Paul, thanks. Well, also note that this new randomized controlled trial ran for 20 sessions, probably a few sessions per week. All his previous papers were just alpha changes observable after a single session. Since your last posted (2nd session) alpha level was in the range .75 uV to 1.25 uV, that is going to be difficult training any lower than that. 1uV EEG is just too close to noise levels, regardless of who makes the equipment or software.
Another thought I had about your Muse / Myndlife 1 uV alpha @ Pz. All EEG is measured differentially. Meaning that the closer the reference electrode is to the site being measured, the lower will be the difference voltage.
Most typical neurofeedback is using the reference on the ear lobe. One or both, if linked.
Muse reference location is not completely clear to me. Is it at the forehead, or at the earpieces? (To better match traditional measured amplitudes?) Earpiece is touching basically at the upper rear portion of the ear, where the conductive rubber 'earpiece' touches the ear surface. This is considerably higher than the ear lobes. And actually over a portion of the cortex / temporal lobes. (Very close to T3, T4 sites.) Whereas ear lobes are mostly OFF the cortex.
I do know that with traditional neurofeedback, ear lobe references, that measured amplitudes at T3, T4 are MUCH lower than at other sites, for just this reason. With Muse and Pz, (if ear references) you're going to see the same type of effect.
A last observation is that Muse ear contacts are conductive rubber. In our tests at OpenBCI with various conductive rubber materials, these typically have a resistance much HIGHER than metallic or silver chloride type electrodes. Further reducing the differential voltage measured. All the other Muse electrodes are at the forehead and metallic.
HI William, I think a good person to answer this would be James Clutterbuck at Mind Monitor. He is not connected with Interaxon but has developed 3P software for Muse for years now and runs a forum at https://mind-monitor.com/ and is quite responsive. He answered a bunch of questions from me when I was trying to get Muse 1 connected to MAX. From what I recall, Muse has a ground and a DRL at the ears however at least in 3P software such as Myndlift the reference in a unipolar placement can be chosen. So for example in the current case I choose Pz as the active channel and then can choose from Tp9, AF7, FPz, AF8, Tp10 as reference. So at least these could be compared and differences noted - however this is not my current focus. I also think Myndlift support might also be quite responsive to a question from a non-subscriber: Try [email protected]. If not, I'm happy to forward a question from you.
As to the larger question about the reliability of any given combo of NFB hardware/software - from my point of view it's somewhat unanswerable. I don't see much independent benchmarking of these systems, and so even with a 10K clinical system for someone like me it would largely be a matter of trust. Based on my experience in the software industry prior to becoming a psychologist, I would say that the big clinical system providers are not trying too hard b/c they have a captive market and often cozy relationships b/w the big clinical names and the equipment producers, so I'm not really convinced those systems are necessarily better. I think it all needs a shake-up from open-source projects like this one, and the consumer market, and maybe things will improve. My strategy is to figure out if there is anything to it clinically - i.e. symptom reduction as promised. Obviously that's predicated on the equipment working reasonably well, and it's complicated by the fact that even sham NFB sometimes "works". But it's not a perfect world - If I can get NFB to work even a little bit with the most modest equipment then maybe I'll look at a more sophisticated system.
Other than Myndlift / Muse, all other commercial / clinical neurofeedback equipment and home training neurofeedback equipment (such as Pocket Neurobics, used with www.brain-trainer.com) that I am aware of -- use ear lobe as default reference. This is also true of the numerous QEEG normative database systems, such as Neuroguide, BrainDX, etc. This was the case when I used Dr. Richard Soutar's NewMind QEEG system, with the Brainmaster Discovery (19 channel) amplifier; with Electro-cap elastic head cap.
So there is some comparability between these systems, if the same reference is used. Research BCI systems are more variable and are not as standardized on ear lobe references. Some of these use references on the scalp, then use mathematical electrode 're-referencing' algorithms, to (for example) remove distortions caused by having a single reference. What are called "reference-free" systems.
Regarding the possible electrodes in Muse that can act as reference, apparently the three central electrodes on the forehead are all part of the 'default' reference system. Leaving these four locations,
So indeed, TP9 and TP10 are the rubber ear contacts. And the rubber is a long strip, with the main contact at the upper part of the ear. That's cool that the various Muse apps can choose the mathematically derived reference location. As long as users / developers can agree on what are the 'standard' reference locations across various applications. Especially those that attempt to replicate QEEG capabilities using 4 + (1 aux) channels.
Getting back to the Tomas Ros alpha down at Pz protocol, I'm continuing to see forward motion with it on my trend graphs. One thing I see is increasing lengths of time where the microvolts stay relatively low. Still have bursts going on, but the low time spans are increasing from session to session. Also seeing less noise overall in the theta and delta bands underneath alpha.
There seems to be a meshing of the neurofeedback protocol, and the daily qigong and mindfulness meditation practice that I do. My hunch is that "less noise in the system" (primarily DMN), frees up more activity to happen in domains that increase life-force and positive interconnections. (As opposed to long standing wiring related to past PTSD experiences that may have reduced chi flow.)
@wjcroft said:
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.
Hi! Would be nice if you can share the OpenVibe script of your Alpha protocol (html i guess). - Best, Pascal.
@wjcroft said:
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.
Hi! Would be nice if you can share the OpenVibe script of your Alpha protocol (html i guess). - Best, Pascal.
Sure you already read the paper by Nicolson, Ros, and Lanius:
https://www.sciencedirect.com/science/article/pii/S2213158220303272
"A randomized, controlled trial of alpha-rhythm EEG neurofeedback in posttraumatic stress disorder: A preliminary investigation showing evidence of decreased PTSD symptoms and restored default mode and salience network connectivity using fMRI"
On the Brain-Trainer forum, a discussion started yesterday:
@mikhaill, thanks. I did probably 10 sessions (over a couple weeks) in the summer of 2020. But did not follow through. I 'thought' I noticed some shifts, but it was subtle. I had some significant DTD (Developmental Trauma) in early childhood. That resulted in strong introversion and other characteristics. Although I feel I've left that shell, there are still subtle remnants that are hard to shake. I'll give this another round of training.
Another factor in any alpha training (up or down), is the placement of the alpha band. While traditionally thought of as 8 to 12 Hz, this can also be highly individualized. And those who have done large amounts of meditation (as I have), frequently have a lower band center point. Determining the accurate per client center frequency, may offer stronger results.
Thanks for posting the final RTC from Lanius, Ros, et al. I did exchange several emails with Tomas last year, when he said the paper was forthcoming. The link to Pete van Deusen's Brain-Trainer forum, also grateful. I think Pete's comments are a bit harsh. Lanius, Ros, and Sebern Fisher (book: "Neurofeedback in the Treatment of Developmental Trauma"), have been working in this domain of DTD, PTSD for decades. So I'm inclined to listen to Sebern when she states this is groundbreaking.
@wjcroft, thank you for reminding me about individual alpha. As I understand it can be found with a test that includes eyes-open and eyes-closed EEG recordings.
Have DTD too, but I'm only at start of the way, a total of 40 sessions. For me, DMN training (theta, alpha, and gamma synchronization at AFz and Pz) turned out to be very useful — for significant reduction dissociation, false-ego weakening and strengthening more real self. It's interesting, that according to Wikipedia, DMN has changed in long-term meditators:
Structural changes in areas of the DMN such as the temporoparietal junction, posterior cingulate cortex, and precuneus have been found in meditation practitioners. There is reduced activation and reduced functional connectivity of the DMN in long-term practitioners.
And the protocol we are discussing also has to do with DMN. I'm curious about your experience with DMN-training.
I agree Pete was a little harsh. I think Sebern knows what she's talking about
I recommend to you the recent book "Neurofeedback. Non-invasive alternative" by Helena Bester. She wrote a chapter about trauma treatment, with many links to Sebern Fisher and Ruth Lanius, and mentioned this Ruth research too. Sebern also wrote the foreword for the book. Can send pdf to you if needed.
Want to try this alpha-down protocol in the future, I will definitely report the results.
This is very interesting topic. I am starting some Pz alpha down NFT myself with Cyton + Brainbay.
@wjcroft, if I interpret your comments correctly it seems your perceived effects effects of alpha down training were a long term decrease in alpha effects (causing reduced DMN chatter/rumination, heightened sensory perception, theoretically less chaotic networks). Does that sound right? As opposed to acute rebound effects posited by Ros.
It's interesting that Ros mentioned alpha down training may only help those with a naturally low alpha power. It is too bad there is no open source qEEG database available for me to check this for myself. I wonder if there are other alternatives?
Here is the more recent paper from the Ros / Lanius team, published online November 2020:
https://www.sciencedirect.com/science/article/pii/S2213158220303272
"A randomized, controlled trial of alpha-rhythm EEG neurofeedback in posttraumatic stress disorder: A preliminary investigation showing evidence of decreased PTSD symptoms and restored default mode and salience network connectivity using fMRI"
Abstract
Objective: The default-mode network (DMN) and salience network (SN) have been shown to display altered connectivity in posttraumatic stress disorder (PTSD). Restoring aberrant connectivity within these networks with electroencephalogram neurofeedback (EEG-NFB) has been shown previously to be associated with acute decreases in symptoms. Here, we conducted a double-blind, sham-controlled randomized trial of alpha-rhythm EEG-NFB in participants with PTSD (n = 36) over 20-weeks. Our aim was to provide mechanistic evidence underlying clinical improvements by examining changes in network connectivity via fMRI.
Methods: We randomly assigned participants with a primary diagnosis of PTSD to either the experimental group (n = 18) or sham-control group (n = 18). We collected resting-state fMRI scans pre- and post-NFB intervention, for both the experimental and sham-control PTSD groups. We further compared baseline brain connectivity measures pre-NFB to age-matched healthy controls (n = 36).
Results: With regard to the primary outcome measure of PTSD severity, we found a significant main effect of time in the absence of a group × time interaction. Nevertheless, we found significantly decreased PTSD severity scores in the experimental NFB group only, when comparing post-NFB (dz = 0.71) and 3-month follow-up scores (dz = 0.77) to baseline measures. Interestingly, we found evidence to suggest a shift towards normalization of DMN and SN connectivity post-NFB in the experimental group only. Both decreases in PTSD severity and NFB performance were correlated to DMN and SN connectivity post-NFB in the experimental group. Critically, remission rates of PTSD were significant higher in the experimental group (61.1%) as compared to the sham-control group (33.3%). Conclusion: The current study shows mechanistic evidence for therapeutic changes in DMN and SN connectivity that are known to be associated with PTSD psychopathology with no patient dropouts. This preliminary investigation merits further research to demonstrate fully the clinical efficacy of EEG-NFB as an adjunctive therapy for PTSD.
Comments
See page 1 of this thread for background and research paper references on Tomas Ros' neurofeedback protocol:
https://openbci.com/forum/index.php?p=/discussion/2647/innovative-ptsd-neurofeedback-tomas-ros-phd-alpha-inhibit-desynchronization-protocol/p1
Paul, thanks. Well, also note that this new randomized controlled trial ran for 20 sessions, probably a few sessions per week. All his previous papers were just alpha changes observable after a single session. Since your last posted (2nd session) alpha level was in the range .75 uV to 1.25 uV, that is going to be difficult training any lower than that. 1uV EEG is just too close to noise levels, regardless of who makes the equipment or software.
Another thought I had about your Muse / Myndlife 1 uV alpha @ Pz. All EEG is measured differentially. Meaning that the closer the reference electrode is to the site being measured, the lower will be the difference voltage.
Most typical neurofeedback is using the reference on the ear lobe. One or both, if linked.
Muse reference location is not completely clear to me. Is it at the forehead, or at the earpieces? (To better match traditional measured amplitudes?) Earpiece is touching basically at the upper rear portion of the ear, where the conductive rubber 'earpiece' touches the ear surface. This is considerably higher than the ear lobes. And actually over a portion of the cortex / temporal lobes. (Very close to T3, T4 sites.) Whereas ear lobes are mostly OFF the cortex.
I do know that with traditional neurofeedback, ear lobe references, that measured amplitudes at T3, T4 are MUCH lower than at other sites, for just this reason. With Muse and Pz, (if ear references) you're going to see the same type of effect.
A last observation is that Muse ear contacts are conductive rubber. In our tests at OpenBCI with various conductive rubber materials, these typically have a resistance much HIGHER than metallic or silver chloride type electrodes. Further reducing the differential voltage measured. All the other Muse electrodes are at the forehead and metallic.
HI William, I think a good person to answer this would be James Clutterbuck at Mind Monitor. He is not connected with Interaxon but has developed 3P software for Muse for years now and runs a forum at https://mind-monitor.com/ and is quite responsive. He answered a bunch of questions from me when I was trying to get Muse 1 connected to MAX. From what I recall, Muse has a ground and a DRL at the ears however at least in 3P software such as Myndlift the reference in a unipolar placement can be chosen. So for example in the current case I choose Pz as the active channel and then can choose from Tp9, AF7, FPz, AF8, Tp10 as reference. So at least these could be compared and differences noted - however this is not my current focus. I also think Myndlift support might also be quite responsive to a question from a non-subscriber: Try [email protected]. If not, I'm happy to forward a question from you.
As to the larger question about the reliability of any given combo of NFB hardware/software - from my point of view it's somewhat unanswerable. I don't see much independent benchmarking of these systems, and so even with a 10K clinical system for someone like me it would largely be a matter of trust. Based on my experience in the software industry prior to becoming a psychologist, I would say that the big clinical system providers are not trying too hard b/c they have a captive market and often cozy relationships b/w the big clinical names and the equipment producers, so I'm not really convinced those systems are necessarily better. I think it all needs a shake-up from open-source projects like this one, and the consumer market, and maybe things will improve. My strategy is to figure out if there is anything to it clinically - i.e. symptom reduction as promised. Obviously that's predicated on the equipment working reasonably well, and it's complicated by the fact that even sham NFB sometimes "works". But it's not a perfect world - If I can get NFB to work even a little bit with the most modest equipment then maybe I'll look at a more sophisticated system.
Paul, thanks.
Other than Myndlift / Muse, all other commercial / clinical neurofeedback equipment and home training neurofeedback equipment (such as Pocket Neurobics, used with www.brain-trainer.com) that I am aware of -- use ear lobe as default reference. This is also true of the numerous QEEG normative database systems, such as Neuroguide, BrainDX, etc. This was the case when I used Dr. Richard Soutar's NewMind QEEG system, with the Brainmaster Discovery (19 channel) amplifier; with Electro-cap elastic head cap.
So there is some comparability between these systems, if the same reference is used. Research BCI systems are more variable and are not as standardized on ear lobe references. Some of these use references on the scalp, then use mathematical electrode 're-referencing' algorithms, to (for example) remove distortions caused by having a single reference. What are called "reference-free" systems.
https://sapienlabs.org/common-average-vs-infinity-reference-in-eeg/
Regarding the possible electrodes in Muse that can act as reference, apparently the three central electrodes on the forehead are all part of the 'default' reference system. Leaving these four locations,
So indeed, TP9 and TP10 are the rubber ear contacts. And the rubber is a long strip, with the main contact at the upper part of the ear. That's cool that the various Muse apps can choose the mathematically derived reference location. As long as users / developers can agree on what are the 'standard' reference locations across various applications. Especially those that attempt to replicate QEEG capabilities using 4 + (1 aux) channels.
William
Getting back to the Tomas Ros alpha down at Pz protocol, I'm continuing to see forward motion with it on my trend graphs. One thing I see is increasing lengths of time where the microvolts stay relatively low. Still have bursts going on, but the low time spans are increasing from session to session. Also seeing less noise overall in the theta and delta bands underneath alpha.
There seems to be a meshing of the neurofeedback protocol, and the daily qigong and mindfulness meditation practice that I do. My hunch is that "less noise in the system" (primarily DMN), frees up more activity to happen in domains that increase life-force and positive interconnections. (As opposed to long standing wiring related to past PTSD experiences that may have reduced chi flow.)
William
Hi! Would be nice if you can share the OpenVibe script of your Alpha protocol (html i guess).
- Best, Pascal.
Ahh it is from Brain bay....sorry
Hi guys!
How are you doing with this protocol?
Sure you already read the paper by Nicolson, Ros, and Lanius:
https://www.sciencedirect.com/science/article/pii/S2213158220303272
"A randomized, controlled trial of alpha-rhythm EEG neurofeedback in posttraumatic stress disorder: A preliminary investigation showing evidence of decreased PTSD symptoms and restored default mode and salience network connectivity using fMRI"
On the Brain-Trainer forum, a discussion started yesterday:
https://groups.io/g/brain-trainer/topic/new_protocol/80945047
@mikhaill, thanks. I did probably 10 sessions (over a couple weeks) in the summer of 2020. But did not follow through. I 'thought' I noticed some shifts, but it was subtle. I had some significant DTD (Developmental Trauma) in early childhood. That resulted in strong introversion and other characteristics. Although I feel I've left that shell, there are still subtle remnants that are hard to shake. I'll give this another round of training.
Another factor in any alpha training (up or down), is the placement of the alpha band. While traditionally thought of as 8 to 12 Hz, this can also be highly individualized. And those who have done large amounts of meditation (as I have), frequently have a lower band center point. Determining the accurate per client center frequency, may offer stronger results.
Thanks for posting the final RTC from Lanius, Ros, et al. I did exchange several emails with Tomas last year, when he said the paper was forthcoming. The link to Pete van Deusen's Brain-Trainer forum, also grateful. I think Pete's comments are a bit harsh. Lanius, Ros, and Sebern Fisher (book: "Neurofeedback in the Treatment of Developmental Trauma"), have been working in this domain of DTD, PTSD for decades. So I'm inclined to listen to Sebern when she states this is groundbreaking.
William
@wjcroft, thank you for reminding me about individual alpha. As I understand it can be found with a test that includes eyes-open and eyes-closed EEG recordings.
Have DTD too, but I'm only at start of the way, a total of 40 sessions. For me, DMN training (theta, alpha, and gamma synchronization at AFz and Pz) turned out to be very useful — for significant reduction dissociation, false-ego weakening and strengthening more real self. It's interesting, that according to Wikipedia, DMN has changed in long-term meditators:
And the protocol we are discussing also has to do with DMN. I'm curious about your experience with DMN-training.
I agree Pete was a little harsh. I think Sebern knows what she's talking about
I recommend to you the recent book "Neurofeedback. Non-invasive alternative" by Helena Bester. She wrote a chapter about trauma treatment, with many links to Sebern Fisher and Ruth Lanius, and mentioned this Ruth research too. Sebern also wrote the foreword for the book. Can send pdf to you if needed.
Want to try this alpha-down protocol in the future, I will definitely report the results.
This is very interesting topic. I am starting some Pz alpha down NFT myself with Cyton + Brainbay.
@wjcroft, if I interpret your comments correctly it seems your perceived effects effects of alpha down training were a long term decrease in alpha effects (causing reduced DMN chatter/rumination, heightened sensory perception, theoretically less chaotic networks). Does that sound right? As opposed to acute rebound effects posited by Ros.
It's interesting that Ros mentioned alpha down training may only help those with a naturally low alpha power. It is too bad there is no open source qEEG database available for me to check this for myself. I wonder if there are other alternatives?
Here is the more recent paper from the Ros / Lanius team, published online November 2020:
https://www.sciencedirect.com/science/article/pii/S2213158220303272
"A randomized, controlled trial of alpha-rhythm EEG neurofeedback in posttraumatic stress disorder: A preliminary investigation showing evidence of decreased PTSD symptoms and restored default mode and salience network connectivity using fMRI"