as the more permanent hard wired ones.> > “Firing together” is another way of saying the signals (that is the EEG “brainwaves”) are in phase. Presumably there are signals (theta, hibeta?) that are best left out of phase by training down.> > Thus is it correct, in light of the above, to infer that the bipolar protocols (or montages) are in fact FUNDAMENTALLY different from the referential ones?> > nick> > > Larry <llewis@...> wrote: Pete,> > I said that bi-polar training may not affect coherence; you argue that it is likely not to. > > Pete: > Actually

what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected> > This seems correct except, perhaps, for the following considerations. > > It has always seemed to me that if a parameter of the EEG is trainable, e.g. amplitude; it must be because the brain self-responds to such changes on its own and then correlates information from its own response (or self-monitoring) with the neurofeedback

information. Its as if the brain says to itself: "Hey! What's going on here, I notice that whenever I register change X, something sounds out there in the environment! This never happened before.!" And this draws the brain's attention to the neurofeedback signal; and the brain tries to understand this novelty better (attention is always drawn to novelty) by playing with it.> > Assuming that something like this is the case, it is likely that the brain responds more to some parameters than others. The fact that phase is very difficult to train suggests that the brain may not do much self-monitoring of this particular parameter; and as a result what you argue may not be the case. The effect of bi-polar training on phase similarity might be quite weak. I think this is something that needs empirical testing. The brain will not relate to the protocol ideas we have about it if those ideas are at variance with how the brain actually

functions.> > I would disagree with your second point. > > Pete::> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.> > Whether using a single or common references, one is recording both reference site activity as well as at that of the so-called "active" sites. In no way is the recording "independent" of the reference(s). There appears to be some logic to the idea that subtracting the same reference activity from each "active" site purifies, as it were, the referential contamination by taking it out; but you do not end up "truly; comparing what is happening at the two sites" but rather comparing what is

happening with the signals at each site after they have been modified in the same way by the subtraction. I can see advantages in this for amplitude recordings; but major disadvantages for coherence recording. > > The modification of the signals from the active sites will result in a much larger coherence value than that obtained if separate references are used. True, separate references will introduce two separate modifications; but because these will be different from one another, they will affect the situation in different, unrelated ways. This should produce a smaller effect on the coherence value than the huge increase in similarity that is caused by modifying each signal in the same way with a common reference. > > As far as I know, no way exists of recording single site brain wave activity independently of any other sites. I've wondered, though, if it might be

possible to use circuitry of the noise canceling sort with two separate reference electrodes traingulating the "active" electrode to derive an "independent" signal. > > > Larry , PhD> 155 East 38th Street (corner Third Avenue)> New York NY 10016> 212-697-5990> llewis@...> Re: Re: Increasing coherence?> > > > > Larry,> > > > I'm far from being an expert in this area myself, but a few comments:> > > >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for

that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> > > >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence

training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > > > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.> > > >

February 6, 2006

Hi Mark: I'm going to have to display my ignorance and ask some stupid questions. Do you also see a corresponding change in the amplitude difference? ie. does the amplitude on the left side increase wrt to the right side? Or does it matter? You're also implying that that left dominance occurs even when left is not referenced to right --- but less often. Is that right? nick Mark Baddeley <baddeley@...> wrote: Hi Nick My pet theory is that bipolar montages particularrly left referenced to right are training the left

hemisphere to be dominant. Mark Re: Re: Increasing coherence?> > > > > Larry,> > > > I'm far from being an expert in this area myself, but a few comments:> > > >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not

coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> > > >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the

linkage of the references automatically makes the two signals more similar.> > > > > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.> > > >

February 6, 2006

Hi Nick

Having found the Othmer's approach somewhat appealing as I was hoping it might lead toward neurotherapy's equivalent of psychotherapy's brief therapy. I have never been able to find the illusive "sweet spot " with clients. Although I notice that others on this list seem to be able to.

Val Brown similarly has people who find a lot in his approach.

A colleague has recently trained with Les Fehmi and reports getting client positive responses that she was not getting before.

I think in the long run one has to become familiar with what is on offer and hopefully develop a creative mix that gels with the practitioner and their client base.

The kind of dialogue that your post will stimulate will also be helpful in this process.

Mark

February 6, 2006

Hi Nick

I no longet get concerned about mapping EEG changes of any kind over time.Once I decide on a protocol and am happy with it ie. think/feel its appropriate for the client and their presenting problem I keep applying it with adjustment within and between session based on spectrum views.

The only statement I am making is that I think bipolar placement is differentiating the hemispheres and promoting left dominace.

Two channel hemispheric training will no doubt do that with appropriate setting and it could also be the end result of some one channel training.

Mark

Re: Re: Increasing coherence?> > > > > Larry,> > > > I'm far from being an expert in this area myself, but a few comments:> > > >> What said seemed basically correct to me; but I would add that inter-hemispheric training, or for that matter any form of bi-polar recording, is likely to affect coherence - but also may not even when the similarity between two sites is affected. > > > > Actually what the bipolar training, interhemispheric or not, is most likely to affect is PHASE, not coherence. When two signals are in phase, subtracting one from the other will result in a very low value; when they are 180 degrees out of phase, it will result in a large value. So, in addition to training down the difference in raw amplitude, it is highly likely that downtraining bipolar signals will move the signals IN phase and uptraining them will move them OUT of phase. Since coherence is independent of phase, it will not necessarily be affected.> > > >> I think one factor that favors the use of bi-polar, one channel recording is that it by-passes the contaminating effect of the 2 reference recording sites that are used in 2 channel coherence training, one for each of the 2 "active" recording sites. . These effects can be large. One particularly huge effect that should be totally avoided is using a linking connector to provide the same reference to each channel. This greatly elevates coherence values because the linkage of the references automatically makes the two signals more similar.> > > > > > Pete> Here I would disagree and say that when comparing two sites against a common reference actually INCREASES the quality of the reading, since we are truly comparing what is happening at the two sites, independent of what might be happening differently at the two reference sites.> > > >

February 6, 2006

From what I can tell, many folks follow Sue Othmer in training bipolar montages

first for everything. For some it rises almost to the point of a fetish. Some

of those may be quite disturbed to hear that now, after a few years with true

2-channel training available to her, Sue is presenting at Winter Brain her new

approach even to interhemispheric training, which is...2-channel training!

Certainly folks would not contine to do something if there weren't getting

results with it, so I assume there is a power to that kind of training.

But for everyone you find who says that " binding " two sites is critical by

increasing phase relationships between them (assuming you don't already have

hyper-coherence between them), you can find at least one who says that reducing

variability in the EEG is critical. So you have 's entropy

training, Val Brown's emergent variability training, squashes, squishes, etc.

My preference, as always, is to avoid for myself any kind of " religious "

adherence to any approach. Whenever I hear someone say that they ALWAYS get

results with ANY approach, I have to turn down the alarm on my BS meter. I

don't hear Len Ochs, or Sue, or or--well, maybe Val sometimes--saying

these things, though some of their followers may. I use bipolar montages for

some types of training as a matter of faith (because I don't know what they are

doing, but they do it fairly predictably) with certain kinds of issues like

tics, compulsivity, etc. I liked Sue's approach as presented at Winter

Brain--and Tom Collura's, which was a variant of it--two two channel training,

and will probably try some of that. And I obviously like standard SMR and Beta

(percent) training. And I really like two-channel squashes, squishes and

windowed squashes. Some may call me a dilettante; I prefer to think of it as

adherence to " what works " rather than to any one system.

I too will be interested in hearing from others on the list about their

experience.

Pete

>

> From: NICK MAMMANO <nickmammano@...>

> Date: 2006/02/06 Mon AM 04:44:10 EST

>

> Subject: Re: Re: Increasing coherence?

>

> Pete:

> You're right of course. I should not have implied that bipolar montages

MUST be training phase (or that the amplitude differences observed MUST be a

result of phase changes). But even if phase shifts are only one of the options,

the possible connection to " time binding " , brain rhythms and to the compelling

arguments of Othmer lit a bulb. I don't any longer view bipolar montages as a

" poor man's method " , or something " less " than the " ear clip " referential

montages.

>

> The proof is in the pudding I guess. There must be a huge amount of " data "

among all the professional trainers who are members of this list. Based on

their experience, and yours, I wonder what the general view is on the relative

benefits of each method.

>

> nick mammano

>

> Van Deusen <pvdtlc@...> wrote:

> Nick,

>

> I think it's safe to say that monopolar and bipolar montages are fundamentally

different in what they measure and do. However, if I said that bipolar inhibit

montages must be training phase, then I did not mean to say that. That's ONE of

the options. It could also simply be training one amplitude down relative to

the other. Sometimes a cigar is just a cigar.

>

> That's what many find discomfiting about bipolar montages: they could be

doing a variety of things, but there simply is no way to KNOW.

>

> Pete

>

> >

> > From: NICK MAMMANO <nickmammano@...>

> > Date: 2006/02/05 Sun PM 01:24:53 EST

> >

> > Subject: Re: Re: Increasing coherence?

> >

> > I bought into the argument that 2C training was preferred to single channel

training – that the latter was the “poor man’s method” or whatever, or that you

get more bang for the buck with 2 channel training, etc. But Pete’s recent

reference to amplitudes and phase differences makes me wonder if simpler single

channel bipolar training might be the way to go in many cases.

> >

> > Pete’s point: (really fundamental physics) it is the phase difference

between the signals from two sites in bipolar training, eg. C3C4g, that is being

trained up or down.

> >

> > Well phase differences have to do with the TIMING between the two signals.

And this reminded me of a very challenging (and persuasive) article by Othmer

(“Quantitative EEG and Neurofeedback” , & Arbabanel, Eds, pp262--)in which

he argues that what he calls “time binding” may be THE MOST IMPORTANT

consideration in EEG feedback training.

> >

> > He quotes “what fires together, wires together” where “wiring” can refer

to short transient or “functional” connections as well as the more permanent

hard wired ones.

> >

> > “Firing together” is another way of saying the signals (that is the EEG

“brainwaves”) are in phase. Presumably there are signals (theta, hibeta?) that

are best left out of phase by training down.

> >

> > Thus is it correct, in light of the above, to infer that the bipolar

protocols (or montages) are in fact FUNDAMENTALLY different from the referential

ones?

> >

> > nick

> >

> > Larry <llewis@...> wrote: Pete,

> >

> > I said that bi-polar training may not affect coherence; you argue that it

is likely not to.

> >

> > Pete: > Actually what the bipolar training, interhemispheric or not, is

most likely to affect is PHASE, not coherence. When two signals are in phase,

subtracting one from the other will result in a very low value; when they are

180 degrees out of phase, it will result in a large value. So, in addition to

training down the difference in raw amplitude, it is highly likely that

downtraining bipolar signals will move the signals IN phase and uptraining them

will move them OUT of phase. Since coherence is independent of phase, it will

not necessarily be affected

> >

> > This seems correct except, perhaps, for the following considerations.

> >

> > It has always seemed to me that if a parameter of the EEG is trainable,

e.g. amplitude; it must be because the brain self-responds to such changes on

its own and then correlates information from its own response (or

self-monitoring) with the neurofeedback information. Its as if the brain says

to itself: " Hey! What's going on here, I notice that whenever I register change

X, something sounds out there in the environment! This never happened before.! "

And this draws the brain's attention to the neurofeedback signal; and the brain

tries to understand this novelty better (attention is always drawn to novelty)

by playing with it.

> >

> > Assuming that something like this is the case, it is likely that the brain

responds more to some parameters than others. The fact that phase is very

difficult to train suggests that the brain may not do much self-monitoring of

this particular parameter; and as a result what you argue may not be the case.

The effect of bi-polar training on phase similarity might be quite weak. I think

this is something that needs empirical testing. The brain will not relate to the

protocol ideas we have about it if those ideas are at variance with how the

brain actually functions.

> >

> > I would disagree with your second point.

> >

> > Pete::> Here I would disagree and say that when comparing two sites

against a common reference actually INCREASES the quality of the reading, since

we are truly comparing what is happening at the two sites, independent of what

might be happening differently at the two reference sites.

> >

> > Whether using a single or common references, one is recording both