More on Neurofeedback’s Brain Training Value

(Note: neu­ro­feed­back is a form of biofeed­back that mea­sures brain waves and that, accord­ing to prac­ti­tion­ers, pro­vides good “brain train­ing” for spe­cif­ic clin­i­cal conditions).
A few weeks ago Dr. David Rabin­er wrote a great post on How Strong is the Research Sup­port for Neu­ro­feed­back in Atten­tion Deficits?, con­clud­ing that

- “It is for these rea­sons that neu­ro­feed­back is under­stand­ably regard­ed as an unproven treat­ment approach for ADHD at this time by many ADHD researchers.

- How­ev­er, these stud­ies do pro­vide a sol­id basis for sug­gest­ing that if par­ents choose to pur­sue neu­ro­feed­back for their child, there is a rea­son­able chance that their child will ben­e­fit even though we can’t be sure that it is the spe­cif­ic EEG train­ing that is respon­si­ble for the ben­e­fits. Thus, although the effi­ca­cy of neu­ro­feed­back has yet to be con­clu­sive­ly con­firmed in a ran­dom­ized, place­bo-con­trolled tri­al, it is impor­tant to place this lim­i­ta­tion in the con­text of the sup­port­ive research evi­dence that has been accumulated.

- Pro­vid­ing this con­text can help fam­i­lies bet­ter under­stand the strengths and lim­i­ta­tions of the exist­ing research on neu­ro­feed­back and enable them to make a bet­ter informed deci­sion about whether to con­sid­er this treat­ment option for their child.”

This post prompt­ed sev­er­al good com­ments, one of which is repro­duced below in its entire­ty, since it adds an inter­est­ing perspective.
Bernard writes: My wife tried EEG neu­ro­feed­back over 10 years ago in the hopes of nor­mal­iz­ing her brain func­tion­ing to over­come life­long epilep­sy. She had a his­to­ry of mul­ti­ple, dai­ly absence seizures and grand mal (ton­ic clonic) seizures once every two years.

After 3 and a half months of twice week­ly ses­sions, we almost gave up on the neu­ro­feed­back. It was burn­ing a hole in our wal­let (no insur­ance cov­ered it) and we were not see­ing any results. How­ev­er, we stuck with it (most­ly because my wife refused to poi­son her liv­er with anti-epilep­tic drugs).

After 5 months, it was like some­one had turned a switch. She stopped hav­ing seizures, was calmer, had bet­ter mem­o­ry and cog­ni­tive func­tion­ing (think­ing clear­er). We stopped the neu­ro­feed­back ses­sions and she went 4 years with­out a sin­gle seizure event and like­ly would still be com­plete­ly seizure free today we had not start­ed a fam­i­ly (her TC seizure activ­i­ty returned, but not the absence seizures, and got pro­gres­sive­ly worse with each preg­nan­cy — but that’s a dif­fer­ent story).

After our expe­ri­ence, I did as much dig­ging as I could about EEG neu­ro­feed­back (see ) and I’m real­ly out­raged that the med­ical indus­try con­tin­ues to “poo-poo” the resound­ing body of evi­dence for it.

Snip­pets from my findings:“Randomized dou­ble blind place­bo con­trolled clin­i­cal tri­als (RCT) are the cur­rent “gold stan­dard” for demon­strat­ing clin­i­cal effi­ca­cy of new drugs or ther­a­pies. It is very dif­fi­cult for new ther­a­peu­tic inter­ven­tions to gain broad accep­tance in the absence of such tri­als. Recent events have raised seri­ous ques­tions about the con­di­tions under which place­bo (sham) con­trols can be used. The inter­na­tion­al stan­dards pub­lished by the World Med­ical Asso­ci­a­tion (Dec­la­ra­tion of Helsin­ki) pro­hib­it place­bo-con­trolled stud­ies when known effec­tive treat­ments exist. Addi­tion­al­ly, there is new inter­est in iden­ti­fy­ing the mech­a­nisms under­ly­ing the place­bo response, which may chal­lenge the “place­bo” as a legit­i­mate con­trol con­di­tion. Both of these events should be of con­sid­er­able inter­est to those inter­est­ed in clin­i­cal psy­chophys­i­ol­o­gy in gen­er­al and neu­rother­a­py in particular. ”

Recent New Eng­land Jour­nal of Med­i­cine reviews of research design have cast doubt on the need for place­bo con­trolled designs. Their review has shown that when there is a pre­pon­der­ance of case series reports, the con­cor­dance between those results and those of the “gold stan­dard” (dou­ble blind place­bo con­trolled stud­ies) was very high. Many in the field are now argu­ing against doing a dou­ble blind study due to the lack of prop­er humane treat­ment of those in the con­trol group (receiv­ing no treat­ment), an approach which is also now con­sid­ered uneth­i­cal by the World Health Orga­ni­za­tion when known treat­ments exist.”

Since the first sin­gle-case study, report­ed over 30 years ago (Ster­man & Fri­ar, 1972), a fair num­ber of con­trolled clin­i­cal stud­ies, stem­ming from many dif­fer­ent lab­o­ra­to­ries, have pro­duced con­sis­tent data on the effi­ca­cy of SMR train­ing in epilep­tic patients. It is par­tic­u­lar­ly note­wor­thy that these results have been achieved in an extreme­ly dif­fi­cult sub­group of epilep­sy patients, those with poor­ly con­trolled seizures who had proven unre­spon­sive to phar­ma­co­log­i­cal treat­ment. We will here pro­vide only a cur­so­ry overview of this clin­i­cal research lit­er­a­ture. For a more detailed treat­ment the inter­est­ed read­er is referred to Ster­man (2000), while oth­er recent sum­maries have also been pro­vid­ed by Mon­der­er et al. (2002), and Walk­er and Kozlows­ki (2005).

In review­ing the data accu­mu­lat­ed in these stud­ies, Ster­man (2000) found that 82% of 174 par­tic­i­pat­ing patients who were oth­er­wise not con­trolled had shown sig­nif­i­cant­ly improved seizure con­trol (defined as a min­i­mum of 50% reduc­tion in seizure inci­dence), with around 5% of these cas­es report­ing a com­plete lack of seizures for up to 1 year sub­se­quent to train­ing cessation. …”

Because of the prob­lems with design­ing a gold stan­dard study, the Asso­ci­a­tion for Applied Psy­chophys­i­ol­o­gy and Biofeed­back (AAPB) has devel­oped their own rat­ing scale for mea­sur­ing effi­ca­cy of neu­ro­feed­back for a giv­en condition:

Rat­ing explanation:

Con­di­tions with ratings:

What real­ly gets my goat is that EEG neu­ro­feed­back has been stud­ied now since the 60s — almost 50 years and there have been no reports of iatro­ge­n­e­sis (a harm­ful effect pro­duced by the heal­er or the heal­ing process): “For­tu­nate­ly, adverse reac­tions to biofeed­back train­ing are over­all rare, and when they occur they are rel­a­tive­ly tran­sient or read­i­ly dealt with by com­pe­tent prac­ti­tion­ers (Ham­mond, 2001; Schwartz & Schwartz, 1995).”

So here we have a treat­ment option that has been stud­ied for over 50 years, has no negative/side/adverse effects, has tons of evi­dence sup­port­ing it’s effi­ca­cy, but does­n’t have a sin­gle com­mer­cial enti­ty that “owns” it in the same way that drug com­pa­nies and med­ical device com­pa­nies own their solu­tions. No com­pa­ny is push­ing for FDA approval — or stud­ies — or mar­ket­ing it, because it’s not cost effec­tive for them.

Cyberon­ics was able to get FDA approval, accep­tance by the neu­rol­o­gy indus­try and insur­ance cov­er­age for their VNS med­ical device for epilep­sy with stud­ies show­ing more dubi­ous effi­ca­cy than EEG neu­ro­feed­back and with well estab­lished, poten­tial­ly seri­ous adverse risks. It tru­ly infu­ri­ates me to see how the com­mer­cial aspect of the med­ical indus­try dri­ves options for patient choice in treatments.


(Note:  I will now bring the few com­ments that fol­lowed, so it is eas­i­er to con­tin­ue the con­ver­sa­tion here).


  1. Alvaro on February 24, 2008 at 11:41

    Dear Bernard, thank you for such a detailed response. If you don’t mind, I will pub­lish it as its own post next week, so that more peo­ple can see it, and respond.

    You raise many good points. And I am hap­py that neu­ro­feed­back helped your wife. Now, noth­ing that you have writ­ten, in my view, con­tra­dicts Dr. Rabin­er’s con­clu­sion. Fur­ther­more, the fact that neu­ro­feed­back has been around for so long, may make one ques­tion why no one attempts the type of study he proposes.

    In fact, in those same rat­ings you link to, one can see only one appli­ca­tion reach­es Lev­els 5, and it is not ADD/ ADHD. Which is basi­cal­ly what Dr. Rabin­er is say­ing, with oth­er words.

    This is not about drugs vs. no drugs. Some non-drug-based approach­es, such as cog­ni­tive ther­a­py or forms of com­put­er-based work­ing mem­o­ry train­ing are start­ing to show effi­ca­cy as com­ple­ment to drugs in well-designed tri­als. And it is not about biofeed­back in gen­er­al, which has clear research sup­port for spe­cif­ic applications.

    You may know that com­pa­nies like Play Atten­tion sell neu­ro­feed­back machines direct to par­ents, for use with kids who have ADD/ ADHD. Have you seen research sup­port­ing an invest­ment of more than $1,000 in such pro­grams? They may work, and par­ents are free to spend their mon­ey as they want- but they deserve to know that it is not proven.

  2. Bernard on February 24, 2008 at 11:47

    Hi Alvaro,

    I don’t mind if you repub­lish the comment.

    Play Atten­tion is not a full fledged neu­ro­feed­back machine like the ones used in stud­ies. It only process­es (IIRC) the alpha wave chan­nel. EEG Neu­ro­feed­back machines like Brain­mas­ter’s Atlantis sys­tem, Zen­gar Insti­tute’s Neu­ro­Care sys­tem, etc. man­age up to 5 wave chan­nels — alpha, beta, theta, delta and (some­times) gam­ma. It also does­n’t per­form QEEG map­pings, so it is not on the same lev­el as a treat­ment option as EEG neu­ro­feed­back as defined in the studies. 

    Fur­ther­more, the fact that neu­ro­feed­back has been around for so long, may make one ques­tion why no one attempts the type of study he pro­pos­es.” — Alvaro


    The most con­clu­sive test of neu­ro­feed­back treat­ment would include ran­dom assign­ment and a con­trol con­di­tion that close­ly matched the neu­ro­feed­back con­di­tion. For exam­ple, chil­dren could receive video game coach­ing from a sup­port­ive adult for the same time peri­od. Or, even bet­ter, they could do exact­ly what chil­dren get­ting the neu­ro­feed­back were doing but not receive direct feed­back on their EEG states.” — Dr. Rabiner


    EEG neu­ro­feed­back is not a “treat­ment” in the clas­si­cal sense. It is a train­ing pro­gram that requires tai­lor­ing over time by the prac­ti­tion­er and active par­tic­i­pa­tion from the patient. It is not pos­si­ble to sim­u­late this effec­tive­ly over time for a “blind” study, much less a dou­ble blind study.

    But the larg­er answer as to why no one attempts the type of study he pro­pos­es is $$$:

    For a very promis­ing treat­ment tar­get­ing such a
    seri­ous con­di­tion as epilep­sy, the num­ber of large-scale clin­i­cal tri­als of neurofeedback
    train­ing to date is dis­ap­point­ing. A like­ly rea­son for this state of affairs is that
    neu­ro­feed­back research is a very time- and work-inten­sive enter­prise that has
    tra­di­tion­al­ly not received exten­sive research fund­ing and has, for obvi­ous rea­sons, not
    been pur­sued by the phar­ma­ceu­ti­cal industry.”

    As I men­tioned pre­vi­ous­ly, no com­pa­ny owns a patent on EEG neu­ro­feed­back — it’s a train­ing process, not a drug or spe­cif­ic med­ical device. It is not eco­nom­i­cal­ly jus­ti­fi­able for a for-prof­it com­pa­ny to spon­sor the nec­es­sary research. There is no ROI. So, it’s up to non-prof­it groups/charities to spon­sor this kind of research. While some have spon­sored small stud­ies recent­ly (such as FACES at NYU), there has­n’t been and isn’t any­thing being done on the scale that appears to be nec­es­sary for the skeptics.

  3. Alvaro on February 25, 2008 at 12:30

    Hel­lo Bernard, you raise some good points but, net net, the sit­u­a­tion is what Dr. Rabin­er describes: neu­ro­feed­back is promis­ing, and may well work, but it is unproven accord­ing to sci­en­tif­ic standards.

    Parents/ oth­er users need to make their own deci­sion on whether to use neu­ro­feed­back or not, as a tool in the toolk­it, with its Pros and Cons. There is not enough clinical/ sci­en­tif­ic evi­dence to war­rant a wider, auto­mat­ic “pre­scrip­tion”.

  4. Bernard on February 27, 2008 at 7:47

    There is not enough clinical/ sci­en­tif­ic evi­dence to war­rant a wider, auto­mat­ic “pre­scrip­tion”.”

    How much is enough?

  5. Alvaro on February 27, 2008 at 8:36

    Hel­lo Bernard, with all my respect, it is not even close.

    One thing is to argue that neu­ro­feed­back may be an appro­pri­ate tool in spe­cif­ic cas­es where a clin­i­cian can sug­gest that inter­ven­tion makes sense and the patient agrees to try, and anoth­er one to imply the tool deserves a wider, auto­mat­ic “pre­scrip­tion”. For the lat­er, you’d need many high-qual­i­ty ran­dom­ized con­trolled tri­als-the type that Dr. Rabin­er was sug­gest­ing neu­ro­feed­back, at least in the case of atten­tion deficits, has none.

    Which does­n’t mean neu­ro­feed­back isn’t a valid option. It CAN be. Let me revis­it  one of David’s sen­tences: “How­ev­er, these stud­ies do pro­vide a sol­id basis for sug­gest­ing that if par­ents choose to pur­sue neu­ro­feed­back for their child, there is a rea­son­able chance that their child will ben­e­fit even though we can’t be sure that it is the spe­cif­ic EEG train­ing that is respon­si­ble for the benefits.”

    If neu­ro­feed­back prac­ti­tion­ers need to go beyond this, they need to enlist some uni­ver­si­ty-based researchers and con­duct a good tri­al. Per­haps the mil­i­tary would be inter­est­ed in fund­ing such a study.

  6. Peter Freer on March 8, 2008 at 10:34

    The infor­ma­tion regard­ing Play Atten­tion is incor­rect. It is not an alpha train­er. It is a phys­i­o­log­ic mon­i­tor that can­not be aug­ment­ed from fac­to­ry set­tings like an EEG device. I want­ed to make it clear that Play Atten­tion is not clin­i­cal neu­ro­feed­back and has nev­er strived to be. Frankly, we don’t try to change brain­waves and are uncon­cerned with their change. We are con­cerned with cog­ni­tive and behav­ioral out­comes. The feed­back used is sim­ply to place the stu­dent in an atten­tive state to learn deficit skills like time on-task, mem­o­ry, fil­ter­ing out dis­trac­tions, visu­al track­ing, audi­to­ry pro­cess­ing, etc. We have a patent pend­ing on our inte­grat­ed behav­ioral shap­ing com­po­nent as well. Neu­ro­feed­back pro­po­nents have main­tained that chang­ing brain­waves facil­i­tates learn­ing of these skills, how­ev­er I find that rather facile. Chang­ing a brain­wave does not insure that I will learn to fin­ish home­work on time, have bet­ter short-term mem­o­ry, etc. If that occurs, it’s sim­ply by coin­ci­dence not by osmo­sis. We strive to teach these skills directly.

  7. Alvaro on March 8, 2008 at 5:48

    Hel­lo Peter, thank you for your comment.

    What you say makes a lot of sense. Now, the ques­tions then become, how exact­ly do you train cog­ni­tive skills, and how do you show the “cog­ni­tive and behav­ioral out­comes” from your prod­uct? We have not come across peer-reviewed pub­lished stud­ies that we can eval­u­ate and dis­cuss in this blog. And we would love to, giv­en the num­ber of years your prod­uct has been in the market.
    Thank you

  8. Peter Freer on March 25, 2008 at 8:13

    What I say does make sense. Specif­i­cal­ly, one teach­es cog­ni­tive skills. We casu­al­ly call it train­ing, but cog­ni­tive skills can be taught. This has been done his­tor­i­cal­ly since the time of Socrates. The feed­back insures we have stu­dent atten­tion. It acti­vates and con­trols cog­ni­tive activ­i­ties like work­ing mem­o­ry prac­tice, spa­tial mem­o­ry, time on-task, visu­al track­ing, audi­to­ry pro­cess­ing, atten­tion sta­mi­na, etc. If the stu­dent fails to main­tain max­i­mum atten­tion, the game stops, alert­ing them to focus-in once again. We then quan­ti­fy how much mem­o­ry improves over time, teach stu­dents to fin­ish assign­ments on-time, mea­sure visu­al track­ing abil­i­ty, etc. Then we spend a lot of time on strate­gies to trans­fer and gen­er­al­ize these skills.

    We’ve received 3 patents based on this approach. We also have a patent pend­ing behav­ior shap­ing mod­ule inte­grat­ed so that stu­dents can learn to self-reg­u­late by learn­ing to focus more. It’s far too dif­fi­cult to explain in a short response like this.

    We have an ongo­ing con­trolled study at Tufts Med­ical School, are begin­ning one at Hard­ing Uni­ver­si­ty, and are start­ing one in a pub­lic school dis­trict next year. Data should be pub­lished in late 2009. 

    The rea­son we’ve sur­vived and ruled this mar­ket inter­na­tion­al­ly for 14 years is because our method makes sense, is found­ed in sound teach­ing prac­tices that are thou­sands of years old and inte­grat­ed them with feed­back tech­nol­o­gy. That’s so rad­i­cal­ly dif­fer­ent and refresh­ing that we are in more house­holds, TBI hos­pi­tals, psy­chol­o­gist’s offices, MD’s offices, learn­ing cen­ters, and schools (over 450 dis­tricts in US alone) that no one in this indus­try comes close. Could we have done this if Play Atten­tion did­n’t work? Not like­ly. Fur­ther­more, what we’ll intro­duce both tech­no­log­i­cal­ly and edu­ca­tion­al­ly over 2008 will alter the future of this indus­try once again. 

    These forums are noto­ri­ous for wrong­ly inter­pret­ing or describ­ing Play Atten­tion which is what occurred here. I only meant to cor­rect that. One should be respon­si­ble or at least knowl­edge­able before mak­ing state­ments espe­cial­ly on a respect­ed forum such as this.

  9. Bernard on January 7, 2009 at 10:34

    I have ref­er­enced this dis­cus­sion in a let­ter to the Amer­i­can Acad­e­my of Neurology:

About SharpBrains

SHARPBRAINS is an independent think-tank and consulting firm providing services at the frontier of applied neuroscience, health, leadership and innovation.
SHARPBRAINS es un think-tank y consultoría independiente proporcionando servicios para la neurociencia aplicada, salud, liderazgo e innovación.

Top Articles on Brain Health and Neuroplasticity

Top 10 Brain Teasers and Illusions


Subscribe to our e-newsletter

* indicates required

Got the book?