"quieting mind chatter / self-referential thoughts at the PCC / Precuneus" - for me this is a bit of a holy grail as a practitioner, because rumination is such a difficult clinical problem. I'm interested in the work of Tomas Ros (Ros et al. Mind over chatter: Plastic up-regulation of the fMRI salience network directly after EEG neurofeedback). He used an alpha inhibit protocol at Pz to reduce chatter - though apparently it may have worked due to a rebound effect.
Very familiar with Ros' alpha inhibit. Have you tried it? I've not but the paper is remarkable and groundbreaking. Seemingly counterintuitive, given all the alpha reward stuff done in the past.
My overall impression of this time period in human history, is that all our shadow material and unhealed ancestral patterns are coming up for integration / resolution. Good for us, if we can survive it and get to the other side. Increased clarity and creativity on the horizon. Planet sure needs it, before we completely trash it. :-)
Hi William, yes I tried the alpha down protocol myself the other day and was impressed with it, first impression. And I am reading the Sebern Fisher book, she mentions it. Honestly at this point I'm finding the learning curve on these protocols pretty steep and have resigned myself to being confused for a while. There is both great specificity and also a paradoxical sense that almost anything can work. Heather Hargraves, mentioned above, is from Western Univ (in Canada) where Ruth Lanius and some of Ros' co-authors are at. She has some interesting protocol ideas too, also based around inhibition: decrease 1-20 Hz at Pz for 15 mins. Yes I agree this is timely and necessary!
He used an alpha inhibit protocol at Pz to reduce chatter - though apparently it may have worked due to a rebound effect.
Goes beyond just alpha rebound effects. It's fine tuning LRTC, Long Range Temporal Correlations -- from a realm of excess chaos, into more orderly communications. Especially prevalent in PTSD. Amazing work. Using the same alpha inhibit at Pz protocol.
Abstract
Brain oscillations exhibit long-range temporal correlations (LRTCs), which reflect the regularity of their fluctuations: low values representing more random (decorrelated) while high values more persistent (correlated) dynamics. LRTCs constitute supporting evidence that the brain operates near criticality, a state where neuronal activities are balanced between order and randomness. Here, healthy adults used closed-loop brain training (neurofeedback, NFB) to reduce the amplitude of alpha oscillations, producing a significant increase in spontaneous LRTCs post-training. This effect was reproduced in patients with post-traumatic stress disorder, where abnormally random dynamics were reversed by NFB, correlating with significant improvements in hyperarousal. Notably, regions manifesting abnormally low LRTCs (i.e., excessive randomness) normalized toward healthy population levels, consistent with theoretical predictions about self-organized criticality. Hence, when exposed to appropriate training, spontaneous cortical activity reveals a residual capacity for “self-tuning” its own temporal complexity, despite manifesting the abnormal dynamics seen in individuals with psychiatric disorder. Lastly, we observed an inverse-U relationship between strength of LRTC and oscillation amplitude, suggesting a breakdown of long-range dependence at high/low synchronization extremes, in line with recent computational models. Together, our findings offer a broader mechanistic framework for motivating research and clinical applications of NFB, encompassing disorders with perturbed LRTCs.
I just sent an email to Tomas asking if they have any larger scale clinical studies underway. Will let you know what he responds with.
However I'm still bit confused as to Ros's explanation of Pz Alpha as it relates to mind-wandering. William, am I correct in saying that people who successfully lowered their alpha during alpha down also showed less mind wandering during the oddball task - which I think ran right after the NFB. But was this because of rebound to higher baseline alpha, or because of low alpha? If the latter then JT's protocol for Quiet Mind should increase mind wandering. Understood that different populations may show different or even paradoxical effects but in any case the implications are not totally clear to me at the moment. I'm referring to this presentation which summarizes several of the studies to date:
The above papers did not do multiple sessions nor follow up on clinical responses of PTSD subjects. But did show single session improvements in measurable dynamics such as LRTC.
--
Ros' 2017 work was referenced in this review, but they did not attempt to replicate. Only to use their own DecNef protocol.
One comment that Ros made which I found insightful, is that the alpha rebound effect only occurs in PTSD clients who happen to be deficient (prior to treatment), in posterior alpha. Unclear whether those clients (who are not deficient in alpha), may experience other benefits such as
the remediation of amygdala connectivity dysfunction
fine tuning LRTC, Long Range Temporal Correlations
etc.
My view is that a wide segment of the population is impacted by various small and large traumas throughout life. An effective PTSD protocol could have substantial widespread benefits.
Ros has written extensively on the neuroscience and strategy behind training the alpha desynchronization (inhibit) at Pz. I do believe that an important component of the improvements seen, is the major DMN Default Mode Network hub in the precuneus, just under the Pz location. Self referential and ruminative thought patterns / ego identification are thought to be a primary function of the DMN. And DMN DIS-engages when one is in "flow-states" with an engaging and activating 'task'.
Jud Brewer has written about his research training subjects with fMRI, in a form of neuromeditation. In this protocol he down trains the BOLD (blood oxygen level dependent) fMRI signal at the precuneus / PCC Posterior Cingulate Cortex.
One advantage to the fMRI neurofeedback, is that the BOLD signal is independent of EEG frequency bands. It just reflects metabolism of the brain area. Adapting the PCC-down training to EEG, one could then choose which band or bands to downtrain. And since alpha has such a primary function in whole brain synchronization, it's logical that this is a good place to start. Tarrant's mindfulness protocol either downtrains beta at Pz or uptrains gamma at Pz. It might be possible to do a combination or sequencing of these that is optimized per client.
I just tried the alpha inhibit protocol myself for the first time last night. I was also monitoring (gently, not too focused), the spectrogram activity during the session. With eyes opened, one normally would expect reduced alpha activity at Pz. And that was the case. But what was striking to me in my own spectrogram, was the amount of 'bursting' alpha (low alpha, 8 Hz) and theta (around 7 and 6 Hz), that was going on.
I have decades of eyes-open (zen-style or vipassana style) meditation regular practice. And it's possible this has both lowered my alpha center frequency (found in long-term meditators), and upped the alpha engagement, even with eyes open. But my suspicion is also that the amount of trauma experiences in my past, may correlate with this bursting. And may be a sign that the DMN just will not completely calm down and 'relax' vigilance. I'm going to continue with the daily sessions to see if I notice less stress and worrying about the future / whats on the other side of this pandemic.
"Tarrant's mindfulness protocol either downtrains beta at Pz...." for sure, but his Quiet Mind protocol up trains alpha at Pz, with the objective of "reducing self talk"..... hence my confusion. I went back to the Ros presentation from April where he discusses this in some detail just to confirm I had it the right way round. https://www.youtube.com/watch?time_continue=5&v=2zv2gBZMu9w&feature=emb_logo
22 mins in:
Alpha acts as a gate for external sensory information so high alpha in the visual and motor areas should decrease visual detection, environmental awareness and lead to more internally focused states such as thinking..... The 2013 paper on the alpha down protocol includes a test for mind wandering during an attention task, immediately after NFB. And indeed the people who reduced their alpha had less mind wondering under task. fMRI results seem to confirm the hypothesis that high alpha at Pz = high DMN = high mind wondering.
So all this is to say, I don't quite understand the Quiet Mind protocol. I think the rationale of it is to increase internal self-observation which is presumed to be a function of the DMN? I might have to watch the presentation again to confirm.
And high alpha has always been found in meditators with eye closed. So it's not surprising that Tarrant's Quiet Mind is keying into that same tradition. I'm not sure Tarrant is aware of Tomas Ros alpha desync research. But Tarrant certainly is aware of DMN down training, because his 'mindfulness' protocol downtrains beta at Pz.
Maybe these are just alternate thought streams about the utility of alpha training. My perception is that quieting the DMN is a topic of much research currently, and it meshes nicely with the other normalization effects that Ros is seeing. Tarrant has his own threads he is following, possibly driven by his experiences training clients in neuromeditation. So is less focused on the kind of deep and therapeutic changes that Ros is reporting on.
Would be interested in your further experiences with the Ros alpha inhibit @ Pz. That is what I want to focus on in this thread.
For sure. I've tried the down training one time. First impression was that it was a quite effortful compared to up training, and that it did seem to lead to an interesting shift after the training period rather than during it. In this respect it reminds me of traditional zazen which is kind of a work out for me rather than a relaxation practice. It was very interesting but I'll have to try a few more rounds of it, and maybe compare with up training, before drawing any conclusion. I'll report back.
I tried alpha down again for 5 x 4 min sessions at 60% feedback. I found visual focus to be a reasonably effective strategy during the session. Subjectively, there was an open, alert sensory focus. However, looking at the numbers for these alpha down sessions my first observation is that I wasn't particularly successful at suppressing alpha. In fact, as the sessions accumulated it looks like alpha is fighting back with bursts of up to 50% increase. This is the tiring aspect I think, because I'm squashing these bursts of alpha which are getting bigger over time. By session five my brain feels pooped. So one possibility is that alpha down is essentially a paradoxical intervention, at least for me. After the session, the subjective sense of bright, open sensory focus remains. The mind feels settled.
Paul, thanks. I've done a several 45 minute length sessions so far (each with 15 minute * 3 blocks). I definitely feel it is quieting the DMN, as well as shifting amygdala patterns. Less worrying, seeing more possibilities and potentials. Also feel like it has an impact on 'intuition' circuits, because as DMN gets quiet, more subtle neuronal impulses (especially sensory, subtle sensory) are not drowned out by noise.
Yes I agree the system seems to resist this downtraining in the short term. But longer term there are significant shifts happening. I assume you are tracking long term trend graphs.
re: tiring. As with all neurofeedback, the paradoxical aspect is to not to 'push' excessively. And more relax into the process. I'm just using a mild intention, and not getting thrown off by the pushback. I have both audio and visual feedback going. For audio I use a 'white noise' sound of a stream recorded in a Canadian nature preserve. The volume of the stream tracks the amplitude of the alpha. So it gets louder as it goes up. The goal being to make the stream quieter. For visual I have several graphs going:
raw EEG
filtered EEG 8-12, so alpha spindles are visible.
envelope of alpha (previous graph), which controls audio volume
I actually imagine I'm coming in for a gentle landing with an airplane with the alpha envelope.
Here is the screen of the BrainBay design I'm using, you can see the three graphs mentioned on the previous comment: raw, filtered / spindles, alpha envelope. Last two graphs show long term trend of alpha and beta. Lower left is a moving / scrolling spectrogram of 1 to 20 Hz (scrolls to the right, over several seconds; left margin is current time), you can see the 1 to 20 freq spectrum on the y axis.
This design has a control ('beta weight' box in the design) that adjusts if beta is inhibited. I have that turned off. So that only the alpha at Pz is being fed back via screen and audio. Object then is to bring volume down and land the plane (envelope). You can see that in the alpha trend graph, there are 5 or more sections where the alpha dips down below 5 uV. The number of these downward areas seems to build as I get more practice.
I'm not using 'thresholds' at all. Just the sound of the stream tracks alpha amplitude directly. I think thresholds may actually alter difficulty.
You know, 'neuromore' supports Muse directly. Might be another option. Might depend if you have Muse 1 or Muse 2, I think the data format might have changed.
I have both Muse 1 and 2. If it supports the extra electrode that would be useful. Mind Monitor does - but raw EEG only I think. However - Neuromore does support Mac I see - one of the issues with the other solutions is they are Windows only so need a new laptop. So I'll give it a try, thanks.
There are several types of threshold schemes. Some are called auto-thresholds, that automatically adjust / track during the session. The other threshold type is fixed. In either case, the feedback is generally triggered when the threshold is crossed in the desired direction. This can be a kind of jarring experience in certain cases, depending on how the feedback is presented and where the threshold is set.
I also notice in your photo above, that the alpha amplitude is already quite low (between 1.25 uV and .75 uV. Depending on the sensitivity of your equipment, some EEG starts getting swamped with electronic / amplifier noise at around 1 uV. So it's possible that your setup is adding to the frustration by mixing your actual EEG with noise. It may not be possible to train your alpha below 1 uV. Does your Muse external electrode show generally accurate results at other bands, other locations? For example compared to the frontal and ear sites, where the wires inside the headset are much shorter and possibly less noise prone.
As you can see from my screenshot, my alpha at Pz varies nominally between about 15 uV and 5 uV. With occasional peaks above that to say ~20 uV.
On this tech sheet (I assume for Muse 1), the noise is listed as 2 uV. Yikes. Implying that possibly your sub 2, sub 1 uV alpha has substantial noise component.
Yeah, I don't know much about the signal processing side yet so I can't really comment, I'm still learning this stuff really. But that does sound concerning. And I have noticed that it's very difficult to compare my data with my prof's. Neuromore is up and running now so I have some more options - maybe a Ganglion? My only concern there is on the FTDI fix - I'm on 10.12.6 Mac OS which may not be supported by the fix.
FTDI only applies to Cyton (RFduino dongle). Ganglion uses a different dongle based on BLED112. I believe Cyton can now talk to all Mac's, regardless of macOS version.
See if there is any difference in microvolts with your Muse 2 vs Muse 1. How many microvolts alpha do you get with eyes closed at Pz?
As usual Muse is a morass of issues. According to the GitHub the Muse 2016 is not viable on MacOS for neuromore - it won't connect due to a bug. The 2014 works but then that doesn't help validate the Myndlift as it's 2016 only. I could validate the frontal electrodes a few different ways but I don't know of any other point of comparison for the aux electrode. At some point, my prof is going to demo Myndlift and hopefully compare it to other platforms so maybe I'll get some clarity then. I'll raise the issue of signal to noise with Myndlift support though.
One way to double check the aux electrode, would be to attach it to the forehead, close to one of the headband electrodes. Then the two signal amplitudes and wave-shapes should closely track each other.
I have decades of vipassana / mindfulness meditation (eyes open), which might explain why my eyes-open Pz alpha amplitude is in the ~10uV range. But in Ros' papers, the PTSD subjects he worked with actually started with lower than normal eyes-open Pz alpha. And they were able to inhibit it during neurofeedback. Then it rebound afterwards. So that's why I think your 1 uV alpha at Pz looks suspicious. Can't go much lower than that, to enable inhibiting further. It's possible your 'real' eyes-open Pz alpha is more like 5 uV. Which would be enough room for down training in spite of the Muse noise of 2 uV.
Anyway, lots of question marks. Hopefully Myndlift will have some insights / suggestions. I did some checking on them earlier this evening. An Israeli company with various staff members (not listed on their own site),
Just a thought - could your values be power [(amp)2] whereas Myndlift are just amplitude?
I can try moving the electrode - though I notice that my alpha sessions on F7 and F8 (Muse electrode )are in the same kind of range. In fact everything's in that kind of range, alpha, beta, etc. I ran the assessment protocol on a friend and his eyes closed central alpha was like 3 uV. The amplitude scale on the assessment report only goes from 0 - 5 uV actually. So I think maybe there is some kind of scaling issue.
A colleague in Seattle uses a clinical grade system and uses Myndlift for remote work. He figured that ML was 90% valid compared to the other system and that was close enough - but I've no idea how he compared these.
BrainBay amplitudes are in plain microvolts, not power (uV squared).
In terms of head locations for strongest eyes closed alpha, it's generally parietal and occipital. Alpha is low frontally in most people. See what your eyes-closed alpha looks like at Pz. Seems like 5 uV would not be unusual. O1 and O2 are other common high alpha locations.
If your 'assessment' range is 0 to 5 uV, then the noise spec of Muse would say that values 0, 1, and 2 are already suspected to be noise contaminated. That only leaves range of 3 to 5 uV, pretty darn narrow.
I ran this issue by my prof and by ML support. Response from prof is that the Muse amplifier is decent for what it is. There will be some noise with dry electrodes. Those low amplitude values are more a consequence of the quantification algorithm used. Support: The Muse noise spec is RMS and applies to the whole signal whereas the amp values are spectral density of part of the signal, so in practice a 4 Hz frequency range would have less than 0.17 noise (RMS amp). So I think I'm okay with that however there is a more general problem of values not really being comparable across hardware/software. The ML assessment package actually only looks at relative values across the spectrum - ratios, etc. - but even that is somewhat problematic:
EEG Theta/Beta Ratio Calculations Differ Between Various EEG Neurofeedback and Assessment Software Packages: Clinical Interpretation - Kerson et al. 2019
I just got the email reply back from Tomas Ros. They have finished a randomized controlled trial of the protocol with PTSD clients, each subject receiving multiple sessions. And it was successful. Paper will be published this year. It was a collaboration with Ruth Lanius. Lanius and Ros have co-authored many papers.
I continue to do a daily session of 3 * 15 min session. 45 min total. Seeing some very interesting effects, particularly in dreams, life force energetics opening up (chi flow), etc.
I read your 'response' from 'professor' and Myndlift. I have to say that the explanation of the Muse noise spec does not make much sense to me, from my understanding of EEG signal processing. But then I did not really grok the weird pdf specification from Muse regarding the 2 uV noise factor. Why would they say that in the first place, it's certainly prone to misunderstanding.
If Ros was successful in his trial, and you are not able to run his protocol whatsoever due to anomalies in the Muse / Myndlift, then that is a statement about your hardware / software combination. Because it is working quite noticeably in my own case.
For reference here is a slide showing typical CAPS score range. In Ros' trial, (next to last viewgraph), pre trial scores were in the ~37 range, severe PTSD. And after all 20 sessions, decreased down to ~25 range, low end of the 'moderate' category. And almost into 'mild' range at ~22.
"If Ros was successful in his trial, and you are not able to run his protocol whatsoever due to anomalies in the Muse / Myndlift, then that is a statement about your hardware / software combination. Because it is working quite noticeably in my own case."
Well I don't think those are the only inferences available to us (i.e. that it's either an equipment problem or else the protocol's wrong) - Ros's data shows a significant effect across aggregated data but individual data vary widely. In the 2013 paper out of 17 NFB participants some had alpha up, some down, some stayed the same. (Also true of the sham group). I think the other papers show individual variation too. I've tried it only twice so too soon for me to conclude anything much about the equipment or the protocol.
Comments
"quieting mind chatter / self-referential thoughts at the PCC / Precuneus" - for me this is a bit of a holy grail as a practitioner, because rumination is such a difficult clinical problem. I'm interested in the work of Tomas Ros (Ros et al. Mind over chatter: Plastic up-regulation of the fMRI salience network directly after EEG neurofeedback). He used an alpha inhibit protocol at Pz to reduce chatter - though apparently it may have worked due to a rebound effect.
Very familiar with Ros' alpha inhibit. Have you tried it? I've not but the paper is remarkable and groundbreaking. Seemingly counterintuitive, given all the alpha reward stuff done in the past.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5051955/
Ros' Twitter feed is also full of good stuff. His mentor Ruth Lanius is legend.
https://twitter.com/neuromeditate
https://www.google.com/search?q=ruth+lanius
Are you familiar with Sebern Fisher's Developmental Trauma work?
https://www.sebernfisher.com/
My overall impression of this time period in human history, is that all our shadow material and unhealed ancestral patterns are coming up for integration / resolution. Good for us, if we can survive it and get to the other side. Increased clarity and creativity on the horizon. Planet sure needs it, before we completely trash it. :-)
Hi William, yes I tried the alpha down protocol myself the other day and was impressed with it, first impression. And I am reading the Sebern Fisher book, she mentions it. Honestly at this point I'm finding the learning curve on these protocols pretty steep and have resigned myself to being confused for a while. There is both great specificity and also a paradoxical sense that almost anything can work. Heather Hargraves, mentioned above, is from Western Univ (in Canada) where Ruth Lanius and some of Ros' co-authors are at. She has some interesting protocol ideas too, also based around inhibition: decrease 1-20 Hz at Pz for 15 mins. Yes I agree this is timely and necessary!
Goes beyond just alpha rebound effects. It's fine tuning LRTC, Long Range Temporal Correlations -- from a realm of excess chaos, into more orderly communications. Especially prevalent in PTSD. Amazing work. Using the same alpha inhibit at Pz protocol.
https://academic.oup.com/cercor/article/27/10/4911/3056439
"Neurofeedback Tunes Scale-Free Dynamics in Spontaneous Brain Activity"
I just sent an email to Tomas asking if they have any larger scale clinical studies underway. Will let you know what he responds with.
Thank you I'd be very interested in that
However I'm still bit confused as to Ros's explanation of Pz Alpha as it relates to mind-wandering. William, am I correct in saying that people who successfully lowered their alpha during alpha down also showed less mind wandering during the oddball task - which I think ran right after the NFB. But was this because of rebound to higher baseline alpha, or because of low alpha? If the latter then JT's protocol for Quiet Mind should increase mind wandering. Understood that different populations may show different or even paradoxical effects but in any case the implications are not totally clear to me at the moment. I'm referring to this presentation which summarizes several of the studies to date:
Here is a partial list of Tomas Ros' publications on alpha desynchronization (inhibit) effects.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5051955/ [2013, Mind over chatter]
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4442612/ [2014 Plastic modulation of PTSD resting-state networks]
https://www.frontiersin.org/articles/10.3389/fnhum.2014.01008/full [2014, Tuning pathological brain oscillations]
https://academic.oup.com/cercor/article/27/10/4911/3056439 [2017, Neurofeedback Tunes Scale-Free Dynamics]
The above papers did not do multiple sessions nor follow up on clinical responses of PTSD subjects. But did show single session improvements in measurable dynamics such as LRTC.
--
Ros' 2017 work was referenced in this review, but they did not attempt to replicate. Only to use their own DecNef protocol.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6650780/
"Current Status of Neurofeedback for Post-traumatic Stress Disorder: A Systematic Review..."
That review also referenced Ros' 2016 paper, which I had missed seeing on the first list,
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5030332/ [2016, Alpha oscillation neurofeedback modulates amygdala...]
One comment that Ros made which I found insightful, is that the alpha rebound effect only occurs in PTSD clients who happen to be deficient (prior to treatment), in posterior alpha. Unclear whether those clients (who are not deficient in alpha), may experience other benefits such as
My view is that a wide segment of the population is impacted by various small and large traumas throughout life. An effective PTSD protocol could have substantial widespread benefits.
Ros has written extensively on the neuroscience and strategy behind training the alpha desynchronization (inhibit) at Pz. I do believe that an important component of the improvements seen, is the major DMN Default Mode Network hub in the precuneus, just under the Pz location. Self referential and ruminative thought patterns / ego identification are thought to be a primary function of the DMN. And DMN DIS-engages when one is in "flow-states" with an engaging and activating 'task'.
Jud Brewer has written about his research training subjects with fMRI, in a form of neuromeditation. In this protocol he down trains the BOLD (blood oxygen level dependent) fMRI signal at the precuneus / PCC Posterior Cingulate Cortex.
https://www.pnas.org/content/108/50/20254
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4529365/
One advantage to the fMRI neurofeedback, is that the BOLD signal is independent of EEG frequency bands. It just reflects metabolism of the brain area. Adapting the PCC-down training to EEG, one could then choose which band or bands to downtrain. And since alpha has such a primary function in whole brain synchronization, it's logical that this is a good place to start. Tarrant's mindfulness protocol either downtrains beta at Pz or uptrains gamma at Pz. It might be possible to do a combination or sequencing of these that is optimized per client.
I just tried the alpha inhibit protocol myself for the first time last night. I was also monitoring (gently, not too focused), the spectrogram activity during the session. With eyes opened, one normally would expect reduced alpha activity at Pz. And that was the case. But what was striking to me in my own spectrogram, was the amount of 'bursting' alpha (low alpha, 8 Hz) and theta (around 7 and 6 Hz), that was going on.
I have decades of eyes-open (zen-style or vipassana style) meditation regular practice. And it's possible this has both lowered my alpha center frequency (found in long-term meditators), and upped the alpha engagement, even with eyes open. But my suspicion is also that the amount of trauma experiences in my past, may correlate with this bursting. And may be a sign that the DMN just will not completely calm down and 'relax' vigilance. I'm going to continue with the daily sessions to see if I notice less stress and worrying about the future / whats on the other side of this pandemic.
William
"Tarrant's mindfulness protocol either downtrains beta at Pz...." for sure, but his Quiet Mind protocol up trains alpha at Pz, with the objective of "reducing self talk"..... hence my confusion. I went back to the Ros presentation from April where he discusses this in some detail just to confirm I had it the right way round.
https://www.youtube.com/watch?time_continue=5&v=2zv2gBZMu9w&feature=emb_logo
22 mins in:
Alpha acts as a gate for external sensory information so high alpha in the visual and motor areas should decrease visual detection, environmental awareness and lead to more internally focused states such as thinking..... The 2013 paper on the alpha down protocol includes a test for mind wandering during an attention task, immediately after NFB. And indeed the people who reduced their alpha had less mind wondering under task. fMRI results seem to confirm the hypothesis that high alpha at Pz = high DMN = high mind wondering.
So all this is to say, I don't quite understand the Quiet Mind protocol. I think the rationale of it is to increase internal self-observation which is presumed to be a function of the DMN? I might have to watch the presentation again to confirm.
Alpha reward training has been a part of neurofeedback going all the way back to the early days of Kamiya (1958).
https://www.google.com/search?q=kamiya+alpha+neurofeedback
And high alpha has always been found in meditators with eye closed. So it's not surprising that Tarrant's Quiet Mind is keying into that same tradition. I'm not sure Tarrant is aware of Tomas Ros alpha desync research. But Tarrant certainly is aware of DMN down training, because his 'mindfulness' protocol downtrains beta at Pz.
Maybe these are just alternate thought streams about the utility of alpha training. My perception is that quieting the DMN is a topic of much research currently, and it meshes nicely with the other normalization effects that Ros is seeing. Tarrant has his own threads he is following, possibly driven by his experiences training clients in neuromeditation. So is less focused on the kind of deep and therapeutic changes that Ros is reporting on.
Would be interested in your further experiences with the Ros alpha inhibit @ Pz. That is what I want to focus on in this thread.
For sure. I've tried the down training one time. First impression was that it was a quite effortful compared to up training, and that it did seem to lead to an interesting shift after the training period rather than during it. In this respect it reminds me of traditional zazen which is kind of a work out for me rather than a relaxation practice. It was very interesting but I'll have to try a few more rounds of it, and maybe compare with up training, before drawing any conclusion. I'll report back.
I tried alpha down again for 5 x 4 min sessions at 60% feedback. I found visual focus to be a reasonably effective strategy during the session. Subjectively, there was an open, alert sensory focus. However, looking at the numbers for these alpha down sessions my first observation is that I wasn't particularly successful at suppressing alpha. In fact, as the sessions accumulated it looks like alpha is fighting back with bursts of up to 50% increase. This is the tiring aspect I think, because I'm squashing these bursts of alpha which are getting bigger over time. By session five my brain feels pooped. So one possibility is that alpha down is essentially a paradoxical intervention, at least for me. After the session, the subjective sense of bright, open sensory focus remains. The mind feels settled.
Paul, thanks. I've done a several 45 minute length sessions so far (each with 15 minute * 3 blocks). I definitely feel it is quieting the DMN, as well as shifting amygdala patterns. Less worrying, seeing more possibilities and potentials. Also feel like it has an impact on 'intuition' circuits, because as DMN gets quiet, more subtle neuronal impulses (especially sensory, subtle sensory) are not drowned out by noise.
Yes I agree the system seems to resist this downtraining in the short term. But longer term there are significant shifts happening. I assume you are tracking long term trend graphs.
re: tiring. As with all neurofeedback, the paradoxical aspect is to not to 'push' excessively. And more relax into the process. I'm just using a mild intention, and not getting thrown off by the pushback. I have both audio and visual feedback going. For audio I use a 'white noise' sound of a stream recorded in a Canadian nature preserve. The volume of the stream tracks the amplitude of the alpha. So it gets louder as it goes up. The goal being to make the stream quieter. For visual I have several graphs going:
I actually imagine I'm coming in for a gentle landing with an airplane with the alpha envelope.
William
Here is the screen of the BrainBay design I'm using, you can see the three graphs mentioned on the previous comment: raw, filtered / spindles, alpha envelope. Last two graphs show long term trend of alpha and beta. Lower left is a moving / scrolling spectrogram of 1 to 20 Hz (scrolls to the right, over several seconds; left margin is current time), you can see the 1 to 20 freq spectrum on the y axis.
This design has a control ('beta weight' box in the design) that adjusts if beta is inhibited. I have that turned off. So that only the alpha at Pz is being fed back via screen and audio. Object then is to bring volume down and land the plane (envelope). You can see that in the alpha trend graph, there are 5 or more sections where the alpha dips down below 5 uV. The number of these downward areas seems to build as I get more practice.
It's a challenge to get comparable level of detail from Myndlift but here is some data from the last of the 5 rounds this morning.
What I think this shows is alpha rebound from the previous sessions and then me working to get it back below threshold for the rest of the session.
I'm not using 'thresholds' at all. Just the sound of the stream tracks alpha amplitude directly. I think thresholds may actually alter difficulty.
You know, 'neuromore' supports Muse directly. Might be another option. Might depend if you have Muse 1 or Muse 2, I think the data format might have changed.
https://www.neuromore.com/
I have both Muse 1 and 2. If it supports the extra electrode that would be useful. Mind Monitor does - but raw EEG only I think. However - Neuromore does support Mac I see - one of the issues with the other solutions is they are Windows only so need a new laptop. So I'll give it a try, thanks.
There are several types of threshold schemes. Some are called auto-thresholds, that automatically adjust / track during the session. The other threshold type is fixed. In either case, the feedback is generally triggered when the threshold is crossed in the desired direction. This can be a kind of jarring experience in certain cases, depending on how the feedback is presented and where the threshold is set.
I also notice in your photo above, that the alpha amplitude is already quite low (between 1.25 uV and .75 uV. Depending on the sensitivity of your equipment, some EEG starts getting swamped with electronic / amplifier noise at around 1 uV. So it's possible that your setup is adding to the frustration by mixing your actual EEG with noise. It may not be possible to train your alpha below 1 uV. Does your Muse external electrode show generally accurate results at other bands, other locations? For example compared to the frontal and ear sites, where the wires inside the headset are much shorter and possibly less noise prone.
As you can see from my screenshot, my alpha at Pz varies nominally between about 15 uV and 5 uV. With occasional peaks above that to say ~20 uV.
On this tech sheet (I assume for Muse 1), the noise is listed as 2 uV. Yikes. Implying that possibly your sub 2, sub 1 uV alpha has substantial noise component.
https://cdn.shopify.com/s/files/1/0414/4905/t/2/assets/muse-tech-spec-sheet.pdf
Another spec sheet, also listing the 2 uV noise factor,
https://www.eegsales.com/Shared/images/General Use/PDF Files/Muse_Technical_Specs.pdf
Yeah, I don't know much about the signal processing side yet so I can't really comment, I'm still learning this stuff really. But that does sound concerning. And I have noticed that it's very difficult to compare my data with my prof's. Neuromore is up and running now so I have some more options - maybe a Ganglion? My only concern there is on the FTDI fix - I'm on 10.12.6 Mac OS which may not be supported by the fix.
FTDI only applies to Cyton (RFduino dongle). Ganglion uses a different dongle based on BLED112. I believe Cyton can now talk to all Mac's, regardless of macOS version.
See if there is any difference in microvolts with your Muse 2 vs Muse 1. How many microvolts alpha do you get with eyes closed at Pz?
As usual Muse is a morass of issues. According to the GitHub the Muse 2016 is not viable on MacOS for neuromore - it won't connect due to a bug. The 2014 works but then that doesn't help validate the Myndlift as it's 2016 only. I could validate the frontal electrodes a few different ways but I don't know of any other point of comparison for the aux electrode. At some point, my prof is going to demo Myndlift and hopefully compare it to other platforms so maybe I'll get some clarity then. I'll raise the issue of signal to noise with Myndlift support though.
One way to double check the aux electrode, would be to attach it to the forehead, close to one of the headband electrodes. Then the two signal amplitudes and wave-shapes should closely track each other.
I have decades of vipassana / mindfulness meditation (eyes open), which might explain why my eyes-open Pz alpha amplitude is in the ~10uV range. But in Ros' papers, the PTSD subjects he worked with actually started with lower than normal eyes-open Pz alpha. And they were able to inhibit it during neurofeedback. Then it rebound afterwards. So that's why I think your 1 uV alpha at Pz looks suspicious. Can't go much lower than that, to enable inhibiting further. It's possible your 'real' eyes-open Pz alpha is more like 5 uV. Which would be enough room for down training in spite of the Muse noise of 2 uV.
Anyway, lots of question marks. Hopefully Myndlift will have some insights / suggestions. I did some checking on them earlier this evening. An Israeli company with various staff members (not listed on their own site),
https://www.linkedin.com/company/myndlift/ [click link, 'see 15 employees']
Just a thought - could your values be power [(amp)2] whereas Myndlift are just amplitude?
I can try moving the electrode - though I notice that my alpha sessions on F7 and F8 (Muse electrode )are in the same kind of range. In fact everything's in that kind of range, alpha, beta, etc. I ran the assessment protocol on a friend and his eyes closed central alpha was like 3 uV. The amplitude scale on the assessment report only goes from 0 - 5 uV actually. So I think maybe there is some kind of scaling issue.
A colleague in Seattle uses a clinical grade system and uses Myndlift for remote work. He figured that ML was 90% valid compared to the other system and that was close enough - but I've no idea how he compared these.
BrainBay amplitudes are in plain microvolts, not power (uV squared).
In terms of head locations for strongest eyes closed alpha, it's generally parietal and occipital. Alpha is low frontally in most people. See what your eyes-closed alpha looks like at Pz. Seems like 5 uV would not be unusual. O1 and O2 are other common high alpha locations.
If your 'assessment' range is 0 to 5 uV, then the noise spec of Muse would say that values 0, 1, and 2 are already suspected to be noise contaminated. That only leaves range of 3 to 5 uV, pretty darn narrow.
Yeah that is concerning.
I ran this issue by my prof and by ML support. Response from prof is that the Muse amplifier is decent for what it is. There will be some noise with dry electrodes. Those low amplitude values are more a consequence of the quantification algorithm used. Support: The Muse noise spec is RMS and applies to the whole signal whereas the amp values are spectral density of part of the signal, so in practice a 4 Hz frequency range would have less than 0.17 noise (RMS amp). So I think I'm okay with that however there is a more general problem of values not really being comparable across hardware/software. The ML assessment package actually only looks at relative values across the spectrum - ratios, etc. - but even that is somewhat problematic:
EEG Theta/Beta Ratio Calculations Differ Between Various EEG Neurofeedback and Assessment Software Packages: Clinical Interpretation - Kerson et al. 2019
I just got the email reply back from Tomas Ros. They have finished a randomized controlled trial of the protocol with PTSD clients, each subject receiving multiple sessions. And it was successful. Paper will be published this year. It was a collaboration with Ruth Lanius. Lanius and Ros have co-authored many papers.
I continue to do a daily session of 3 * 15 min session. 45 min total. Seeing some very interesting effects, particularly in dreams, life force energetics opening up (chi flow), etc.
I read your 'response' from 'professor' and Myndlift. I have to say that the explanation of the Muse noise spec does not make much sense to me, from my understanding of EEG signal processing. But then I did not really grok the weird pdf specification from Muse regarding the 2 uV noise factor. Why would they say that in the first place, it's certainly prone to misunderstanding.
If Ros was successful in his trial, and you are not able to run his protocol whatsoever due to anomalies in the Muse / Myndlift, then that is a statement about your hardware / software combination. Because it is working quite noticeably in my own case.
Ros' December powerpoint, with a few details of the clinical trial:
https://www.dropbox.com/s/vhkcz5n5ur8i4ak/rtFiN 2019 Maastricht - Plastic Modulation of EEG Resting-State Dynamics following Neurofeedback.pptx?dl=0#
Presented at the rtFIN 2019 conference,
https://www.rtfin2019.org/107020
For reference here is a slide showing typical CAPS score range. In Ros' trial, (next to last viewgraph), pre trial scores were in the ~37 range, severe PTSD. And after all 20 sessions, decreased down to ~25 range, low end of the 'moderate' category. And almost into 'mild' range at ~22.
Well I don't think those are the only inferences available to us (i.e. that it's either an equipment problem or else the protocol's wrong) - Ros's data shows a significant effect across aggregated data but individual data vary widely. In the 2013 paper out of 17 NFB participants some had alpha up, some down, some stayed the same. (Also true of the sham group). I think the other papers show individual variation too. I've tried it only twice so too soon for me to conclude anything much about the equipment or the protocol.