Thread: Published in Nature Neuroscience: 77% induction rate with electrodes on the scalp at 40 Hertz

Originally Posted by Zthread

I also participated in that Mortal Mist study. I tried it a lot, with various frequencies, including the recommended one. Unfortunately, it didn't do much of anything for me, at least for inducing LDs. I guess nothing works for everyone. Or I might have been doing something wrong.

How often did you get lucid with CES? Was it anywhere close to the 77% success rate reported in the Nature paper? Also, are you someone who gets lucid fairly easily?

I don't have the details about exactly how often I got lucid, but I can say that I am a natural lucid dreamer and typically have 1-3 lucids a week without any induction methods. I definitely go through dry spells too where I won't LD at all for a month or so, but then they randomly come back for whatever reason. The CES for me definitely qualifies as an induction method so it's only natural that my LD frequency would increase a bit just from experimenting with the device and being in the mind set to get lucid, regardless of the device's actual effectiveness. The correlations I was able to make were more indirect...in that I was lucid in general more often during the times when I was experimenting with the device, rather than becoming lucid directly as a result of having the device on during a REM period if that makes any sense.

Originally Posted by nina

I don't have the details about exactly how often I got lucid, but I can say that I am a natural lucid dreamer and typically have 1-3 lucids a week without any induction methods. I definitely go through dry spells too where I won't LD at all for a month or so, but then they randomly come back for whatever reason. The CES for me definitely qualifies as an induction method so it's only natural that my LD frequency would increase a bit just from experimenting with the device and being in the mind set to get lucid, regardless of the device's actual effectiveness. The correlations I was able to make were more indirect...in that I was lucid in general more often during the times when I was experimenting with the device, rather than becoming lucid directly as a result of having the device on during a REM period if that makes any sense.

So it may have been at least partially the placebo effect, in your case. Must be nice to be a natural lucid dreamer!

Great - thanks Zoth!
Will work myself through it and see, what it is all about with the 77% and lucidity indicators vs. actual lucid dreams.
Not instantly obvious from a glance over it.

Originally Posted by StephL

Great - thanks Zoth!
Will work myself through it and see, what it is all about with the 77% and lucidity indicators vs. actual lucid dreams.
Not instantly obvious from a glance over it.

Considering Voss was one of the authors, you can probably get some more insight by reviewing her "Lucidity Scale". I won't be surprised if it keeps showing in future studies of this nature. Here's the article.

Thanks again - I was just about to edit in, that I found nothing else whatsoever on the measuring method in the paper than an attached reference number, which is really not in order for something absolutely central and crucial, actually.
And strangely - the references in your link start with number 20.
Now I'll check the other paper.
Did you find out how many actual lucid moments were triggered by this? Per how many test runs?
I very much doubt, it was 77% and I very much dislike obscurantism in scientific papers.
I read quite a lot of them things in original - and I got to say - the OP's one is very hard to digest - how can it be, that they don't make it clear, what their actual results were?
On the other hand - it made it into Nature ..?

Well - got to read them both completely.
What they did in that second paper, seems to be comparing reports of lucid and non-lucid dreams for in the end 8 criteria, two of which wouldn't be really useful, since not significantly different, namely realism and negative emotions. Below they forget to say that memory is also significantly better, but they do in the text. Hence - 6 factors:

The goal of this study was to identify the phenomenological correlates of primary and secondary consciousness in dreams
and to introduce the LuCiD scale as a reliable means for the measurement of lucidity and consciousness in dreams. According
to our findings, dream consciousness can best be described through the factors (1) INSIGHT, (2) CONTROL, (3) THOUGHT, (4)
REALISM, (5) MEMORY, (6) DISSOCIATION, (7) NEGATIVE EMOTION, and (8) POSITIVE EMOTION. Whereas normal REM sleep
dreams lack those factors requiring secondary consciousness, lucid dreams are defined through those, namely insight,
thought, control, and dissociation. In addition, lucid dreaming seems to be accompanied by positive emotion, suggesting that secondary consciousness adds cognitive functions and positive emotionality onto primary consciousness.

They found out, that in these 6 factors on a scale from 0-5, lucid dream report reap significantly higher values.
I guess, it's really like the WIRED article suspects - they woke people up and inquired about the (only 3) criteria, and analysed the values.
If they were significantly higher under stimulation, than without, where LDs have higher values - then that constituted a degree of lucidity.
And if the absolute mean values were as low as WIRED cited (can't easily find them) - that's pretty disappointing for all practical purposes let alone justifying diy experiments.
Proof of principle only.

I hope, I'm wrong..
Correct me, if so - I'm going to have to read both properly, before coming to a serious conclusion.
And I'm so disappointed, I am too lazy.
Don't get me wrong - if it's "just that" - it's fascinating anyway - but the media hype around it takes the fun out of it for me.

And if the absolute mean values were as low as WIRED cited (can't easily find them) - that's pretty disappointing for all practical purposes let alone justifying diy experiments.

The numbers mentioned by Wired are shown graphically in Figure 3 of the Nature paper.

I hope, I'm wrong..
Correct me, if so - I'm going to have to read both properly, before coming to a serious conclusion.

Might be worth a try to email a question or two to the principal author. Her email address is provided in the paper at the bottom of page 1. She might just mark such an email as spam. But maybe not.

Thank you Zthread!
I might just do that!
If I will - of course I will report!

Well well - here a nice graphic, which the authors did not provide so concisely - it stems from converting diagrams of theirs, so the data can be interpreted easily - from here: The Neurocritic: Does Gamma tACS Really Induce Lucid Dreaming?
Cheers, Zoth!



Mean scores for LuCiD scales for non-lucid vs. lucid dream reports - NOTE: each scale goes from 0: strongly disagree to 5: strongly agree.
The yellow bars indicate means after 25 or 40 Hz tACS.


Blue is normal dreams - red is real lucids - and yellow is where the people ended up after stimulation.
The mean insight level reached with stimulation is below 1!! So they very much disagree with the notion that they were aware of the fact, they were dreaming.
While they did find something, I am aware of that - the LDing community is again confronted with a huge disappointment in my eyes, which could have been avoided by less sensationalist presentation - especially in the secondary sources, of course.

Originally Posted by StephL

The mean insight level reached with stimulation is below 1!! So they very much disagree with the notion that they were aware of the fact, they were dreaming. While they did find something, I am aware of that - the LDing community is again confronted with a huge disappointment in my eyes, which could have been avoided by less sensationalist presentation - especially in the secondary sources, of course.

On the positive side, though, it's still pretty encouraging that they were able to induce statistically significant increases in proven indicators of lucidity in subjects who were not LD enthusiasts, using weak electrical stimulation (0.25 mA peak-to-peak) for a duration of only 30 seconds. In addition, at least one of the participants had an actual LD:

Example of lucid dream report following 40-Hz stimulation. I was dreaming about lemon cake. It looked translucent, but then again, it didn’t. It was a bit like in an animated movie, like the Simpsons. And then I started falling and the scenery changed and I was talking to Matthias Schweighöfer (a German actor) and 2 foreign exchange students. And I was wondering about the actor and they told me "yes, you met him before," so then I realized "oops, you are dreaming." I mean, while I was dreaming! So strange!

So it's possible that by optimizing the tACS amplitude, frequency, duration, and time of application during the sleep cycle, a high probability of inducing actual LDs could be achieved.

I just sent the following email to the principal author of the paper:

Dear Dr. Voss,

As someone who is interested in lucid dreaming, I very much enjoyed your recent Nature paper on induction of self awareness in dreams using electrical stimulation of the scalp. Your finding that the LuCiD factors of insight, dissociation and control can be increased through tACS scalp stimulation is extremely interesting and potentially valuable in clinical applications. However, it wasn't completely clear to me what percentage of the dream reports following 25 Hz and 40 Hz stimulation indicated that the subject actually had a lucid dream. Your paper states that: "Regarding subjective ratings of lucidity, lucid dreams were most prominent during stimulation with 25 (58%) and 40 Hz (77%) suggesting that the rate and/or periodicity of oscillatory activity in the brain is causally relevant for higher cognitive functioning[13,14] and that lower gamma-band activity may indeed be a necessary condition for the elicitation of secondary consciousness in dreams, perhaps even in waking." Does this mean that in the case of the 40 Hz stimulation, for example, 77% of the subjects actually had lucid dreams? Or does it mean that 77% of those subjects had statistically significant increases in LuCiD factors, but that they didn't all have what would normally be considered to be a lucid dream? If the latter is the case, what percentage of the subjects had dreams that would normally be considered to be lucid, for both 25 Hz and 77 Hz stimulation? Thank you for doing this valuable work. I hope you have continued success in your future research projects.

Best regards,

[Zthread]

Fantastic Zthread - thanks - we should really do that - take up direct contact to scientists - also wonderfully put!
Looking forward to find out, how they will react - and of course to finding out the number of actual lucid dreams having been elicited.
The lemon cake citation must mean - they elicited at least one. How many, though?

Originally Posted by StephL

Looking forward to find out, how they will react - and of course to finding out the number of actual lucid dreams having been elicited.

She might just delete it without replying. We'll see.

Originally Posted by StephL

The lemon cake citation must mean - they elicited at least one. How many, though?

Right. Definitely one. Whether or not there were more depends on what they really meant by this statement:

"Regarding subjective ratings of lucidity, lucid dreams were most prominent during stimulation with 25 (58%) and 40 Hz (77%) suggesting that..."

One of my users is trying to reproduce the findings. You can see the effect of the tACS on the raw EEG reading:



And this is what he has found so far:

Originally Posted by Jef

I am beginning to realize that zapping the brain synchronizes it to the desired stimulated frequency (alpha, beta , delta etc.) much more powerfully than any sound and lights.

Originally Posted by IAmCoder

One of my users is trying to reproduce the findings. You can see the effect of the tACS on the raw EEG reading:

And this is what he has found so far:

I am beginning to realize that zapping the brain synchronizes it to the desired stimulated frequency (alpha, beta , delta etc.) much more powerfully than any sound and lights.

Not sure I understand this. The horizontal scale must be seconds. What is the vertical scale? Is the gray trace the tACS stimulation? And the blue and green curves show that the brain state was changed and REM was induced? Is the NeuroSky device capable of doing the tACS? Doesn't look like it, based on their website. So was another device used to produce the tACS stimulation?

Yes, the horizontal scale is in seconds, so the printout shows one minute. The vertical scale shows the raw EEG signal from the NeuroSky device. You can read more about it here: Brain Wave Signal (EEG) of NeuroSky and use this formula to convert the raw values to voltages: (rawValue * (1.8/4096)) / 2000. This is due to a 2000x gain, 4096 value range, and 1.8V input voltage.

The blue channel is just a slowed down and averaged out version of the raw data. The raw channel is sampled at 512 Hz, but the blue channel only runs at 10 Hz. The green channel is a count of eye movements, which is normally used to trigger audio tracks and now tACS devices. The subject was awake during this test (at 18:39) and manually turned on the tACS device - the block of gray shows the effect of the tACS on the raw EEG. But the latest version now does that automatically after counting six consecutive eye movements.

And yes, the NeuroSky device is not doing the tACS. This device was used: g.Estim - g.tec's programmable electrical current stimulator - it is now connected to an Arduino board which forwards the signal from the app after it detects the eye movements, but that can also be done with USB relays. This one should work too: JACE Systems Electrotherapy TriStim. And this is the one that was used in the study: neuroConn - DC-STIMULATOR PLUS for tDCS and tACS.

Originally Posted by IAmCoder

Yes, the horizontal scale is in seconds, so the printout shows one minute. The vertical scale shows the raw EEG signal from the NeuroSky device. You can read more about it here: Brain Wave Signal (EEG) of NeuroSky and use this formula to convert the raw values to voltages: (rawValue * (1.8/4096)) / 2000. This is due to a 2000x gain, 4096 value range, and 1.8V input voltage.

The blue channel is just a slowed down and averaged out version of the raw data. The raw channel is sampled at 512 Hz, but the blue channel only runs at 10 Hz. The green channel is a count of eye movements, which is normally used to trigger audio tracks and now tACS devices. The subject was awake during this test (at 18:39) and manually turned on the tACS device - the block of gray shows the effect of the tACS on the raw EEG. But the latest version now does that automatically after counting six consecutive eye movements.

And yes, the NeuroSky device is not doing the tACS. This device was used: g.Estim - g.tec's programmable electrical current stimulator - it is now connected to an Arduino board which forwards the signal from the app after it detects the eye movements, but that can also be done with USB relays. This one should work too: JACE Systems Electrotherapy TriStim. And this is the one that was used in the study: neuroConn - DC-STIMULATOR PLUS for tDCS and tACS.

So the main significance of the chart you posted is to show that the tACS has an effect on the brain state? How do we know that it's not just the tACS current that's being directly detected, rather than brain state? It looks like the brain state goes right back to normal after the tACS is shut off. Or am I misunderstanding it?

Thanks for providing the links to the tACS devices. How difficult is it for someone who's not a medical professional to buy one? And how expensive are they?

It is the tACS current that is being directly detected. And the brain state does go back to normal, so all we have to go on for now is the subjective statement. It would be interesting to see a longer recording with the power bands turned on.

Not sure what they cost or how to get one as a mortal, will see if I can find out.

The following blog from July 18, 2012 mentions prices for a couple of tDCS devices. Prices for tACS devices are probably similar.

Device | DIY tDCS - Part 2

There are presently only two stand-alone devices that produce tDCS. They are: the Eldith DC Stimulator by Neuro Conn, of Germany, which sells for €3000 (about $4,000US) and the CESta, by Mind Alive Inc., of Canada, which sells for $350US. [now $450. + accessories]

Wouldn't buy one for $4000, but maybe for $450.

Hm - so it isn't even sure, that the brain state adapted to the current, because the current directly would overshadow it anyway?
Are we sure, that they took care of that problem in the original paper?
No e-mail yet, Zthread?

In the original study they applied a notch band filter at 40Hz to take care of the problemstudy.jpg

This is the setup I used from the lucidcode website

setup.jpg

Thank you jeff!
Making your pictures big:

This setup seems to work only in the early morning between 4-5am by entrainiment of 40 Hz gamma waves. This makes the frontal and occipital regions artificially active during the REM sleep stage and viola. Part of the brain is asleep and part is wide awake

LucidREM.jpg

Elite. Was wondering if we couldn't also filter out the 40 Hz at some point.

When you play the 40 Hz while awake, do you feel more aware?

Can you record a minute while awake and relaxed with the EEG power bands on, followed by the tACS and then another minute directly after the tACS. Was wondering if any of the channels change afterwards...

Originally Posted by IAmCoder

Elite. Was wondering if we couldn't also filter out the 40 Hz at some point.

When you play the 40 Hz while awake, do you feel more aware?

Can you record a minute while awake and relaxed with the EEG power bands on, followed by the tACS and then another minute directly after the tACS. Was wondering if any of the channels change afterwards...

The setup works well now using the Zeoscope , modified to trigger the arduino Lucid.jpg(tACs) play along with the music track and only arm the tACs after 5am
During lucid gamma goes up from 0.08 to 0.2 , while during the wake stage is usually 0.3
Attached is the actual capture of the lucid dream, the REM sleep before 5:am only triggers audio and after 5:am the waveform gets erratic coz of tACS interference

Was thinking about using my Zeo Bedside +Serial cable to detect REM and then have my computer trigger flashing LED's + 40Hz tACs for 30-45 seconds during the beginning of REM stage. Still waiting on some of the parts to arrive and I'm looking for a cheap 3.3v Arduino + BLE. According to the data in the paper, it seems tACs stimulation (as applied in the experiment) would not be enough to achieve what would be considered a lucid dream state by many members of this site. I believe they only applied tACs for 30 seconds or something like that. Perhaps adjusting duration or other parameters would be enough to achieve better lucidity.

How long until we see a LDing tACS kickstarter, I wonder?