ADNF and NeurOptimal : my opinion and some research

For many weeks now, I've been chatting almost daily with Jean-Loup Drouet, webmaster of biofeedback.fr and neurofeedback-information.fr. I discovered the biofeedback.fr website many years ago (it opened in 2007!), when I was interested in building a homemade EEG helmet. The very relevant information of Jean-Loup (whom I did not know at that time), quickly dissuaded me in view of the complexity of the thing and the technology not yet developed (in particular dry electrodes).

So it is with pleasure that I exchange with him now, especially since he has a lot of relevant resources and reviews (which helped me greatly in writing this article, thank you), on NeurOptimal. For example, on this page he compares the technical specifications of NeurOptimal with those of a neurofeedback system.

More specifically, in his comments, he notes (I fack-checked a NeurOptimal site here) :

In addition, each user is encouraged to create a website and then encouraged to create a link to a directory in order to influence Google (see ADNF website). So you have every chance of arriving on "Dynamic Neurofeedback" by looking for information on Neurofeedback. The abundance of these sites has the effect of making this method virtually particularly interesting and thus certainly very effective...

• This association aims to spread neurofeedback in France but this object does not exclude interventions abroad. (statutes)
• The association provides its members with neurofeedback equipment at the association's headquarters or at home. (statutes)
• Neurofeedback is not a medical treatment but rather a healthy brain gymnastics. (IR)
• The information provided by the ADNF concerning the benefits of neurofeedback is the most faithful possible reflection of the english literature on the subject but is by no means a guarantee of results. (IR)

The statutes of an Association define the legal framework of action of the Association, they protect it and must be transmitted to the Prefecture. It is quite rare to change them, because it means that the Association would take a different course than that initially defined by the founders. But they can be updated following a vote in an Extraordinary General Assembly where the members are convened specifically for this purpose. This is often to update small things (head office address, website address, etc.....). In my sports club, for example, we updated our statutes dating from 1988 (!) in 2016, to integrate new technologies: e-mail address, website, domain names....

The Internal Regulations, on the other hand, aim to specify the daily functioning of the Association and to specify the criteria for the application of the statutes. It can be changed quite easily at the Annual General Meeting. It can therefore evolve with the Association and remains a good reflection of what the Association does.

Several things already bother me in these statements.

Spreading neurofeedback in France

Already, the Association wants to disseminate Neurofeedback in France, but its site explicitly states : ADNF uses the professional neurofeedback equipment NeurOptimal™ from the Zengar Institute (The President of the ADNF explained to me that initially, ADNF had not chosen NeurOptimal, the solution was imposed for reasons of cost / simplicity over time).

So they broadcast neurofeedback, but only using NeurOptimal (Jean-Loup explains on this page why NeurOptimal is not neurofeedback and I use the same arguments in my previous article).

I naively asked to be explained how NeurOptimal was Neurofeedback, and I got the following answer:

There are several kinds of neurofeedback and NeurOptimal is one of them. NeurOptimal sends the person back information about his brain function, but the resulting learning is done unconsciously instead of by operant conditioning. (...)

The definition of neurofeedback may specify that the resulting learning must be done by operant conditioning but it is unfortunate if it excludes other forms of learning.

What bothers me about this statement is that neurofeedback is a particular branch of biofeedback. And biofeedback, in its very definition, is based on operant conditioning. So already, big problem. It would be like saying that there are different types of archery and that pistol shooting would be one variant among others. This would probably be correct if the pistol in question fired arrows rather than bullets (for the anecdote, crossbow shooting depends on the Federation of Shooting (pistol / rifle), because the posture / speed of the projectile is closer to shooting than archery).... If NeurOptimal is not based on operant conditioning, calling this technique "neurofeedback" is false. After, yes, "it is unfortunate" that the definition of Neurofeedback doesn't fit with the functioning of NeurOptimal, but well, science guys.

I also contacted BCIA, the international certification in Biofeedback, with a very simple question: Is neurofeedback based on operant conditioning? Can a solution that is not based on operant conditioning be called neurofeedback? and I got the following answer:

Yes – neurofeedback is based on operant conditioning.

The BCIA does not like to talk about material, they prefer to talk about neurofeedback protocols, but the answer has the merit of being clear....

(In detail, neurofeedback also relies on classical conditioning, and all underlying mechanisms of neuronal plasticity, but the major ingredient remains the operant conditioning)

Neurofeedback is healthy brain gymnastics

Second, they specify that neurofeedback is not medical treatment.

When we know that Neurofeedback has demonstrated the highest level of efficacy and specificity in the treatment of ADHD.... well, personally, I find that neurofeedback is indeed a medical treatment (if practiced by a person with a medical degree, eh).

I specify the diploma in medicine, that one does not reproach me the illegal practice in medicine, but Louis Mayaud of Mensia (which develops a system of automatic Neurofeedback for the ADHD, scientifically robust and medically recognized), specifies on this subject :

"I don't think that's strictly necessary. There are many technical and psychological reasons why this [a qualified medical practitioner] has been indispensable so far and why it may also have a benefit today. It must also be taken into account that this presence also presents the most important limitations of which are the inter-individual variability and therefore the fact that effectiveness is a function of the practitioner."

NeurOptimal, on the other hand, is definitely not, Zengar, the company that markets the product writing: "The NeurOptimal is not a medical method, device or treatment. This system has not been validated for any medical application by the FDA or any other government agency." (source on biofeedback.fr).

The information provided by the ADNF reflects English scientific literature

Finally, the ADNF certifies that the information it provides is as faithful a reflection as possible of English literature. By going to the ADNF website, under the heading "Some scientific studies and other analyses on NeurOptimal dynamic neurofeedback", the first link led to a powerpoint presentation by a high school student who was testing NeurOptimal. So it was certainly English, but scientific literature level, it was very debatable (since removed from their site -and appearance of the mention "other analyses" in the title to include their powerpoint-, following my remarks).

I would add that currently, most of the resources(*) provided as "English literature" (and following my comments, as "other analyses") by ADNF are only PowerPoint presentations / Blog article. It has therefore not been published in an international peer-reviewed journal and cannot therefore be considered as "literature". In science, literature refers to published articles that have been peer reviewed. At best, the term that can be used is "exploratory study" or "preliminary results that need to be completed".

(*) In detail, the ADNF site gives 11 resources:

• 4 are PowerPoint presentations, so not scientific literature ;
• 3 are blog articles / a thesis that has not been published, so idem ;
• 2 abstracts of scientific articles, impossible to find on the journal's website, I wonder if the authors had their article retracted / withdrawn afterwards ;
• 2 scientific papers published in bad reputable journals, whose authors admit that more data would be needed. The hardware is based on the old model (with the ProComp and not the zAmp that we don't know anything about), so it doesn't seem to be up to date anyway.

In order not to leave anything to chance, I also looked at the "Research" resources given on the site of Zengar, manufacturer of the NeurOptimal. There are 25 of them:

• 15 are PowerPoint presentations, so not scientific literature ;
• 3 are the same blog / thesis articles as on ADNF website, so idem ;
• 4 are abstracts whose article is not found / conference abstracts ;
• 3 are real scientific articles, which deal with Neurofeedback, without ever mentioning Zengar's solution.

So here too, it is quite poor scientific literature.

ADNF does not rely on the scientific method

As a matter of measurement, not everything can be proven by scientific study, even if that would be ideal. We often have to settle for less for practical reasons. (...)

Yes, the fact that it is scientifically proven is not our first selection criterion because there are good products that do not meet this criterion and that we would not like to deprive users of.

Let us be clear: not being able to prove the existence of God and therefore appeal to faith is correct. We can't prove God exists. On the other hand, to prove the effectiveness of a material (NeurOptimal or Neurofeedback) associated with a good training protocol, this is done. So there is no reason to settle for less. I reminded ADNF that it was personal ethics to question the effectiveness of NeurOptimal when you are an association that promotes it. Obviously, this would be "outside the scope of ADNF", because it considers that it is often necessary to be satisfied with less than science. You've been warned: ADNF's selection criterion is not clinical efficacy, it's user testimonials (and a testimonial is far from proof, because of the placebo effect).

I touched briefly on the case of Mensia, which found 3.6 million euros from Europe to conduct a clinical study on their Koala solution. ADNF explains that, for them, it is easy to do science. But why has Zengar never submitted or obtained a project to validate its solution? According to Societe.com, Mensia was founded in 2012, which is ten years later than Zengar.

NeurOptimal: a mystery machine with many virtues

Then I asked technical questions about NeurOptimal: ADNF only promotes this solution, so I thought to myself, that they had explored the subject.

Here again, the ADNF is kicking the ball: "Indeed, Zengar does not communicate on his technical implementation of the zAmp", "Val Brown does not communicate precisely his algorithm for calculating the occurrence of turbulence(*)".

So, like just about everyone else, ADNF ignores how NeurOptimal works. And here, honestly, it's really a question I ask myself: why doesn't Zengar make a patent to protect their technology and disseminate it? To my knowledge, this is the classic case. At my faculty, as a researcher who worked on gene therapy, I regularly received emails from our legal department so that we could be followed to file patents.

(*) In 2018, no researcher was ever able to observe these "turbulences".

That is my overall criticism of ADNF. I don't talk about their website, I could spend my life there, so much there is confusion between NeurOptimal, Neurofeedback, what science shows.... At the limit, my previous article takes well the communication bias of all these NeurOptimal sites.

Oh yes, one thing on the ADNF website. One cannot blame them for not communicating around Neurofeedback, if only in view of the long page of links, which makes available to the user, quality links (unfortunately drowned in other links much more debatable). And I'm not saying that, because there are links to Jean-Loup's sites or mine (although, it's always nice the flattery, huh).

Jean-Loup and I have a special treatment, with a link to this footnote :

It is important to specify it: Neither Jean-Loup, nor I, commercialize material of Neurofeedback.

We will never recommend specific equipment (we do not use the same equipment), because in Neurofeedback, it is not the equipment that is essential: it is the quality of the EEG signal and the protocol used (NeurOptimal does not offer any tools to verify the EEG signal and its protocol is "trade secret"). Similarly, on the economic level, we have a local activity (we can't rent equipment that will be used at the other end of France like NeurOptimal), so to tell you clearly: these articles are a volunteer activity and 95% of people who contact me to do Neurofeedback are referred to other (quality) practitioners because this is not my geographical area (and I don't take any commission, eh). So I can't value a "product": there isn't any. What I value is neurofeedback. And besides, I don't denigrate NeurOptimal, I try to inform you about the opacity of this solution / their business model.

And once again, as a therapist, in the end, what is important is to base yourself on scientific facts, validated, verified, in order to provide you with the best therapeutic management.

NeurOptimal is open to science

Afterwards, the President of the ADNF is well aware that his Association is promoting a solution that is not scientifically proven. And even if I see well in our exchanges that he does not really understand what the scientific method is, he puts forward a certain will of Zengar to go towards more scientific proofs :

I rely on 3 facts to say that the Zengar Institute is open to scientific evidence:

1) Zengar hired a research director to accompany the ongoing studies.

2) Zengar makes its equipment available free of charge to researchers wishing to carry out a scientific study.

3) Zengar has integrated in the new version the possibility of triple blind sessions.

Concerning ADNF, I confirm that we are open to scientific evidence for NeurOptimal and that we regret that there is no more. The testimonies of practitioners and their clients are used for want of better.

Concerning NeurOptimal practitioners, it seems quite obvious that if there were a good hundred scientific studies on NeurOptimal, it would be an excellent marketing argument. Why would they be closed to that?

You can say that the effectiveness of NeurOptimal is not scientifically proven and that, failing that, we are satisfied with testimonials. That sounds much more accurate to me than your formulation.

At this point, I ask myself a question: if Zengar makes everything available for researchers to validate NeurOptimal, how can we explain that no serious study demonstrates its effectiveness after 17 years?

From my personal opinion, which belongs totally to me and only involves me, I see the problem from several angles:

• Already, a scientist knows that NeurOptimal is not Neurofeedback;
• No scientist could observe these "turbulences" (always prove the existence of a phenomenon before studying it, I speak about it here) ;
• Everyone doesn't know how NeurOptimal works (and in science, we are for the transparency of our protocols so that others can replicate and verify them);
• In science, there is a publication bias called "negative": it is much more difficult to publish if the result is negative. And as a scientist, our CV is our publications. Refused papers that show that NeurOptimal does not work, it is possible, we would just have no trace of it. And embarking on a project like this (with no information on the material) is risky, because again, publishing is vital for a scientist.
• When NeurOptimal was released (early 2000), it was much more difficult to check the information (I coded my first site in 2003 with a 56k connection and I was 15 years old to tell you....), and Neuroscience was much less fashionable. In a sense, it reminds me of homeopathy (with which I bathe you, I know): at the time homeopathy was created, it was an alternative to medicine (with bleeding and company doing more harm than good): not doing harm was already better. I advise you this magnificent video which explains this (you will learn in particular that you should not consume coffee/tea, do not ride a horse, do not play, do not masturbate, do not read pornographic texts... otherwise homeopathy does not work. Another time, I tell you).

So I just think that if NeurOptimal can't get the scientists on board, maybe it's because the researchers who work in the field know it can't work the way Zengar wants it to.

For example, James Randi (a professional illusionist who is more particularly known as a demystifier of pseudo-science and other paranormal phenomena) offers through his foundation a million dollars to any medium who would prove according to a scientific protocol this "power". Strangely, no one took up the challenge.... So it's not that scientists aren't interested in a subject, it's sometimes just that people talk bullshit....

I have also received a preliminary report showing that NeurOptimal improves brain function in stuffed animals. The authors of this preliminary report are certified by the BCIA. So they know the scientific method, Neurofeedback and what they do. For all practical purposes, I would remind you that this is a preliminary report, not a scientific study published in a journal (the number is too small). I found the author (Rubén Pérez-Elvira, from Nepsa Rehabilitation Center) on LinkedIn, then we chatted on Facebook. He confirms to me that indeed, he observes with his preliminary results that the NeurOptimal manages to improve the cerebral functioning of stuffed animals. We laughed about it among ourselves, but obviously many people asked for his nomination for the Ig-Nobel Prize, which rewards scientific research which seems unusual but which leads to secondary reflection. The Ig-Nobel committee asked him to complete his study and publish it in an international journal before accepting his appointment.

He also told me that he managed to improve brain function, thanks to the NeurOptimal, from a.... piece of wood.

Rubén Pérez-Elvira has a much drier opinion than me on the NeurOptimal : for him, NeurOptimal records nothing at all and uses locally stored data and cooks especially with regard to the age of the patient (he tested stuffed animals with different ages included in the software and obtained different results). It also highlights the pyramid system business that I mentioned in my last article.

And he made a qEEG before / after NeurOptimal: no cerebral change was perceptible. Once again, this is not scientific evidence, but NeurOptimal relying on testimonies, it seems honest to me to disseminate contradictory testimonies to theirs. Science clarifies in general, all this.....

It was a great meeting and we'll check if the signal picked up by NeurOptimal is from the EEG (a conference at ISNR, two years ago, supposedly showed the opposite, I contacted them to get the documents, request refused because I'm not a member... here, here... Ruben, through Thomas (Director of the Neurofeedback-EEG Institute Munich, -an accredited training institute BCIA-), member of the ISNR, will apply. Thanks to both of them!). This will bring a little more water to the mill and I will update the article.

I also contacted Louis Mayaud, VP Research at Mensia who works on Koala, an automatic medical neurofeedback system for ADHD patients. The question was simple: "What if we could put a Mensia Koala on the head of a stuffed animal? We'd get thrown out in the first place because of the impedance? The system would then throw us away because it would realize that the system does not detect EEGs?".

Impedance should not pass in the absence of intracellular fluid;

If it did, the first session is performed by a specialist who should not consider that it is an EEG;

If he considered it EEG by mistake, then our back end diagnostic tools would report an error within the next day to the practitioner;

If the calibration was done on a real animal then subsequent sessions would detect that the signal does not "resemble" the assessment data (editor's note: assessment = first EEG evaluation carried out on the patient).

So in an automatic system, which has benefited from real scientific research: the system would not be fooled by a stuffed animal.

At a given moment, you can be friendly with people, but at a different moment, you just have to realize that :

• ADNF promotes NeurOptimal as neurofeedback, which is false;
• ADNF completely ignores neurofeedback equipment used in scientific research, in favour of a proprietary solution of which it knows nothing;
• ADNF advances a lot without proof, so be critical of what it advances and ask for proof!

Now it's up to you to make up your mind in the light of all these elements.