Mapping the Mind: How EEG Reveals Distinct Brain Phenotypes

A diagnosis can name the problem, but it often can’t tell you what will actually help. We sit down with legendary EEG and QEEG expert Jay Gunkelman, who has read hundreds of thousands of raw EEGs, to explain why “endophenotypes” change everything about brain-based interventions. Instead of chasing symptom lists and billing codes, we zoom in on clear EEG patterns that reliably show up when brain function fails and, more importantly, point toward interventions with better odds of success.

Jay breaks down what an endophenotype really is: a measurable neurophysiological pattern that sits between genetics and behavior. We unpack why that middle layer matters for clinical outcomes, especially when two people share the same DSM label but need completely different approaches. You’ll hear how simple pattern recognition can produce much higher agreement across clinicians than traditional interpretation, and why that reliability is essential for treatment planning and for any future AI-driven EEG tools.

We also dig into one of the most compelling examples from Jay’s early work: the low-voltage fast EEG pattern frequently seen in chronic alcoholism, its connection to overarousal and GABA systems, and why alpha training through neurofeedback helped some people regulate without relying on alcohol to feel “normal.” Finally, we explore research showing neurofeedback can drive real brain plasticity, including changes in both gray matter and white matter, reinforcing that training the brain is not just theory but measurable physiology.

If you care about neurofeedback, EEG patterns, brain health optimization, or evidence-based mental health care, hit play, then subscribe, share this with a friend, and leave a review so more people can find the show.

To learn more about Transformative Neurotherapy visit:
https://www.TransformativeNeurotherapy.org
Transformative Neurotherapy
570 Lincoln Ave. 
Bellevue, PA 15202 
412-204-7397  

Transcript

Welcome And Big-Picture Brain Health

SPEAKER_00 0:02

Welcome to the Transformative Neurotherapy Podcast with your host, Dr. Heather Futter, founder and executive director of Transformative Neurotherapy. This is the place where feelings happen faster. Because let's face it, your brain doesn't come without. Here we take a holistic approach to brain health, bringing the other science, mind-body harmony, and the tools you need to optimize your well-being. Whether you're a high performer, executive, athlete, hacker, or just someone tired of your brain working gifts. Dr. Buddy is here to help you unlock your full potential. From brain fog to chronic stress, we are covering it all so you can finally experience brain health, mind harmony, and total well-being. Ready to get on stock? Let's get started.

SPEAKER_01 0:56

Understanding phenotypes changed the way we approach brain-based interventions. Today we'll explore how those distinctions emerged and why they matter for clinical outcomes. Welcome.

Jay Gunkelman’s EEG Origin Story

SPEAKER_01 1:06

I'm Heather Putney, founder and executive director of transformative neurotherapy, here in the studio today with the infamous Jay Gunkelman, a top leader in the field of EEG and QEEG. So, Jay, for those of us, for those who don't know you, um, you're one of the people that has probably read the most EEGs, raw EEGs in the world. How many think it's been so far?

SPEAKER_02 1:29

I quit counting at 500,000 in the 1990s, and I have no clue. Um, you know, it and it doesn't really matter. I mean, it most people don't even believe the 500,000. How do you get that kind of a number? I worked in the busiest EEG lab in the world. We did over 100 a day. And uh there were days we topped 200, but if you just use 100 a day as a calculation, because it's easy math, uh you know, uh I was there uh from 1996 to the mid-90s. And if you just do the math, my doctor was too cheap to give more than two weeks of vacation a year. So 50 weeks, uh, I did call on weekends, but just don't count weekends, just to make the math easy, five days a week, you know, um you know, 100 a day, uh, times that many years. The math comes out at about 500,000. So um, but uh again, that's that's a conservative estimate, but uh uh nobody believes that anyway. So uh but the lab we were in uh had 400 hospitals uh from all over the U.S. that could transmit. Um we did the first transmission of an EEG over a telephone line uh and reconstructing it at the receiving center in 1974 was the first time it was demonstrated. And I was working in that retraining people how to how to hook up EEGs properly for the first eight months, but after that I ran the lab in San Francisco for all those years.

SPEAKER_01 3:07

Yeah. Well, you truly are a pioneer in the field. And one of the things that you're very well known for is um kind of the discovery, or I or I should say kind of the um I don't know, it's not so much the discovery.

SPEAKER_02 3:23

Stumbling upon stumbling upon is more accurate, I think, than anything else. So the the the endophenotypes. Yes. And you know, uh if you've been looking at EGs and paying attention for all of those EGs, eventually patterns come out of the data and your data kind of speaks to you. And when your data's speaking to you, you have to publish about what it's saying because either you're crazy as a loon and your data is speaking to you, or it's saying something important. And until you publish it, nobody can test the hypotheses that would be generated by the by the data.

Endophenotypes Over DSM Labels

SPEAKER_02 4:08

And, you know, I I simply looked for patterns that were common. And the basically when you'd see somebody that had a failure of brain function, what was the pattern that they had? And instead of paying attention to the diagnosis, I paid attention to the pattern and what therapy worked with it. Because unfortunately, as well intended as the DSM was, and DSM one was just before I was born, and now DSM six is being proposed soon. But the the the intention is diagnosis, therefore therapy. If we get the diagnosis right, it'll tell us exactly how to treat you. But there's no predictive validity at all. They've pushed for diagnostic specificity, but there's no benefit other than a billing code. I mean, you do get administrative benefits for getting a diagnosis right and everything, but it doesn't, you know, it doesn't tell you what to do and how to treat. And I simply looked for common failure modes, and eventually I had this list in my head of these common failure modes, and I was doing lectures about common failure modes, and I noticed one day that two of the 11 that I had identified had known genetic correlates. So those patterns were expressed genetics. An endophenotype is a step in between genetics and behavior. It's an expressed gene. You can have a gene that's not expressed and never have anything associated with it, you know. But if it's expressed, you end up having a neurophysiological pattern that's identified, and that matches you up with the thing that will fix it, both with neurofeedback as well as medication.

Alcoholism Patterns And Alpha Training

SPEAKER_02 6:08

And so the first one I spotted was the low voltage fast EEG, and it's extraordinarily common in alcoholism. In 1972 to 1975, before I worked in this big busy EEG lab, I had the first state hospital-based neurofeedback, biofeedback laboratory in the world. We wrote a grant as kids. I mean, there was no IRB. There was no proven efficacy for anything. There was there was just kids with equipment and turned loose in a hospital. And oh goodness, uh uh, they should have had better sense than that, really. But, you know, luckily we weren't uh uh badly motivated and and we we've we found some very good uh outcomes. When you hook up an alcoholic that's uh been there a number of times and they volunteer to come into the lab to be tested, and you look at their EEG, one of the most common patterns in chronic alcoholism is that they they have an EEG with no apparent rhythmicity, a low voltage, fast EEG. And when you see that, it's gene number four. And gene number four makes GABA receptors. GABA is an inhibitory neurotransmitter, and if you have more receptors than you can saturate, you're over aroused. So these alcoholics are so over aroused they have to consume alcohol to get GABA to saturate them and wind down their arousal level. They start with no alpha, none. And if they have two shots, I mean, literally two ounces of hard liquor. We're not talking cook two sips of beer, we're talking shots. If they do two full shots of alcohol, they start to make some alpha, but it's not as much as you would have in the beginning. By four shots of alcohol, now they're making alpha like you were when you started to sit down for your first sip. Now, that's a slippery slope. If it takes you four shots to be normal, and you're gonna be feeling normal. It's a seductive thing. To feel normal, you want to feel normal again. And you know how you got there uh through sipping on alcohol. Now, it's not a long-term productive way to do it. I mean, you can learn how to regulate your own body and saturate your own GABA receptors. Uh the the alpha can be trained. When we saw no alpha, remember, 21-year-old kid, stupid, and I mean, educated but stupid. I mean, you you you have book learning, but y you've never seen clinical operations before. So you you look at the EG and see no alpha and you think, oh, let's train alpha. There's two kinds of training you can think about at that point in time in 1972, SMR and alpha. So uh when you look at an EG that has no alpha, you think, well, let's do some alpha training and see what that does. And as simplistic as that was, it worked like a miracle. The people that learned how to make their own alpha, 14 to 20 sessions worth of training, which is you know, a few times a week. So and in a state hospital setting for alcoholism, you're there for a while. This is not a check in for a week in Malibu with massages and good food. No, no, no, no. This is the state hospital in North Dakota. And, you know, it it wasn't a malicious facility, but i you know, it was a state hospital. You know, that uh uh if you ever saw one flew over the cuckoo's nest, you saw the state hospital in North Dakota. Exactly. It's the same characters, uh the uh the it was, you know, uh uh uh an identity, you know. So um we trained alpha and they got better and didn't come back. And the most obvious thing in the alcoholism drug division was recidivism. The the recidivism in a state hospital setting is not like a good 12-step program that might have a 30% success rate, and the good ones do. This was maybe getting 10%, if that. I mean, people would just come back. And if they're uh down in the gutter alcoholics, they're gonna be back fairly soon. They don't really have any career to do in between visits, so and when these people didn't come back, it was noticed, you know. Well, gee, wh wh where's you know, Mike? You should be back by now, you know. Um and after doing that for a few years and seeing people we worked with not come back through, um, in 1974, two years worth of work with this, we wrote a grant to NIH to see if we could get this studied because it looked like it was working like a charm. And this is 1974, which is before Penniston published in '89. So this is a decade before uh uh the alpha theta were coming out of Penniston's. So um, you know, we we had stumbled upon alpha training and alcoholics as something that was beneficial. Uh we we did EMG relaxation, we did temperature training, we worked, we were on the front page of the Jamestown Gazette uh in 1974 in a color picture on the front page above Nixon, uh below the uh fold, you know, um he was having some economic difficulties at the time. So um anyway, what that that worked out well. Um but I paid attention when I moved from the state hospital treatment neurofeedback-oriented center to just straight medical EEG. And the 100 a day was not neurofeedback, this was straight medical EG. And after you see your first few hundred thousand, you start to actually know what you're looking at. Um there's nobody I know of that's seen as many as I have. And uh uh the people who are trying to train algorithms to look at EG need big data sets. So most of them have come to me asking for my data sets. And I'd have to tell them the data doesn't belong to me. I can't give it to you because it doesn't belong to me. It belongs to the clients who are clients of my clients. I was writing a porch for the therapist, and it's their patients, not mine, and I can't give you their patient data, and and so no, I'm sorry. I it would be lovely if I could, but I can't.

Why Simple Patterns Beat Algorithms

SPEAKER_02 13:21

And they need that mass of data to train their algorithms in order to be able to replace human uh reading of EEG, which is at this point, neurologist and neurologist, the agreement side by side, pull two doc two doctors that both have the same degree. One was trained in uh uh Rutgers, and one was trained at UC San Francisco, or any two schools, di different schools. The agreement between neurologists side by side, reading a bunch of EGs, 30 to 40 percent agreement. Well, you know, that's not even good enough odd to get me to go to Vegas, you know. I mean, I'm not gonna get a good outcome here. So uh uh the good thing about the phenotypes is it's not a diagnosis we're looking for. We're looking for a pattern that's easily identifiable that can predict an outcome. Now, after I published the paper, the paper has no single waveform in it, not one. It just you know talks about the EEG patterns, frontal alpha, frontal theta, temporal, uh slowing. I mean, the these are all simple patterns you can see in the EEG. The interrator reliability between myself, the apparently the one they thought might know what they were looking at, you know, it was my my model, so uh uh they used me as a standard of what was correct. Martine Arnes invited me to uh lecture in Europe, and when I showed up, I had to sort out a whole bunch of uh phenotypes and on on EEGs. He'd already done it himself, having read my paper. We had better than 90% agreement. So the the the Cohen's Lambda, which is a statistic of of how reliable is your or how repeatable uh the the the diagnosis would be, or categorization in this case, not diagnosis.

SPEAKER_01 15:36

So the development of these or the discovery of these endophenotypes really is a way to kind of make sense out of what the clinicians are seeing, and by by having that framework, it it helps people um to see the same thing or to kind of come to the same conclusions. So like that reliability improves a lot then with the field. So that in and of itself is a huge contribution.

SPEAKER_02 15:59

Yeah, it's and it's simple pattern recognition. We're not talking about some subtlety or a fancy wiggle that you have to recognize or something. This these are all real, real simple patterns. And that uh if uh an untrained raider who's never seen a single EG with me pointing at what the phenotype was, read the paper and got 90% accurate, better than 90% accurate, um, that it's easy to see, you know. Uh uh, if you can't see that there's alpha in the front, you should perhaps get some glasses, you know. I mean, it's not these are not hard to detect patterns. Just real this is uh the face validity of these patterns is absolutely right in your face. So uh it's easy to spot the patterns, and the patterns have reliable predictions.

Neurofeedback And Real Brain Plasticity

SPEAKER_02 16:55

And that you know, ultimately predicting therapy, the prognosis is more important than the diagnosis. Diagnosis is gonna give you a name to call somebody, and sometimes they're used pejoratively. You know, euro this or euro that, you know. But nobody's ever waved waved their finger, you've got frontal alpha, you know. So, you know, uh these patterns aren't pejoratives. You can't complain about them as a as a finding. And they're they're uh uh neurofunctional patterns that can be trained. Uh you can uh get rid of or enhance brain patterns, the brain's plastic, and I know plastic has got a terrible name nowadays, you know, microplastics and all that, but plasticity is malleable, formable, reformable, and the brain is the organ that forms itself. And, you know, I used to think that, well, gray matter, yeah, that's real plastic, but that white matter, that's your wiring, once it's wired up, that's kind of permanent. And boy, was I wrong. You know, uh uh I I do appreciate uh moments that uh uh make me feel like I'm stupid because I think it's a reflection of reality for a moment, you know. But Mario Beauregard did two studies that showed both gray matter and white matter are sculpted by the neurofeedback. Uh Montreal Neurological, major institution, major researcher, not selling you a damn thing, just doing research. And I didn't, it still didn't sink in. It took the third whack in the head to to finally penetrate. Dr. Panada, UC San Diego, uh, they did diffusion tensor imaging on autism, and they had done neurofeedback. And the areas that were being trained in neurofeedback sculpted the white matter in those areas. I expected the gray matter improvement. I did not expect the white matter. And you could see the DTI. I mean, the if if the MRI's got a big finding, big thing over here and suddenly it's gone, uh the the white matter problem is no longer an issue. You cleared a quadrant of the brain uh um because of um because of the training, the the that that's astounding. And you know, I uh um it's it's so important because you know, if you have something not working and it's not possible to train it, and in neuroscience they said the brain is part of the involuntary nervous system, so how the hell can you voluntarily change it? Well, their naming for it was a naming convention error. You know, our voluntary nervous system is mostly voluntary, but some of it's not. Our muscle tone is actually sympathetically innervated, it's part of the involuntary nervous system. So our voluntary stuff isn't fully voluntary, and our involuntary stuff isn't totally involuntary. Uh Neil Miller in the 1970s uh uh it was proving that you could take classically conditioned things that they thought were involuntary, and you could classically condition it like a dog salivating Pavlov. Uh it you but you could now operately condition them. And uh that was switching over to voluntary nervous system kind of function for these involuntary functions. And it was pissing off the neurophysiology people because it was the involuntary nervous system, after

Voluntary Control And The Closing CTA

SPEAKER_02 20:45

all. Well, uh uh uh they said to Neil that your your your publications are false publications because uh you're using subtle muscle tension patterns to control the blood pressure. You're not really voluntarily controlling your blood pressure. How do you prove that they're wrong and that you're not defrauding science with your work? A student, Leo DeCara, got hold of some Carrari and no no students and drugs, he took the Carrari, paralyzed himself, so they had to bag him to have him breathing. And he demonstrated voluntary control of his blood pressure while paralyzed with Carrari. No, you can't you cannot get that through an IRB.

SPEAKER_01 21:40

No, please not. So would recommend trying.

SPEAKER_02 21:44

You know, but it proved beyond a shadow of a doubt that this was actually not mediated by muscle tension. And Neil Miller realized the importance of students, and he started FERB, the Foundation for Education, Research and Biofeedback many, many years ago, back when the Biofeedback Research Society was going. And I picked up on that, and I've been trying to support students ever since I heard about that. I'm I met Leo.

SPEAKER_01 22:15

I knew Yeah, and and truly, truly, you have been helping develop uh the um the students moving forward in the and the uh future of the field of EEG and your discovery of um endophenotypes has massively changed how we approach the way we work. And I think just like what you were saying before, um, when you can understand the phenotypes, you can also, it implies potential interventions, you know, and what you can do with the neurofeedback or the other interventions. And in our next episode, which is coming up right after this, we're going to be talking about how you use those endophenotypes to help uh predict medication interactions and um you know improvements or sometimes what to avoid uh to avoid. So if you're interested in hearing more from Jay around around these things and the applications with the um uh with medications and improving outcomes, then tune right into our next session. But thank you, Jay, for your contributions to this field and um and all that you do for for all of us.

SPEAKER_02 23:14

My pleasure. Thank you. Nice to be invited.

SPEAKER_00 23:22

You've been listening to the Transformative Neurotherapy Podcast with Dr. Heather Putney. Remember, your brain isn't supposed to hold you back, it's supposed to power you forward. So stop letting it crash your party and start letting it do its job. If you're ready to optimize brain health, sharpen your focus and age like a five five. Schedule your free consultation today at Transformative Neurotherapy.org or call us at 412-204-7357.