Understanding Neurofeedback: Brain Optimization - Dr. Andrew Hill

Understanding Neurofeedback

This episode of the podcast features Dr. Andrew Hill, a cognitive neuroscientist at UCLA, founder and CEO of the Peak Brain Institute, and an expert in the field of neurofeedback. Dr. Hill discusses the science of neurofeedback and using neurofeedback to increase cognitive performance.

Show Notes

0:00 intro
1:52 What is biofeedback
4:57 Biofeedback: Associative learning
6:09 Neurofeedback: Operant conditioning
9:38 On EEG-based cognitive neuroscience
14:02 Quantitative EEG (QEEG) analysis of the brain
16:06 QEEG-driven neurofeedback
20:21 Infra-slow and ultra-low EEG
22:10 Low Energy Neurofeedback System (LENS), High Performance Neurofeedback (HPN)
23:12 Sensory motor rhythm (SMR)
26:40 EEG for ADHD
27:28 Primary mechanism during feedback
29:19 Risks of trying to optimize a system we don’t understand that well
33:28 Using neurofeedback to help the system increase its own regulatory capacity
43:49 Neurofeedback and clinical research studies
53:40 Resources for neurofeedback research
1:01:53 Neurofeedback and alcoholism
1:08:16 Who neurofeedback is good for
1:26:55 How to find a neurofeedback training practitioner

Links from the Episode:

• Peak Brain Institute
• PubMed
• International Society for Neurofeedback and Research
• Peniston-Kulkosky Brainwave Neurofeedback Therapeutic Protocol
• The Willpower Instinct - Kelly McGonigal

Transcript:

Daniel S.: Welcome to the podcast for Qualia. My name's Daniel Schmachtenberger. I work with Qualia in research and development. We are delighted to have Dr. Andrew Hill with us today. Andrew Hill is a cognitive neuroscientist from UCLA and really an expert in the field of neurofeedback, amongst many other things in terms of bio signal processing and cognitive performance, performance optimization, and gerontology. We'll see where we get but we're really gonna focus on understanding neurofeedback, which is the method for letting the brain get signal about its own function and optimize itself, which is super interesting.

And we're going to start at the beginning. Both because this isn't a topic we've talked about for our listeners before, and because it's actually a topic that I want to make sure I understand the foundations well, because it's a topic of high interest but not high familiarity for me. So Andrew, thank you for joining us.

Dr. Andrew Hill: Oh, thanks for having me, Dan. I really appreciate it.

Daniel S.: Andrew is the CEO, Founder of the Peak Brain Institute, which is a place that's really kind of advancing, synthesizing the science of neurofeedback and associated neuroscience discipline, and then also bringing best in class neurofeedback for a variety of topics to the public. We're gonna talk more about ways that people can learn further, beyond this podcast.

So let's start with, neurofeedback is a form of biofeedback.

Dr. Andrew Hill: It is, yeah. All neurofeedback is biofeedback, although not all biofeedback is neurofeedback.

Daniel S.: So do you want to just define briefly, what is biofeedback?

Dr. Andrew Hill: Biofeedback at its basic level ... Or feedback in general, let's see if you can just drop the bio for a second. Feedback within living systems or within all systems are control mechanisms. They're ways of tuning and adjusting and stabilizing and providing different sort of stability resting points for complex systems to live in. Sort of balance in these stable basins. And when it comes to EEG and other physiological signals, you're in biofeedback when you are measuring something that is not normally perceptible. Biofeedback in a psychological cognitive psychology sense is taking a signal that's not normally perceptible, be it your heartbeat, your galvanic skin response, your hand temperature, your muscle tension somewhere, and if you measure these symbols in real time, you provide a control mechanism essentially, that really maps to how the body and brain learn in general.

Not to be specific, this is not some magical property of the brain, this feedback learning. And it's not even a magical property of living systems, because all systems that have stability tend to learn in this way. They tend to have feedback mechanisms that provide a certain shape and character and quality to that complex system. And for us, adapting to the environment is so huge that we continue to couple our brains to the environment and learn from it, almost as an extension of our own brain.

So in biofeedback, if we simply provide the signal of, "here's the information" back to the brain, it can usually go, "Oh, that's interesting. That's me." And sort of couples that information. When doing audio based EEG biofeedback, we found that the coupling signal kicks in within 10 or 15 minutes, and you can find it in the EEG. I assume something similar is happening in things like heart rate variability and galvanic skin response, but I'm not quite as familiar with nuances of those signals as I am in EEG, especially where it is. But essentially by providing a signal to a person, in biofeedback, meaning peripheral nervous system biofeedback, that's the line I'm drawing between biofeedback and neurofeedback is, which nervous system are you working on?

If you're working on the brain, I'm calling it neurofeedback or the central nervous system, and if you're working on the body, peripheral. And the line between them, for anyone unclear, is that anything that's central nervous system is encased in bone. And if it's outside of bone, it's peripheral, broadly speaking.

So when you're measuring these peripheral mechanisms that are regulatory, and you're providing a signal that's changing back to the brain, it goes, "Oh, interesting." Then you do something that is different a little bit in biofeedback and neurofeedback. It's a form of associative learning. And in biofeedback, the associative learning is learning to perceive the feeling of the signal. The subtle nuance sort of sensation of taking control of the parameter. It's very sort of hard to describe until you've done a little bit of it, but you practice arrhythmic almost, responding to the equipment mirroring back to yourself something going on. And after a bit, you kind of get a sense of control. That's called skill transfer, essentially.

You're sort of initiating a regularization of the feedback, of your sense of it. If you practice that enough with the peripheral biofeedback mechanisms, you get some control over HRV, GSR, hand warming, and a whole host of other features because you've given the brain some information about the PNS that it can use to control the PNS, just like it's controlling the PNS with a thousand other signals. It's extract that information from that signal, and use it to tune the system.

There's a little bit of shaping or operative conditioning or instrumental conditioning in peripheral biofeedback. But neurofeedback or central biofeedback is almost entirely operant conditioning, instrumental conditioning. There are some small ways in which there are some types that are not just that. But the vast majority of neurofeedback as it's practiced in the clinical world and the research world, is a form of operant conditioning. Now let me break that down for a second, because this is true in peripheral biofeedback a little bit.

You're basically setting a threshold and saying, "Hey, system, go that direction." And so whenever you exceed a threshold or cross a threshold, the system gets applause in some way. Audio/visual feedback is pretty classic. The trick about bio peripheral feedback when you're giving a shaping operant conditioning award is, you kind of know what's happening when you cross that award. You get a sense of it, and that's the transfer. You learn to sense that moment of achieving some change in threshold in the parameter. The brain maps a weird sense for you, almost like adding a sense for your meet so that you get like ... It's almost like adding an extra oil gage or rev meter to your car. It's like, "Ooh. More information."

And if you practice it in peripheral biofeedback, if you practice it, you get control of the skill. And if you keep practicing it, you retain control over the skill or access to that initiation of the skill. But if you don't practice it, you generally lose the skill. That's sort of one of those things. This is why voluntary willful forms of bio hacking, if you will, that are subtle, that include measuring parameters that are not usually under physiological control, that even measure parameters ... They're not measured. Things like meditation is almost a form of biofeedback. The control signal is probably more negative ... Sorry. It's still positive feedback, but it's probably adversive, "Oh, I'm not doing it right." Kind of experience when you're meditating. That's the feedback. But generally neurofeedback and biofeedback are positive. You apply a stimulus.

Now, I mentioned operant conditioning. I do want to say one thing about that. This is not Pavlov's dog. These are Skinner's pigeons. So Pavlov's dog in classical conditioning, you take two things that aren't normally associated, and you link them. You associate them. So a dog salivating with a bell ringing. Not normally something that makes dogs salivate, but if you pair it with a salivation induced stimulus, you can eventually transfer the learning from the steak to the bell. Skinner on the other hand, took behavior that already existed, pecking and stepping, and shaped the direction, the amount, the quality of that behavior so pigeons would ... He really wanted them to peck three times on a bar. So for a while, he'd reward a single peck on the bar, and then eventually a double peck, and then eventually a triple peck, and to shape behavior. Pigeons are able to learn fairly complex behavior, even though they have fairly tiny little brains, and they're actually rather superstitions.

So if they do something one time, and they get an extra reward, an extra dopamine hit, they may actually put a stylized behavior into ... A behavior they've learned operantly conditioned, without any need for it. Basically they're like a baseball player, "Oh, my God. I hit a home run. Let me do this every time." And they remember those things especially well. And so superstitious pigeons is kind of a problem in research.

Neurofeedback, the big difference here, is that you can't perceive this stimulus that's being measured. You can't perceive your theta changing, your beta amplitude changing, your average frequency of alpha changing, your coherence or connectivity between two regions fluctuating in time. And that last bit is the most important, fluctuating time. These are not static things. These are ... Not even one of these measures as static or independent. There's literally billions of things happening to produce this one brainwave. You're measuring some parameter of, in most forms of neurofeedback. Be that the DC Coupling on which it's riding, which would be things like ultra slow or slow [inaudible 00:10:24] potential frequency. Things like the amount of specific bands. Things called theta or beta or alpha. There's other forms of neurofeedback that might look at relationships between regions in these signal bands. There's lots of things you can train. There's a near infinite variety that you can get.

One of the things about neurofeedback is that we don't really understand the brain, I mean scientists in general, don't understand the brain. It's still the most complex thing we come across, probably will ever come across, unless we succeed far beyond Ray Kurzweil's dreams, and we're not really sure the way EEG is fully produced either. As a neuroscientist, as an EEG based cognitive neuroscience, cognitive neuroscience studies how mind and brain are meshed. Brain produces the mind. They are usually resource focused. For me it's about attention and also laterality, how things are distributed in left/right hemisphere and how that changes how you are, who you are, what resources you have, which is also trainable to some extent.

But the skill of cognitive neuroscience and EEG is an EEG methodology. Science is methodology. I tell students of mine when they're thinking about grad school, don't worry about where you go or even really all that much, who you work with, as long as you're not too early and too late in your career. Beyond that, the questions you should be asking yourself are what problems do I want to work on and what methodology do I really want to use? Do I want to learn? Do I want to get deep into? Because PhD, as the joke goes, is the art of learning more and more about less and less until you know everything about nothing. You get there through an exploration of a space, of a science of knowledge using a specific lens and tool set, and these things are complicated and take a lot of effort.

Well, EEG's an incredibly deep field, and we don't understand at it's base level what's really going on, we understand the cellular level how EEG is produced. It's firing of mini columns or micro columns in the cortex. These are mini CPUs. 30,000 neurons. A group of glial cells. Some vasculature. Some long term and short term connections. This is us doing a basic functional computational unit of the cortex. We know that. And we know that all these cells tend to fire together within this group, and when they fire at certain rates, the ones right angles perpendicular to the scalp or the skull, their signal aggregates enough it's perceptible through the attenuation layers to get picked up. That's probably a third to a half of the electrical signals, the dipoles being generated by micro columns that are actually pointing in the right direction. At least have are pointing into the sulci and not on the tip of a gyri, so you miss a lot of that.

FYI NEG measures the other horizontal access, but you still get only partial information. So EEG science gives you partial information from the scalp, of a complex system of a couple of billion CPUs, but are actually mostly tied to each other recurrently, where if you push on one, they all change.

So what's happening is we're measuring a phenomena, and especially as human focused cognitive neuroscience do EEG, especially neurofeedback people and attention driven people. We assess the brain from the perspective of what is the phenomena we're seeing and what might optimize? What might this mean? This is one of the things that people don't understand necessarily about things like EEG analysis of the brain. QEEG is one example, Quantitative EEG. Typically, in QEEG, which is a term that means simply, measuring, but it typically means measurement in contrast or in comparison to a normative database. That often raises a lot of eyebrows. "Wait a minute. Well, who are these people? These are average people? Well, I'm not average." Or, "I don't want to be average, so why are you comparing me to average?" That's the sort of typically knee jerk, and sometimes people assume that the goal of training from the QEEG, is to train to zero or z Scores or standard deviations to become average. Not typically. There are forms from our feedback where that is the goal, training everything at once auto magically, and you end up with a certain shift in z Scores.

That's not the traditional way to train even though, traditionally in good science we do use QEEGs. But we use them as tools to understand the brain, prognostic versus diagnostic. Tools to say, "You know, this pattern at a population level often means this." "Oh, hey. You're experiencing that. This might be where it is." So we use the sense of the shape of patterns that we get to get a sense of the brain that we're looking at and try to figure out from the hypothesis that generates, what's actually real in the data. An EEG of course, is very noisy data, too. Blinks and coughs and twitches and sleep deprivation and caffeine and illicit substances. All these things have a pretty profound effect. And age, a pretty found effect on the EEG.

So in the case of EEG there's an intermediate step in doing biofeedback. You typically need somebody to look at the brain patterns. Assess them against the normative database and then figure out what it is you want to do. Traditional if you would have to field 3/4 of the field would say, "Let's find the problems in QEEG. Let's do sort of neuro psyche level assessments, continuous performance tasks." Depending on their specialty, people use different assessment. But for the most part, it's the general neuro psyche, the sort of battery, that people tend to use and CPT is the most common, and then you figure out what you want to do.

Now, segway into different forms of neurofeedback, there's a QEEG driven form, if you will, that's sort of purely QEEG. That was the auto magic quip I made earlier. You tell the software, "Okay. I'm measuring all the brain or all of the scalp at once. We're gonna calculate in real time, standard deviations or z Scores from this comparison database. And when 70% of the ones that are more than a standard deviation and a half, move towards zero, provide a reward." But the next moment, it could be a different 70% if you're setting that kind of bucket threshold.

And so in that case, you're providing a fairly blunt push on the brain, and you're hoping it organizes in a way that it needs to for the most part. It's blunt but specific. What I mean by that is it's not controllable. You're sort of providing a, "Lemme see where the weak forces are. Let's see if the system rearranges when I push on the stuff that is most obvious." That does tend to work. QEEG driven, meaning z Score and neurofeedback. Talking about the z Score because you're in real times training from the database. It often works very well, and it's often very powerful.

I do QEEG driven neurofeedback, but I don't do it in real time from the QEEG. What I do is develop a pretty deep analysis of your brain. Hundreds of pages of norms of analysis of how unusual your brain is. We figure out, "Okay. Well, what makes this brain data valid and relevant for you? Are there problems here? Are there interesting, curious things you might find interesting about your interconnectedness or your speed of processing or what might this stuff say about how your resources are functioning?" I usually do that as a cold read. I do that as sort of a party trick, but it works 95% of the time the way I do it.

What I mean by that, not true like speaking to the dead, Jonathan whatever his name was, leading questions. I don't do the leading questions, but I do sort of a blind, "Well, okay. Here we go. I think this pattern means this. This is probably this or maybe that for you. This is this. And, oh, that's going on. Okay. We should work on this and maybe that." Usually by the end of the first or second sentence, people are like, "How did you know?" Because I don't do a clinical entry, I'm not a psychologist. I'm not saying, "Let me do a deep history. Let me figure out your relationship with your mother." We don't ever ask you about your mother, unless she wants to train, and then we ask a lot. If she wants to come in and get a brain map, it helps us inform you of our idea about your brain. That's the kind of stuff we want to know about your mother, not so much, anything else. We don't need to.

And so, that layer of neurofeedback as an industry, which is probably at least half of the industry, will do science based, evidence based assessments. Look at your brain, try to figure out what's going on, and then through the use of a combination of things suggested from clinical practice, basically lore, and things suggested by science, which is one step removed from lore, so tested lore, and some times purely suggested by the data. That happens sometimes too. You blend all these things with your sense of how things change, whether you're sensing what people might want to do, and then I go after singular or two or three spots at once and try to work on areas of connectivity. Big sweeps of the brain. Big regulatory markers and push on specific things. One or two or three or four at a time, not 97 this moment. 84 that moment. Six this moment. Three that moment.

And so those are the two flavors if you will of QEEG training. One is z Score driven, they're often multi channel and one is usually less, fewer channels and is typically a little more precise. And either relies on magical abilities and neurofeedback or it relies on deep neuroscience. You can do either way. You can be the most incredibly gifted, intuitive person who has done this for 30 years and knows everything about how brains work, but doesn't know how they know that. There are a few people like that in the field and they're getting amazing results. They mostly do the same thing as the science based people do, which is deep assessments and evidence driven decision making.

And then there's a few technologies that some have elaborated in the past 10 or 15 years. A few of them in the last couple of years. That have less understanding about their mechanism and probably don't have the mechanism that their proponents would expect or suggest that they have. Those include things like it's either infra slow or ultra low. Things below one hertz. The reason for my skepticism is there's something in physics, just the way electromagnetism works, where if you want to measure a frequency, you need to measure it twice as long as the sample you want to measure. It's called the Nyquist theorem. So if you want to measure a theta wave, which is four times a second, you've got to measure at least twice, 250 milliseconds or at least a half a second to get the theta wave. When you're below .1 hertz and we're at like .0001 hertz to get a single iteration of change, and to measure something in like an FFT, a frequency analysis, how much is this stuff changing kind of thing, you need hours, and you're doing sessions that are 20 minutes.

So I have a hunch there's some mechanism that is not well understood and even the proponents of these mechanisms will say things like, "Yeah, the physics breaks down." The lead, if you will, person who is one of the first people in this first ultra low training regiment is also one of the greats in the field who created what everyone else is still mostly using. Sort of traditional band training that I do and many people do. He's a physicist and he says, "Oh yeah. You're right. It can't be training down there, but these results we're getting, these are great results." It's not that they're chasing some weird model, they just happened to find something that works in some interesting way and they're building more resources. I still prefer some evidence based stuff where we can keep going back to assessments and get a sense of how things are changing, which is why I work the way that I do.

Then there's a few technologies that are sort of more one size fits all, which don't work all that well. And there's some technologies that mix or use more micro stimulation. Things like LENS, which is Low Energy Neurofeedback System. A variant of that, a new variant called HPN, High Performance Neurofeedback, I think. And those are sort of micro stimulation of a series of sites done on the head on both sides, for everyone. Short sessions. Not much electricity, about as much as in an old school radioactive watch dial, in terms in the amount of ... it's pico or nano or some incredibly small unit of measure of energy, but it's enough to perturb the system and it produces change. It seems to work really well for a few conditions like traumatic brain injury. And it seems to work less well for a lot of the other things that neurofeedback works really well for, has been my experience. I don't use LENS or HPN because we get pretty good results going after specific things, and I also do it from a neurofeedback.

The style is mostly something called SMR, sensory motor rhythm. This is a magical rhythm in the brain. 12-15 hertz in humans, usually. I have a hunch that it's integrated in a bunch of different ways that make it really useful to hack. One of those things is learning. Sleep regulation is another. Stability of the system is another. So, if you train up SMR, you end up with reduced seizures. The seizure threshold gets dramatically lifted. And research suggests that humans that do this end up having 30%-50% reduction seizures. Often 5%-10% of the people get reduction of more ... Elimination of more than a year of seizures in the literature reviews.

But this was a discovery in the field that sort of started the human neurofeedback field about 50 years ago. 1967, Dr. Barry Sterman, UCLA, was a learning scientist that had worked on cats to see if they could get a milk dropper, maybe a chicken broth dropper, a little thing triggered whenever a brainwave went up. Now, cats make SMR in an obvious way, which I'm guessing is why Dr. Sterman chose this rhythm. You've seen a cat lying in a window sill watching birds, you've seen SMR. That incredibly inhibited body but very alert mind is an SMR state. ADHD is the opposite of that. You're mentorically hyperactive. You're mentally hyperactive. Disinhibited. I view SMRs as a inhibitory tone in the brain. And trimming that up, Sterman found through doing another experiment months later, makes the brain stable.

So Sterman was hired by NASA to essentially figure out how dangerous rocket fuel is because astronauts were complaining of hallucinations and nausea and other things. So Sterman would expose cats to rocket fuel, and found that most of those cats had a really predictable dose curve. Ataxia, crying, drooling, seizure, coma, death. This is the late 1960s. We don't generally kill cats anymore in research, but a few years ago, slightly different landscape. And one group of his cats refused to have seizures, they had exposure times two and a half times as long. So it was like an hour to seizure for most cats, and 240 minutes in the cats who had received some neurofeedback training in another experiment a while back.

The next step was, somebody unaffiliated with the lab had some uncontrolled seizure disorder, lots of seizures, and she was like, "Okay. We're gonna do some of this on me." And this started maybe somewhat [inaudible 00:25:44], but I mean, the data's there, so this actually happened. So she, over time, reduced all her meds and was seizure free for many, many months after doing some training. And then Sterman ended up getting some more research out there over the years on learning and epilepsy. Several more people did SMR research on epilepsy and seizure. Sterman had his funding pulled soon after he submitted his initial findings about seizure control with SMR. Very rapidly, without too much communication. So there's some suspicion in the field that there was some ulterior motive from Pharma to control alternate mechanisms. Because of the time, the late 1960s, the neuroleptic class of major drugs was pretty profitable, you know?

Daniel S.: Mm-hmm (affirmative)

Dr. Andrew Hill: It was before Prozac era, so we didn't have the 800 pound gorilla in the Pharma room then.

From there, from epilepsy, from instability, we discovered that ... Because EEG's a sleep realm as well. When you're cognitive neuroscience, you're one step away from a sleep scientist. Because the history of EEG is a sleep history, sleep studies, sleep science, and so very quickly we discovered, EEG scientists discovered in the 1960s and 1970s that we regulate sleep pretty profoundly and, "Oh, look at that. It sets impulsivity under control suddenly, profoundly quickly." So now it was an ADHD target when that became a known thing in the 1970s and 1980s, and it works incredibly well on things like ADHD now. It's sort of the biggest use.

So neurofeedback, SMR training, [inaudible 00:27:16] feedback will eliminate 80%, 90% of ADHD in a matter of weeks in most people. Usually a long term change, done.

Daniel S.: Okay. So you said ... That was a great intro overview. You said a lot of things. I want to double click on quite a few of them there. I want to start by making sure the primary mechanism of what's happening regarding feedback itself across both peripheral and central nervous system feedback really makes sense to people. To explain it another way, we have sensory systems that sense the outer world and we have sensory systems that sense the inner world. And so, just like we've got the ability to extend our exterior senses, our eyes or our ears, whatever, through telescopes, microscopes, our ability to talk over the phone right now, which is a technological extension of the sensory system into the environment, our proprioceptive systems that are sensing what's going on inside of our body, pretty much everybody feels a neurofeedback when they're learning a new thing like balance. Right?

Dr. Andrew Hill: Yeah. Yeah.

Daniel S.: And they're ... It's also similarly hard to describe. Like when you're trying to do something that's almost impossible, like walk on a slack line. And it's impossible, impossible, and then all of a sudden, it's kind of possible and you can't quite describe exactly what it was, but your system figured out from a form of feedback, falling, how to regulate better.

Dr. Andrew Hill: Exactly, yeah.

Daniel S.: What you're looking at is that all the biofeedback technologies kind of are what a microscope or telescope is to your eyes, to your external senses. These are those kinds of tools to your own internal senses, to actually increase internal capacity so that the system can pay attention to its own states and learn and regulate faster.

Dr. Andrew Hill: Yeah. Any signal you provide to the brain that has information in it, the brain would like to extract that information and learn from it. If it discovers that information is its self or for the body, it finds the information a little more compelling, and it learns faster couples, yokes that information. Once you have that feedback loop, you can shape it by applauding shifts in the data.

Daniel S.: Right. So now one of the most important things that you said is, the brain is this radically complex system. We've got these millions of CPU-like things and how good that analogy is, is even a question. But we have these millions of CPU-like things that are deeply coupled to each other. So when we're trying to do optimization, we typically, we learn about optimization in complicated systems rather than complex systems where the causal dynamics were more obvious, and we're trying to optimize, meaning either increase or decrease, some number of metrics in relationship to each other where we understand the causation better. When we try to optimize a system that we don't understand all that well, there's always a question mark. Are we optimizing for a model that is a subset of what's happening and some of what we don't understand actually potentially getting damaged in the process. Because that's the externality, the side effects.

Dr. Andrew Hill: I mean, that can happen. If you train a [inaudible 00:30:18] protocol that produces the wrong effect, then you can do that. And you ignore the side effects, if you white knuckle it, "Oh, this is good for me. I'm doing the thing I should do." And you keep going, you will permanize the weird side effects. I mean I had ... There was a parent that came into a place I worked years ago who had done 80 sessions of neurofeedback on her autistic son where they canned the protocol in a system she didn't know how to use especially well, and he went from a high functioning autism to regressed, non-verbal under a table. He had been not that autistic and did a lot of work with him and moved him really forward. And then in a month and a half, two months, three months, boom, back to two year old status.

Or you can disrupt your sleep profoundly, or you can make yourself anxious, or you can ... Yeah. It's a pretty powerful tool that if you manipulate in ways that don't ... The model doesn't meet the brain, you can do harm.

Daniel S.: Right. It's like if you're doing posturally corrective exercise and you do exactly the wrong ones for you, then you can move a partial pattern in the wrong direction.

Now, my question though is, in that case, the symptoms were increasing, negative symptoms were increasing and they were white knuckling. I'm curious about where somebody is actually increasing a positive, but it is at the cost of externalizing somewhere else that they might not know, right? Because the brain does a lot of things.

Dr. Andrew Hill: Zero sum kind of stuff or?

Daniel S.: In that possibility.

Dr. Andrew Hill: I don't think neurofeedback has that risk. Not to the same degree as something like TDCS or TMS or one of those zapping technologies. There's some evidence those are zero sum because of how they work.

Daniel S.: I think it's worth saying more because we mostly know about medical interventions that cause side effects because they work on a fairly narrow band of things we want to optimize without understanding the whole complex system and so, other side effects are common and our electrostim things also have that. We may be increasing cognitive performance but maybe decreasing some aspect of top down neurologic control of the immune system. But we wouldn't even fucking know because it's a long term thing.

Dr. Andrew Hill: Yeah, I mean, that seems to be happening with zapping technologies. Where if you push a bit of the cortex hard, you might be able to boost briefly, if you will, in one resource, but you seem to get some limit as the brain shuttles over to react to that in the short term. It's such a profound change.

The neurofeedback to some extent is exercise. And exercise done properly doesn't produce hypertrophy and postural problems and knee issues and a weird neck problem and adrenal fatigue, if that even exists. It means from your shakes from people taking too many pre workouts or whatever. But done right, exercise optimizes the system and it does not ... You really can't do too much variable exercise. If you run all the time, there's a way you're gonna do damage eventually with exercise, you're gonna overtrain, and you can overtrain with neurofeedback. That's a valid analogy. But if you train the right way, physically, you don't get side effects. It's the elimination of side effects, of sleep issues, of stress. The homeo-dynamics, a word that I know you enjoy, improve the regulatory range of the system becomes more flexible and able to re-regulate from the baseline.

Daniel S.: So this is what I was hoping you would speak to because this is what I think is so fascinating about biofeedback and neurofeedback, specifically, is that you're actually helping the system increase its own regulatory capacity and not giving it the stimulation to try and say, "Oh, it's too high. We're gonna exogenously lower it or if something is too low, we're gonna raise it." You're giving it feedback to learn how to raise or lower or adapt itself, which really does much more likely have the propensity to be comprehensively positive and on zero sum.

Dr. Andrew Hill: I mean there are some forms of neurofeedback that are what Roberto Pascual-Marqui, who's a guy that invented source analysis. Loretta, he calls hotspotology. There are some forms of that where you go, "Oh."  And I call it some neofrenology. This pattern means this, therefore let me turn it down. If it's too high, turn it down. If it's too low, turn it up.

A mentor of mine in QEG, a gentleman named Jack Johnstone, he died last year. He would say, "You don't just turn things up and you just don't turn things down." He was also in the same role I am, EEG neuroscience, psychology. He got me thinking a lot more about how this all fits together and much less about, "Ooh, is there a data marker that's off?" Because that's so easy to misinterpret, over interpret. I mean so, yes, we're looking at regulatory capacity and the forms of neurofeedback that I enjoy, that I improve of, that I like to use, are forms that work on global mechanisms initially, at least. SMR re-regulates the whole system at attention, sleep, a few other things, seizure, obviously.

Alpha training or Alpha Theta, which is a special type of training, it involved entering a sort of hypnogogic sort of state, essentially. It is also extremely useful for certain things, including re-regulating over arousal, including causing an immune surge in T-cells. So if you're T-cells are low, there's some evidence that you can take them and hack them up continuously with alpha training, actually. Alpha theta is a near cousin of alpha training. The literature, the published paper by Dr. Schumer, shows alpha training brings up T-cells. I've seen the effect in people doing alpha theta, so I know it's a similar mechanism.

Daniel S.: So I have a question here, when we think about neurofeedback as exercise for the brain and central nervous system and we kind of play with that analogy a bit, someone can do a pretty tremendous amount of curls, increase biceps and not affect lower body at all, right?

Dr. Andrew Hill: Yeah.

Daniel S.: And so, in a similar ... We wouldn't say doing curls is increasing the adaptive capacity of their musculoskeletal system, we'd say it's increasing a fairly narrow capacity.

Dr. Andrew Hill: Yeah. I have two things to say about that. One is, that the neurons and the micro columns are not in biceps, and they're interconnected much more profoundly. And they're [inaudible 00:36:19] electricity, which spreads. And so half the signal that you're measuring at any spot under the electrode is from anywhere, and half the signal is from right under the electrodes. There's some spacial sensitivity but not profound spacial sensitivity for training. So that's one thing, you're getting some global effects when you do band based training in multi electrodes. You aren't unmixing the signals and doing real time source analysis, which you can also do. And it has some benefits, when you understand what you're going after. It's huge to do source analysis training, but in general, you're measuring a signal at a scalp electrode. It's a dirty signal, if you will, in terms of spacial resolution, half of it's from everywhere. So you're getting some global effects, when you do regulatory frequencies.

Two, there's a pretty dramatic and measurable plasticity boost throughout the brain that seems to happen when you do a single session of neurofeedback, and therefore, continuing neuro sessions. So you're getting a BBNF bath when you provide this stimulus the brains like, "Whoa. Interesting information. Let me try to make sense of that." And it appears like BBNF flood is one of the things you get, so plasticity just goes up. Although, motor potentials with TMS were easier after a neurofeedback session than before. That's a function of the amount of BBNFs. It's a proxy measure experiment that's been done with one, single session of neurofeedback by doubling ... dropping the threshold activation of the tissue, which just means more plasticity. It's more interconnected than biceps and it's also a more global effect. It's almost like if you worked your bicep out and all the bradykinins and other inflammatory molecules got released, caused your other muscles to go, "Oh. Let's all tone up." Then it would be a better analogy for neurofeedback.

I do use a bicep analogy. When people ask if it's permanent. I'm like, "Yeah. Yeah. Yeah." But you're not stuck in that position either. You are turned into somebody with strong biceps. You're turned into somebody that when there's heavy stuff to pick up off the floor, discovers they have strength, but it isn't necessarily flexed all the time.

Daniel S.: So maybe rather than biceps versus lower body, we'll look at sprinting versus distance running, which are utilizing the same things in different ways. The reason I'm asking, has to do with what you would call, general capacity to adapt to any useful state versus the ability to get into specific states well. So if someone does alpha training and they go into a specific state, that might be super useful for immune boosting or maybe anything needed through alpha. Obviously we know some people who have excessively high alpha and can't get into beta, is doing alpha training going to necessarily increase someone's capacity to go into beta or theta or any state versus just capacity to go into that state?

Dr. Andrew Hill: It depends on how the brain works. And we think the brain works in ways where certain frequencies, if they're stuck on, inhibit the change. So doing any training that breaks those stuck patterns up, does improve regulatory capacity. There was a study not too long ago that showed the absolute change, training a band in a certain direction, may actually not happen for a lot of people. That's why the QEEG studies in neurofeedback are a little weak. It's because the absolute change across a group of people before and after a handful of sessions isn't dramatic or even all that significant. It happens for a few of them, but if you measure a complexity in the data, you can look at functional change happening within session, what you discover is the range, the ability of people to change states while doing a session is increasing. So the exact thing you're asking has been demonstrated mathematically that there's increased functional range in the EEG, even if the absolute band power isn't changing before or after a session always, in all people.

Daniel S.: So this is an interesting question if we ask what is the equivalent of crossfit as opposed to sprinting or distance running for methods of neurofeedback? If our goal is to increase adaptive capacity for any state based on whatever relevant stimuli, how would you think of that?

Dr. Andrew Hill: It's fully the way that I work. I mean, that's the short answer. Now, I mean, SMR, alpha-theta, alpha, beta, even broadband past training does fit that category because in alpha-theta training you improve that ability to hit that liminal state between consciousness and unconsciousness, hypnogognia to some extent. It breeds to other things as well, like insight. Like re-regulation of arousal, it's hitting a bunch of things. It's actually where the relaxation response, I think, must come from. Hypnogogia sort of in between state states. It's a transition state. We hold you in transition state, you get better transitioning.

So you're unable ... This is why alpha-theta is used for Creativity. Creativity I would argue, big C Creativity, one of these things that takes holding vast amounts of possibilities in your mind, which is regulatory capacity of ideas at least, Creativity.

Alpha-Theta because of training ability to do hypnogogic liminal shift, produces increased access to that generative wellspring. So I'd argue that started feedback. SMR, making the brain more metastable against the seizure threshold. Improving the transition states, awake to asleep, asleep to awake. Improving a few of those other things and used in those mechanisms. I think you can pick three or four styles in your feedback and do almost anything with any of them. But if you do SMR, alpha-theta, beta, alpha training, hit the big bands and suppress data usually, this used to be called the author method before they invented the infra slow method below one hertz. But it's the old author method, it's band training where you say, "No. No. Theta's inhibitory. You make a lot of theta, you don't inhibit. Let's tamp that down." And damaged tissue might be theta or delta with your eyes open so you can tamp it down. Alpha if it's present everywhere might make everything stuck, you need to get rid of that.

Alpha-theta tension like you went into earlier, if you're stuck in alpha you're kind of like, "Huh? What?" Your timing's slow. You process information. You bind information slowly. Your word finding may be off when your brain ... When your alpha slows down with aging, that's one of the causes of normal cortical aging. That's word finding problems, you hit 57, 60, 62, 65 and you can't find the word and it gets bad, that's usually the speed of processing. So your alpha speed index is that.

So this is the QEEG tool, is getting a big but vague picture of what's going on and then going, "Oh. Okay. Well here's your problem right here." Kind of like your mechanic will. You have to trust the neurofeedback person the QEEG person, like you do your mechanic. They can sell you anything in terms of looking at your complex systems. But it's also not pure supposition, it's not pure clinical lore. So we start with the evidence, generate a hypothesis. Now the hypothesis subject to experience, which is weak, missing clinical practice, and then refine those hypotheses as we get effects from training. And so we steer people better and better towards the facts. I think that's better than holding onto a model how this works, because it's ironclad. Because they're all imperfect and when you describe the core beliefs of all the schools of thought of neurofeedback, you find they're in conflict and they can't all be true. And yet, most schools of neurofeedback get relatively good results.

Daniel S.: So let's talk about why you say most schools of neurofeedback get relatively good results and yet most neurologists aren't going to say that neurofeedback is a particularly meaningful thing, and Wikipedia has a mostly negative slant on it. Can you?

Dr. Andrew Hill: Sure.

Daniel S.: Obviously when you talk about how complex it is, how many metrics might be affected or not affected but, yes, speak about that a little bit?

Dr. Andrew Hill: Yeah. I mean the question this is unfortunately this is asking is, "Why isn't there more gold standard blah, blah, blah in placebo controls, high end clinical data, clinical research studies published?" Because that is the decision criteria for the neurologist, for the psychiatrist. Is, "Well, it's not in PubMed at a high level, therefore it doesn't exist." So it's a truism, but I think a valid one that research validation of clinical technique lags by a significant amount of time. It's also true that there are thousands of papers on neurofeedback. Now, there's not a thousand good papers, there's like one or two or three decent papers. And that's the real question, is, "Why are there not a lot ... Why are there not the same kind of data backstopping? Why is the game not played at that level by the neurofeedback interventionists the same way it is played by the biochemical interventionists?" Because it seems like there's a game to be played, if you prove this, you get CPT codes for your insurance billing. You prove validity, efficacy, specificity and you are off to the races because the insurance system is the way to go.

So there's a few assumptions in there that we don't have to challenge to point out that's how it's been and neurofeedback has not met that particular criteria and there's some really good reasons for that, if you understand neurofeedback a little bit better. One of those reasons, the most valid, the most real, is probably that it's extremely difficult to do double blind EEG analysis. In real time, you need a signal that needs to stay clean when the person moves, when the software does weird things. As a technician, you're watching stuff. You could single blind it. You could hide the signal screen from the client and just have a game screen. That would work. Not very good science. Double blind.

So up until about 2012 or 2010, you couldn't do double blind research. There was lots of ways that people worked around it but there it was always some other physiological signal that was contaminating the person's EEG. It was still them. You were still doing something that wasn't double blind. And then Howard Lightstone, who's the Chief Engineer Lead behind the primary software in the field called, EEGER, E-E-G-E-R. It used to be called Neuro Cybernetics and it's now EEGER. Howard recognized this problem. There was lots of clinical use, 5,000, 10,000 people using it, but nobody actually doing good research because of this blinding problem.

So I helped Howard a little bit in terms of flushing ideas, but he ended up building this beautiful ... It's called the Study Manager, and what it does in real time is it grabs several segments of prerecorded EEG off the disk. So you do those yourself, you make them clean. It takes the live data off the person's head and looks at the power envelope. It takes the segments and shuffles them. Stitches them together. Scales them to the person's EEG. Blends them and then generates training metrics, training data, feedback off the storage segments only and generates screen display off the blend of the EEG. So if the person coughs, or moves, or unplugs a wire, you get what you expect and it looks like it's a real EEG. But the actual amplitude, spacing and band things bare no mathematical relationship pretty much to the person's real EEG. None, actually, because the training parameters are not coming from the blended signal. They're coming from static segments that are stitched together and scaled.

Daniel S.: Mm-hmm (affirmative). That was a tricky thing to figure out how to put together.

Dr. Andrew Hill: Yeah, so Howard did this, so now there's five multi-centered studies going on in ... Using EEGER and it's still becoming a bit more of a research tool. That being said, in 2012, because of the preponderance of the research in things like ADHD, in 2012, the American Academy of Pediatricians raised neurofeedback back up to Level I - Best Support. So a table of what's an intervention for pediatric problems, and this is the table for ADHD, and the things in Level I before that, were psycho stimulants. The things in Level I now, psycho stimulants, neurofeedback. Now this research publication, American Academy of Pediatricians, which is a review board of doctors, essentially, pediatricians, is not an insurance company. It's not the insurance machine. And so, that wasn't enough to get it covered by ... very well. It's very poorly covered by insurance. 5% of people might get something back, rarely. Diagnostic EEG is sometimes a little bit more, but QEEG is not exactly that. So usually you don't get coverage from the insurance end of the pool.

There have been 5, or 10, or 15 really good studies since 2012 across different problems and different complaints. Things that cost money, things like pain, things like ADHD, but nothing profound yet. This gets to the other reason why their research is lagging. Because to really get coverage in this current way the game is played, you need FDA clinical trials, Phase clinical trials. Here's where it breaks down. Because the money to produce the clinical trials in the way the game is played now, is in the $1 million to $5 million range for a small study. And, being neurofeedback, this is more expensive than a drug study because you can't just give people their own drugs to take and do the study. Well you have to actually do it. So you're talking about 20 sessions, 30 sessions, 40 sessions, 50 sessions, very expensive way to do it. And, "Oh, no." Neurofeedback is a heavily individualized process when it's done the best.

So now you need 1,000, 2,000, 5,000, 10,000 end study of participants. Now we're talking about a $5 million to more like $20 million study. So while there are some academics doing studies of the 40 people, 50 people, there's no one doing big Phase clinical trials because, where's the money coming from? The equation breaks down. Pharma pays for its drugs to go through this process. They invest $5 million or $10 million in research and $5 million or $10 million in studies and they have 10 of those products that they're developing, and three of them become drugs that are sold. They make $50 million each. The math mostly works out. Great. But there's no money coming in, in the tens of millions in neurofeedback research. And the people doing the neurofeedback research who are getting the best results, are mom-and-pop clinicians. Usually psychologists, sometimes social workers or family therapists or nurses or neuroscientists, and some are MDs and some are psychiatrists but they're probably only a fraction of a percent of the people doing this work.

And so it's this gray area, that is not medicine, because it's not only done by MDs, and it's not psychology because you don't have to have a clinical psychology license in almost every state to do this. And it's not sleep EEG, your polysomnography, and it's not epileptology, but it's still EEG. And it's not really personal training because you kind of have to know neuroscience and a pretty heavy duty tech skill set to do this well. But it's closer to personal training in terms of how it's carried out. A quarter of it can be personal training with your very attentive, very intuitive but very well-educated technician helping assess, identify problems, dial in possible solutions, working you through them, evaluating them and fine tuning if necessary.

So for me the natural evolution of the field is into personal training, not more and more medical. And a lot of the clinicians, half the people out there are clinicians or more, they want this to be fully covered by insurance, and at the same time most of them are dumping insurance and running the other direction. So it's a very scattered field of 5,000 people in the U.S., let's say, and it's not very cohesive and most of them are small clinicians and there's no money in doing research but they're getting great results.

Daniel S.: So we have talked about this in the context of functionalized medicine, personalized medicine, the same way where we're not looking at does a drug have some effect beyond placebo across some large number of randomized population, but as a personalized medicine protocol, which is gonna be different for everyone, have some meaningful effects on some pattern of lots and lots of metrics. And this, now, is something that requires scientific process beyond clinical trial.

Dr. Andrew Hill: Yeah and lose that to machine learning and big data, and we're in that space suddenly. This is the biggest data you can imagine because it's EEG. EEG is bigger data than anything. It's bigger than image data. It's bigger than MRIs. It's the most chaotic. It is the most dense data set. The reason my PhD took me a couple of years, or many years, is because I had 20 or 30 terabytes of data that I had to process through to get my answers. The joke goes that a physicist will spend a year planning an experiment, do it for an hour and have an answer. An EEG scientist will do an experiment for an hour and analyze the data for a year and a half. It's this incredibly dense space that requires phenomenal reduction, but reduction that actually is to some model that is meaningful, so you can learn things and use the data in some [inaudible 00:53:31] break it's reality part of the individual, that's a problem.

Daniel S.: Okay, so if there are people who want to learn more beyond what we talked about here to really have a sense that the epistemology is good even though it doesn't look like double blind placebo controlled trials yet for all the reasons that you mentioned. We've only been able to do double blind trials for a little bit of time and it's personalized effects and what metric do you look at and et cetera. What would be resources to learn more for someone to really have a sense ... Because you're saying things like, "80% of people with ADHD have positive results." Et cetera. Where would people go to learn more?

Dr. Andrew Hill: I mean, there's of course stuff on PubMed.gov and you can search there and you can find things there under neurofeedback. EEG and SMR, those two keywords, produce a great deal of information that is backstopping a lot of what I'm describing as the history of the field in the past 50 years. You can find some things there actually, quite a few good papers. But you can also go to the ISNR website. That's the International Society for Neurofeedback and Research. They're sort of the trade organization for the neurofeedback world. They have been maintaining a bibliography of all of the research that's available across complaint. Categorized by complaint. So you can look it up for Tourette's or ADHD or autism or seizures or migraines, peak performance or age or anything else you want to look at.

We also have a website, the Peak Brain Institute website have what we think is a good sampling of some studies. We've highlighted that talk about it for anxiety, or pain, or ADHD or things like that. But it's also ... I encourage people to think about this, again, it's closer to exercise. Yes, the research is there for exercise because it's a little easier to do research when doing prolonged exercise for longer than neurofeedback one would argue, at least technologically assisted neurofeedback. But exercise is the sort of thing that you do and pretty quickly you go, "Oh. Oh, okay. Here's what's happening. This actually works." And there's no placebo effect when you're sleeping better and you're much stronger and you've dropped 20 pounds of adipose and your pants fit different. The same can be said of neurofeedback. For most forms of neurofeedback the effects you get show up initially, within the first couple of weeks. And so, it's not a phenomenally huge investment to do a month of neurofeedback, 10 or 12 sessions or something, and go halfway into it, "Wait a minute. Oh, this is what this process in neurofeedback is doing."

Now you do sort of have to pick your neurofeedback technology a little bit, and I have some bias here. I'll fully admit that. But I do think using QEEG is the threshold for adequate neurofeedback preparation. Oh, this person records a Q and uses it to understand my brain. Okay. They're at least not operating off in la la land fully yet. It takes time and effort and energy and money. It's very expensive to learn QEEG. The machine's expensive. The databases are tens of thousands of dollars. The knowledge to process and interpret QEEG in reporting, metrics, takes year. So that's so expensive if you're gonna be somebody that settles for woo, who settles for purely magical thinking and maps of the world and models of the world there are little resemblance how things actually work, if you settle easily for those things, you generally don't invest heavily in new onsets of skills science.

And so, if you're somebody who's really dug into QEEG and you're doing recordings and interpreting them and using them to guide your practice, even if you started off believing neurofeedback was magic fairy dust, you've learned it isn't. Because you've learned so much about the brain, in spite of yourself that you now do evidence-based neurofeedback. It's such a high investment that becomes this sort of weed out factor in some ways, I think. Because if you're like, "No. No. I have a bunch of great protocols. They always work and I never look at the person's brain." Well, you may think that they always work, but I know that your efficacy is lower than it could be and your side effects are higher than they can be at the very, very least. The person may get great results. You have a great trainer. But if you're picking somebody randomly, pick somebody who probably uses the best tools and QEEG, at least as an assessment device, is a big clue towards that.

So I would say, you could look up PubMed, you can go to isnr.org and look at their bibliography, Peak Brain Institute and look at our selected articles. But honestly, you can find a practitioner and just do some training. One thing about training and biofeedback or neurofeedback, is that this is again learning, not some special type of intervention. Like a drug, drugs often work best. Chemical interventions often work best when they're profoundly well targeted. Chemotherapy is now becoming targeted to cancer, a genetic sample of your cancer, and get engineered chemo that only attacks the genetic signature of your cancer. That's incredibly exciting. That's like beautiful typesetting of science. But many earlier drugs for mood and psychiatric stuff are big giant shotguns, where they have hit stuff and, "Oh, we often get a benefit. That's kind of interesting." But they're not clean in their mechanism, not precise. So you can do some of those things with neurofeedback and just see what happens because there are often very early and quick effects. Or you can dig into research and decide for yourself that there's enough there.

We typically have people do three months of training initially because it takes time. That's another reason why neurofeedback is not profoundly well adopted. Why don't we have incredible cord muscle strength, abs, profoundly great cardiovascular range and toning capacity, and perfect blood lipids, perfect hormones. Why don't we have those? Because we almost know everything we need to do, to get those things in range. Oh, wait a minute. It takes dropping all sugar and going and doing exercise, maybe it's three or four times a week for at least two hours a week. Wait a minute. It takes some effort? That's why neurofeedback isn't well adopted. Everyone finds out what it can does and a month in when they're feeling like all of the limits they thought were permanent are no longer permanent, are lifting, everyone asks me, "Why doesn't everyone know about this? Why isn't it everywhere? Because this is amazing and I need to do this." Then answer is simply, "Why don't we all have great cardiovascular and core strength, great abs, whatever? Because it takes time and you have to go throughout."

Three times a week is what we do, maybe for three months, four months to get a permanent change for many things. But it's this process of going and working with a trainer and getting some exercise. You go work with a personal trainer, you or I, for three months, three times a week working out strictly and getting some support around how we lived our life, some bio-hacking support. "Oh, you probably shouldn't eat Krispy Kreme Donuts more than once a week." Some broad strokes like that. Those things will change you profoundly in three months and the abs will show up if you're getting the right kind of support. That's what neurofeedback is. It's not analogous to adderall or ritalin or modafinil. Or even things like piroxicam and other nootropics. It's not in the same category. It's much closer to, "Well are you gonna tune your system up and exercise it and change how it works, or are you gonna just react to the stressors, the motivators, the demands, the things that are coming in and out of your body?" Anyways-

Daniel S.: Which means that it has the capacity for much broader, more comprehensively positive and more enduring positive effects and takes more input. Now I would imagine, you might not have formal data on this but you might have an intuition or a clinical sense, why don't we all have good core strength and abs? Some of the behavior changes are not only like they take energy but they take actually going against some fairly life long addictive difficult patterns. So what is your experience with if someone is actually doing SMR training first and they're increasing their capacity for impulse control that then translates to not just the benefits of SMR but increases the capacity to do other things like exercise.

Dr. Andrew Hill: Absolutely, and it does. I mean, if somebody comes in with "alcoholism" and they need two things. They need SMR training and alpha theta training, at the very least, they may need more than that. The reason they need alpha theta is because they need to re-regulate the brain's inability to produce GABA or Alpha, essentially, which has been suppressed through exogenous drinking for many years. So the brain stops producing it, becomes stuck in this beta heavy state, so we regulate without the theta and we do the SMR, not really to work on the alcoholism directly but maybe to work on the impulsivity the anxiety the difficulty with sustaining stillness with your emotions, et cetera. All the inhibitory tone stuff.

So I would argue that instead of thinking about it like a specific problem, you think about it like, again, resources. Where SMR is inhibitory tone and if you have good inhibitory tone, yeah, it means you aren't gonna be ADHD, but maybe you weren't to start with. Maybe you're only impulsive around that extra yummy learning thing that happened when you got alcohol reinforced for a bunch of times. Maybe you have that gene that makes you find alcohol even extra rewarding, so you learn faster. That's a process that happened and SMR will give you just a touch more inhibitory tone and make you more successful controlling behavior.

If OCD or PTSD like problems and you're anterosingular or posterosingular, it's stuck in a beta mode and spinning your mental wheels and burning cycles and being really inefficient and monitoring the environment constantly. Well, you can slow that down by tamping those betas down, replacing them with a little bit more alpha. So none of those models about how the brain works, are accurate mapping out the territory here in the slightest, but they're useful knowing that there's an over arousal syndrome that involves gaba and probably alpha. You can work on it with alpha theta neurofeedback successfully most of the time.

Peniston, who developed alpha theta, the Peniston protocols, discovered that in a population of alcoholics, you could reverse recidivism. Move it from 3/4 or 2/3 down to 1/4 or 1/3 in terms of relapse rates and it's reversed. And Doug Quirk did the same thing with alpha theta in prison populations in Canada. 7,000 prisoners reverse recidivism. It went from 75% recidivism to 25% after alpha theta neurofeedback. And then David Kaiser and other people probably [inaudible 01:04:11] was involved, did SMR training interspersed with alpha theta and the Peniston protocols, the first alpha theta protocols had visualization audio tracks along with the actual biofeedback. And a couple of guys came later and just did the SMR interspersed with alpha theta and got the same kind of behavioral re-regulation.

So it's really a question of finding models that are roughly accurate more than necessarily valid.

Daniel S.: So this is interesting, when people read Jim McGonigle's book on the willpower instinct and just the sheer amount of data of impulse control related to quality of life metrics everywhere, right, starting with the two marshmallows experiments. There's a question I think a lot of people have regarding all the things that they want to invest their time in, which ones should they do first? It increases the capacity to do others. Because they can't really increase their learning in specific areas and make their finances better and make their health better, and their relationships, and optimally at the same time. So a lot of people say, "Let me learn how to speed read so then I can read other books better." Except speed reading mostly sucks.

So this is actually an interesting kind of proposal. If someone invested in some deep neurofeedback work early on, increased capacity to actually do anything else and stabilize those things, is a pretty interesting argument for putting it low on the stack.

Dr. Andrew Hill: Yeah and this is why about 1/3 of our clients at Peak are peak performance clients. No pun intended there, but that's the reason why we picked the name. Because, yes, I have ADHD people and people with autism and cerebral palsy and brain injuries and migraines and anxiety and PTSD, all kinds of "diagnoses", but about 1/3 of my clients are people who are incredibly high performers who are trying to remain or gain from there, or elders who've decided they don't want to experience slow down or maybe experiencing non-pathological changes and they want shore them up. But I would say about half and half from the young, high powered bio-hackers CEO, Ben Greenfield, whatever, let me squeeze out the extra 5% or 10% or 15%. And those guys train forever and train for a few months intensely and then drop intensity and keep going long term. Some of them on their own and some of them at our centers. And some of the elders will just shore up the slight drop in processing speed and get the word finding back and get the sleep re-regulated, back on track, slow down the aging trajectory, probably, and they're good to go.

Many of my clients move from a fix to a fitness mentality over time anyways. Or they'll have ADHD and nothing works in my life because I'm profoundly ADHD. I'm that kid in the family. I'm the one who can't do anything right. Who can't get out of the house? Who's fallen out of school?  Who's made all the bad decisions, apparently? That kid comes in, that adult kid, as an ADHD person with profound ADHD, you've got a couple of those. We fix the ADHD and all of a sudden they're profoundly capacitive in their ability to take control of their life and they end up becoming the person who's like organizing the family dinner and keeping everyone calm and communicating well during the chaotic environments and things like that. Is this treatment? Maybe if I was a psychologist, but otherwise it's something different.

Daniel S.: So I want to go through a bunch of questions fairly quickly that I think might be on peoples' minds. You've mentioned it but let's categorize the categories of things for which this is interesting. Because you've already mentioned cognitive things, you've mentioned psycho emotional things, you've mentioned physiologic things like immune on the negative side-

Dr. Andrew Hill: Pain.

Daniel S.: ...obviously performance on the positive side. And you haven't mentioned it but many people have kind of spiritual states, right? Numinous states that they ... So meditation training. And we have both decreasing dysfunction and then increasing function or capacity. And just kind of speak to what each of those categories, some of the main things that we've found or-

Dr. Andrew Hill: So I often look at where my practice is and who comes in. It's flexion bias, but I end up with a third roughly in each category. And I roughly categorize it as psychology, neurology and peak performance. Neurology includes things like autism and other developmental disorders, and seizures often, and migraines, and brain injuries, all of that. And then psychology is the biggest bucket including ... ADHD is probably neurology, as much as anything else. But then, psychology or sleep things often and anxiety ... I'm not saying they're in the mind, I still think they're in the brain. But they're a function of brain learning to be in that state because of internal/external demands that made it learn that way. Depression fits in that category. It's learning. All of this is learning. Addiction. This is all learning.

Then in the peak performance are things like processing speed, anti aging stuff and vigilance and creativity for alpha theta. Pain fits in that category somewhere. But I'm not sure if it's in the neurology or psychology space. It might be in both. It might be in peak performance. A hardcore athlete who's only sore because they workout like a Ben Greenfield every day for four hours or whatever he's doing these days. Or a Tim Ferris, who's like throwing himself off buildings now and climbing and doing crazy stuff all the time. These guys are probably in pain a lot because they're hard charging individuals. If the pain is in your body, it's probably also in your mind a little bit. So we can do psychology stuff for pain, which is actually the same rough area according to Leon Eisenberger and others.

On the right front there's some emotional pain and physical pain centers, and if you train them down and you have chronic pain syndromes, they drop. And if you train them down in a different way and you have chronic emotional pain syndromes from attachment problems or deep heartbreak or PTSD or whatever else is going on, we can tamp those down. So it's not a question of what we can work on, it's almost more of a question of what we can't work on.

Daniel S.: Sure.

Dr. Andrew Hill: The list becomes small. [inaudible 01:10:31] you can work on it very, very well with neurofeedback. Attention, sleep, stress, mood, attention switching, which covers things like OCD and Tourette's. Those are both sort of switching problems, if you will. Seizure, migraine, PTSD, rack and attachment disorder. You can work on things even like bipolar disorder and schizophrenia. I generally don't because, A, I'm not a psychologist or a psychiatrist. B, because especially schizophrenia, you're talking about something where there's an active disease process almost fighting against you. If you stop accommodating, it stops working. And so it's almost ... It's harder to get a permanent change. Much, much harder with an active disease process.

I also don't generally work on severe Alzheimer/dementia kind of thing because once you've gotten that far, you've lost so much of the scaffolding of the system, the tissue, that you probably would lose the ability to learn and remember. I mean, when people ask about Alzheimer I point them to other things like Dale [inaudible 01:11:33] men program, which can apparently reverse Alzheimer by controlling metabolic factors. But once you're very, very, very, profoundly impaired like that, it may not be the best solution. That being said, you're profoundly brain injured, having lots and lots of seizures, it's exactly the right solution. So regulatory systems. If the machinery for them is shot like in Alzheimer, you can't do much, but otherwise, you can do almost anything.

Daniel S.: Yeah, I mean it sounds like neurofeedback is kind of the top down regulatory systems, like genomics work is to the bottom up systems. I mean there's very little in the psyche or the physiology that's not effective if not profoundly regulated by the top down system.

So someone could come in to see you for anxiety, but they also happen to have sleep issues and they also happen to have some immune issues. So you just factor everything going on is how do we help their top down systems increase regulatory capacity. I'm curious.

Dr. Andrew Hill: Yeah, I do factor some of them in. Other ones I don't factor in, I just take approaches that are broad.

Daniel S.: Mm-hmm (affirmative). So I'm interested in the interface of comprehensive neurofeedback trainings like you're doing, that might include alpha theta training, SMR, et cetera. With other forms of comprehensive training like functional medicine, like the Peniston protocols, if someone is sitting with some kind of neuro degen, like good methods of psychotherapy. So the intersection of either psychotherapy neurofeedback or physiologic work in neurofeedback depending upon people's need, and how much has that been actively explored?

Dr. Andrew Hill: I mean, it's explored only in that often health and optimization and wellness professionals will do more than one thing and will encourage more than one thing. I mean, we do a lot of mindfulness meditation in all the centers [inaudible 01:13:26]. Feedback we always offer meditation or mindfulness, because we find that when people do both, both work better. My Peak Brain Culver City office shares a wall with a great Ashtanga yoga studio and so some of our clients will do ashtanga yoga six days a week in the morning and then they'll come in and do neurofeedback right after. Very aggressive if you will, physical training. I've also had clients that do physical therapy sort of hand in glove with neurofeedback. And for the exercise clients, the clients that do physical therapy, when you add the neurofeedback stuff, the physical stuff starts to change faster. So physical therapy that wasn't really working that well suddenly unfolds, that's probably plasticity.

Other clients with traumatic brain injuries are getting better but still have ... TBIs you often have sort of the two step forward, one step back or three step forward, one step back kind of effect where it's taking longer to make change, working the brain harder too only to get it to stick in traumatic brain injuries. It's not quite as simple as ADHD or anxiety.

And so as we're training these guys, sometimes I'll say, "Okay, why don't you come next door and start working out like hardcore for a month or two?" And a week or two into doing yoga and then coming doing neurofeedback, the gains that we're getting earlier on are back and moving faster. It's at least partially doing all these down dogs with your head down and driving blood into the brain, it must be.

But it's also true if people workout at their gym and then come in and train. Or vice versa, do neurofeedback almost as a pre-workout. Get all pumped up. Get focused. Get plasticity factors going and then hit the gym hard, you find things change. But these are all subjective experiences that are pretty ungraded, unmeasured. But I hear it again, and again, and again, so I encourage people to add whatever other interventions that are behavioral to their life, be it speech and language for a kid, or tutoring or language learning or yoga, meditation, whatever it is. Whatever modifiable behavior, behavioral hack you can do to increase skill learning, change in the right direction of the brain, plus neurofeedback [inaudible 01:15:37] because everything seems to work better. I mean, this is not necessarily ... It's at least partially because of plasticity and neurofeedback, plasticity enhancers.

But it's also sort of the science of human transformation. When you add more intervention strategies, when you hit a certain magical threshold, it's like for human hacking intervention type things, the chance of having a discontinue list change. The linear becomes almost certain. You add one or two things you get change. At three, you get a lot of change. You add four, you're gonna get some profound, amazing thing happening.

And so I'm a big fan of people adding lots of interventions when they're trying to make change, but I actually tell people usually not to take nootropics when they're doing neurofeedback for the first couple of months. Because, at least, I don't want them to start nootropics. If they're on them, fine. If they have them dialed in, great. But lots of people want to start and they want to get everything done.

Daniel S.: Sure.

Dr. Andrew Hill: I'm like, "Great. Well, I'm happy to suggest some nootropic strategies for you to dig into those things. And even like your sleep hacking and your diet and exercise and everything else, those are all important things, but why not change anything else right now and you can just do some neurofeedback for a month and let's see where we can shift your baselines. And then we'll take it up another notch with nootropics and whatever else." And we usually dial in lifestyle factors, so we're not purely hacking the brain, we're also doing a little bit of light hacking for people. Usually supporting that vision for people.

Daniel S.: So a few last questions for people to have a sense how to move forward. Home units. Obviously a lot of people are using units right now and most of these units are dry leads and not that many leads.

Dr. Andrew Hill: And the wrong locations for training, most of them.

Daniel S.: Are there any home units that are good? Are there any that are worth using? Are they problematic? What is your sense on that? Would someone be better to just do HRV training?

Dr. Andrew Hill: My sense is, yeah, you're better to do HRV training. Is probably true if you wanted some biofeedback at home. The EEG, neurofeedback systems that are home consumer based, have many problems like you've described and those are all big problems and true problems. Wrong location. Bad signal quality. Too few connections. Yada, yada, yada. The bigger problem is that neurofeedback is not one thing and that you might not want to do one thing for one person. You might not want to stay with what you're doing once you get a certain [inaudible 01:18:00]. And so, learning how to do a neurofeedback protocol is not rocket science. Learning how to do neurofeedback ... Even if the tool you bought for $100 off the shelf did what it said it did, and none of them do, you still wouldn't just want to do that thing.

Daniel S.: Now, if someone is coming and they're training with you and they want to do some ongoing training on their own in between, are there some home units that people can use that are meaningful once they've already got some training and maybe with check-ins?

Dr. Andrew Hill: Yeah, and we do that. Many practitioners do that. There are devices that are at the intersection price and complexity wise of clinical and pre-consumer level uses. But they're not consumer pricing. We put together a system at a pretty "low cost" and it's pushing $4,000. If it includes HEG, Hemoencephalography, which is infrared blood flow treating for migraines. It's about a $4,500 system and that's just the hardware. We've got to spend a few days with you, working with you to do an assessments and teach you to set up systems and set up your protocols and then we send you home. And for three months we do weekly check-ins, chart reviews, support, like you're describing. The bio-hacker can get a lot of mileage that way. And the motivated mom or dad who has got an ADHD or autistic kid can get a lot of mileage that way. The profoundly anxious or ADHD person trying to train themselves cannot.

If your thing is a thing that gets in the way so much that you can't be productive and you can't do things without a certain amount of doing them and brazening through and figuring out some computers and some technology, then it's not ideal. But, yeah, we do a lot of that for our clients who want to start in our office and then want to be come home trainers, or clients that are only able to come to one of our offices for a few days. In three days we can do QEEG and do two practice neurofeedback sessions with you a day and teach you how to log sessions and do that kind of thing. We do a lot of that.

And of course, we also open more offices all the time. So we're trying to open up in lots of cities as well. But our goals are a little different than the average neurofeedback clinician is gonna work specifically on more of a therapeutic psychological medical perspective. We're trying to make this like the brain gym or the brain spa. And teaching someone to use this sophisticated equipment and get a good training program going fits that methodology even if they're not working out in our center. So we're happy to extend the knowledge management piece of it to clients that are self-directed, computer savvy and want to hack away at their own brains.

Daniel S.: So one of the goals of Qualia is being able to identify, I mean, kind of like the first part of neurofeedback, being able to separate signal from noise, amplify the signal and then synthesize the signal in the whole space of what's actually effective for people. And so we want to be able to help identify and curate resources that people can use to actually reliably help themselves. I really love the approach you're taking of being able to get well-trained people who aren't trying to do one thing, but are trying to actually help increase an active capacity of the top down systems across a lot of goals done well and increasing, having centers in more places I think is a hugely important thing. Is there anything that could help you with this mission?

Dr. Andrew Hill: Sorry, I just lost you there. I'm back.

Daniel S.: Where'd you lose me? We'll edit that out.

Dr. Andrew Hill: You really like how I'm doing it, getting well-trained people.

Daniel S.: So, coming back in, I really like the approach of not just doing one therapeutic method, but really helping practitioners understand more complexly how to be model agnostic, be well trained in the models, have good equipment and actually be able to do good comprehensive neurofeedback with people and be able to interface other things like whether it's yoga or other methods for traumatic brain injury and so I think training at home is a difficult enough thing as you mentioned, if someone's already got very difficult things going on. But there's a reason yoga studios exist, is that it's just hard for people to do for any reason, right? For many people to do for any reason.

So I'm delighted that you're getting more centers in more places, getting more people trained, getting better software systems so people can do higher quality work at the level they're training. Is there anything that you could use to help advance what you're doing? Like if there's a city that you're not in yet and people are interested in helping it get to that place or?

Dr. Andrew Hill: Yeah. That's a big question. I mean, we have a pretty aggressive goal. I keep praying for growth. Right now we're in four cities. We're in San Diego, Downtown Los Angeles, Culver City Los Angeles, and St. Louis. Then we have some concierge technicians who live in Portland, Oregon, the Bay Area. Our goal is to open up 50 centers internationally. In 2018, the goal is to open another five to ten throughout the U.S. and then in 2019, to open another 20-40 throughout the world.

We're trying to become a big company but we're still a small start up. We're less than two years old. We've been running for 21 or 22 months. I've got my books finally sorted out and saw that we made $8,000 for the company, but I've also put a couple hundred grand into growth. We're not flat revenue in that way. We still made some money but we're putting all of our resources into growth. And some point, we're gonna have to break out of that, me as a mad scientist trying to take over the world and save the world from a neurofeedback perspective, into a large multi national company. I'm not quite sure how to get there. We're trying to build data tools first. And that's probably the thing I need the most help with is, I need to find myself a good CTO to manage that piece of it. But maybe just a good UX/UI Jenga full stack developer right now, that would be great. If there's one listening who wants to get involved with some good bio-hacking.

Daniel S.: So whether it was someone that could help with the data science, the software, the UX, the UI, real estate or fundraising, they can contact you?

Dr. Andrew Hill: They can contact me. And we have some pretty big ideas and I think we're well positioned. I mean I think there are only two companies in the neurofeedback space that are best positioned for big national presence and one of them is Peak Brain. Because we've got four centers and because I happen to be kind of old in the field. I've done it for a while. I know how it's done well and I know what mistakes not to make. And the other one is actually Betsy DeVos's Neurocore, because she's the Secretary of Education, and her [inaudible 01:25:00] company has a pretty heavy investment in Neurocore, which is a network of neurofeedback centers in Michigan and Florida. They've announced they're gonna double their footprint.

So from my perspective, if there's governmental level, high level individuals who are pushing this and they're trying to become big players in this space and I'm trying to take the little solo clinician and make it the equinox or Gold's Gym of the brain, those are different goals. An education [inaudible 01:25:20] center, which is what her things are. Or Peak Brain, which is more peak fitness. I think at some point we're gonna have enough societal pressure that will be these massive organizations. People will demand this stuff when they realize what it can do. But there was a time before there was a Gold's Gym on every corner and I think that's where we are for the brain. No one quite realizes they want to look like, or feel like or be like, they can be.

And access to technology is what I'm trying to solve. I'm not sure yet how to solve the piece about awareness. Full brand, full marketing, if you will, not from a point of view. More guy kawasaki evangelism than true marketing in this case, but I'm always happy to talk about neurofeedback. I never care. I mean, I care a little bit. My CLO would kill me if I said this, but I never care if someone trains with me. For me it's an evangelism thing where, "Oh, my God. This stuff exists." You don't have to be satisfied with your brain. Shift happens. Take control. It's gonna happen anyways. The brain's whole damn job is to shift. Why don't you decide what direction it's gonna move in. This can do that. Find somebody. Let me help you find somebody. Come see us.

I'm pretty passionate about this idea that you can do it. And you can do it with a psychologist, or a personal trainer, or a psychiatrist, or an autistic specialist, ABA therapist that does neurofeedback or something. It doesn't really matter. You're losing out if you're sitting on your butt and not doing neurofeedback and watching TV or something. You might as well sit on your butt and do neurofeedback.

Daniel S.: So make sure they do QEEG first and other than that are there any other guidelines of what to look for, what not to look for if someone wants to train with someone and it's safe or not?

Dr. Andrew Hill: There are a couple of keywords that are usually red flags, and those include words like quantum and detox. Generally in the health space when those words are invoked, they're invoked as obvious percipatory marketing language. Non-linear. Did I say that one? Non-linear is a big one. It's an important one. So if the reason they tell you it works is language you don't understand and you ask them what the words mean they've used, they can't explain them, then you're probably not working with somebody you should keep working with. And if they don't say, "I don't know." As often as they say, "I do know." That's also a problem in this space. You know when somebody has all the answers, you want somebody who has all the tools and is gonna help deobtiscate or demystify the neuroscience for you. You don't want a magic practitioner of wizardry. You want a very skilled mechanic or personal trainer who can break it down for you and give you a sense of what your options are. Not take control.

Not like in the 1950s, which we used to do with medicine, which is, "Oh, okay. Whatever you say. I'm just gonna do that." This is your brain. It's never the brain of the person working on you. You should be making these decisions so the more they include you in the, "Here's what your data means. Here's what we're doing. Here's what we're trying. Where do you want to go?" The better you should feel about the process.

Daniel S.: So QEEG and basic critical thinking of if they are doing a realistic scientifically informed job.

Dr. Andrew Hill: Yeah. And also cost. I mean this stuff is often very expensive. The average person charges like a psychologist, about $150 an hour. And so cost should not be the reason you do this, even though it's quite variable. You'll pay between $5,000 and $10,000 for a full course of neurofeedback. But if that's $10,000 spent towards woo woo hardware that doesn't do anything and has broken interface and a technician is using all one size fits all stuff, it's a waste of $10,000.

But if it's $5,000 or $10,000 that eliminates your ADHD, fixes your sleep problem, limits your seizures, improves your productive output, deepens your sleep, makes you Mr. Creative, it'd be cheap at $50,000 or $100,000. Or it lifts your kid's ADHD or it lifts your kid's autism symptoms by 10% or 20%, that's priceless to people with problems. So it's not a very price sensitive field, and it's also a niche with about 5,000 practitioners in the U.S. at most, which keeps the pricing relatively high.

So people have to navigate these things, unfortunately, when deciding who to work with and how to work. Hopefully they can get a sense that there's a lot of different technologies out there and there's a little bit of a difference between them.

So starting with QEEG, at the very least, partitions that space towards reasonable practice.

Daniel S.: So, we'll see what kind of questions come up with this podcast. There's a bunch of follow questions regarding FMRI neurofeedback and HEG and the ponds unit that's happening at the University of Wisconsin and intersection with transcranial stem and all kinds of fun things.

But this was a fantastic introduction to the space and the field.

Dr. Andrew Hill: Oh, good.

Daniel S.: I really appreciate you coming on and I really appreciate both the work you're doing in the space and the work that your come from is, you don't really care who they're working with, you care they're getting good work. This is our exactly come from too and we-

Dr. Andrew Hill: Yeah. I want to change the world. All the adderall, all of the ritalin, all of the methamphetamine, all of the poorly working psycho active drugs we use for 90% of things, sleep drugs, anxiety drugs. My goal is to eliminate those for everyone, and I can't do neurofeedback on every one, and even if I open 50 centers, I can't work in every one. So I'm trying to effect a sea change that I happen to sail on.

Daniel S.: Well, we share that mission and are coming at it from different directions, but I love the direction you're coming at it through. And if we can be of support ongoingly, we will. Thank you for this today.

Dr. Andrew Hill: My pleasure, Daniel. Thank you.

Daniel S.: Thanks, Andrew.

If we find a product or service we love, we want to share that with our community.  In some of these cases we will partner with the provider in an affiliate relationship which may result in a payment or benefit to Qualia.  We won't ever enter into such an arrangement or recommend any product or service we haven't researched or stand behind.

 All content provided on this website is for informational purposes only. This information is never intended to be a substitute for a doctor-patient relationship nor does it constitute medical advice of any kind.