In this episode, we explore: ADHD and Medication Efficacy π Neurofeedback for Brain Performance π The Cost of Mental Health Treatments π° Brain Mapping: Why It Matters π§ Ethical Concerns and Charlatans in Neurofeedback π¨ π Donβt miss this engaging conversation full of clinical expertise and real-world applications! πΊ Watch here: https://youtu.be/RJT5jDVJtnI?si=ELRjDcsC48hiPoM3 Key Moments: 0:00 β Pre-show Tech Troubles 0:30 β Introduction & Disclaimer 1:49 β Dr. Hill on ADHD Research π 3:34 β Why Neurofeedback Research Lags π§ 5:50 β Challenges in Neurofeedback Studies 10:00 β ADHD, Medication, and Parity Debate 11:48 β Conditions That Neurofeedback Helps π©Ί 17:05 β Measuring Progress and Brain Changes π 23:25 β ADHD Medication & Neurofeedback Synergy 28:00 β Brain Mapping and Personalized Treatment π§ 35:00 β The Role of Tools and Techniques in Neurofeedback 40:02 β Medication vs. Neurofeedback for ADHD 46:12 β Psilocybin and Brain Plasticity π 51:00 β Long-Term Medication Use and Neurofeedback Effects 55:00 β ADHD, Stress, and Brain Regulation Discussion 1:00:00 β Santiago's Upcoming Events π #MentalHealth #Neurofeedback #OpenChat #BrainHealth #QandA #MentalHealthAwareness #NeuroNoodle #JayGunkelman #DrMariSwingle #JoshuaMoore #AndrewHill #JoyLunt #AnthonyRamos #JohnMekrut #LiveQandA #Neurofeedback #ADHD #MentalHealth #BrainMapping #Neurotherapy #ADHDTreatment #Neuroplasticity #MentalHealthAwareness #Psilocybin #BrainHealth #NeurofeedbackTherapy #BrainScience
Episode Summary
I joined the panel on NeuroNoodle for a live Q&A about neurofeedback, ADHD, brain mapping, and where the evidence actually stands. This piece pulls together my own answers from that conversation. You can watch the original conversation for the full discussion with the rest of the panel.
What follows is my view as a cognitive neuroscientist and brain-training coach, drawn from 14 years at Peak Brain Institute and more than 25,000 QEEG brain maps. I will be clear throughout about what is well-established research, what is my own observation, and what is extrapolation.
Why does the neurofeedback research lag behind what practitioners see?
The published literature on neurofeedback trails what experienced practitioners observe by decades. There are five sane reasons for that gap.
First, applied neurofeedback is individualized. You assess, you start somewhere, and you iterate toward the person's goals as their priorities and symptoms shift. What you do for one person looks different from what you do for the next. Two athletes can walk into the gym with the same goal and two different trainers will get them there by different routes. Almost all of the research, by contrast, runs the same protocol across a group. A few studies individualize alpha frequency, but most apply fixed protocols to small samples. That design does not reflect neurofeedback as practiced. It reflects something closer to the cutting edge of 1978.
Second, the work is slow. Real, measurable change at the individual level often takes 30, 40, 50, sometimes 100 sessions. Group designs running short protocols break down against that timeline.
Third, money. A tier-one drug trial costs about five million dollars, and that is a relatively simple study with a single compound, blinding, placebo control, and active shams. Nobody owns neurofeedback. The field is a set of niche hardware and software providers with little historical pressure to fund that kind of trial.
Fourth, blinding and sham control are genuinely hard with neurofeedback. I ran one of the first sham-controlled studies for my dissertation around 2009 and 2010. For a technology that is 100 years old as EEG and 55 to 60 years old as neurofeedback, our first blinded studies are about 15 years old. That is recent.
Fifth, there was active opposition. In the 1970s and 80s, Sterman's seizure work drew pushback (Sterman & Friar, 1972). Later, medication companies sent physicians to ADHD conferences specifically to discredit neurofeedback.
None of that means the technique is weak. I started in this field before there was efficacy proof of anything. If neurofeedback did not show real power in the research, I would have moved on to something else. Meta-analytic work on ADHD neurofeedback has shown effects on attention and impulsivity that approach what medication produces (Arns et al., 2009). If you want the deeper case, see my neuroscientist's guide to neurofeedback for ADHD and the research overview on whether neurofeedback is legitimate.
What conditions can neurofeedback actually help?
Research shows neurofeedback has utility across brain-based regulation problems. Studies find it can reduce seizure frequency (Sterman & Egner, 2006). The research shows it can improve attention. It can improve affective regulation. The research on neurofeedback for anxiety shows real signal.
I will say plainly: neurofeedback is not in the business of curing anything. What I am confident about, based on the research and on what the maps show, is that neurofeedback can have an impact on most brain-based presentations. For rare conditions where there is little or no research, I tell people directly that it is experimental and I do not know to what degree it will help. Neurofeedback is a skill-acquisition training platform, not a drug that triggers hives, so the downside is usually low. With many seemingly intractable presentations, I have seen results that surprised me.
How many neurofeedback sessions will I need?
This is the first question almost everyone asks, and there is no honest number I can give before I have looked at your data. Your psychiatrist cannot tell you how many appointments you will need. Your orthodontist cannot give you an exact date. The 20-session or 40-session figure people read online is a myth applied to a population, not to a brain.
Here is what I tell people. For a garden-variety, limited attention presentation, expect somewhere around 40 sessions to reach meaningful improvement. After that, if you want the gains to stick, we titrate away and confirm the brain can hold the new capacity on its own.
I take numerical measurements every session across every frequency band. Each visit I check whether the brain retained the change and whether the gains held while we stretched the spacing between sessions. When the numbers stay good and someone can go two weeks between visits, I am confident the gains will still be there years later. Structural MRI work, including Ghaziri and colleagues in 2013, has shown that neurofeedback can produce measurable gray and white matter changes (Ghaziri et al., 2013), which fits what I see in the durability of these gains.
Why does the diagnosis matter less than the brain map?
Parents often want a clean number tied to a label: ADHD is a five, autism is a two, this one trains easier than that one. The problem is that EEG and QEEG are not diagnostically specific. We are not going to claim efficacy matched to a DSM category, and we do not need to.
I work below the level of the diagnostic label. Instead of the DSM bucket, I talk about the brain's regulatory resources: attention, the stress response, rumination, perseveration, impulsivity, processing speed, sleep onset and maintenance. Those resources are valid, you can see them in the data, and they cut across diagnostic labels. Phenotypes are buckets too, just buckets that map onto brain function rather than behavioral checklists (Johnstone, Gunkelman & Lunt, 2005). You can read more in my piece on biohacking with EEG phenotypes and in the QEEG brain mapping guide.
I care less about diagnosis than about prognosis. On average, the research and my own tracking suggest about a standard deviation of change on the QEEG, on performance testing, and on whatever we train every 20 to 25 sessions. That is a conservative estimate. With significant brain injury, major illness, schizophrenia, nutritional deficits, or developmental factors, change can run at half that speed.
A brain map also shifts the conversation. When someone says "I have depression," I ask what depression means for them. Plenty of people feel sad. I want to see how it is showing up in the brain, not how a person compares to the population. I have seen maps from people with depression who show epileptiform discharges in the EEG despite never having had a seizure, and those discharges may feed their symptoms. In those cases the conversation goes to the prescribing professional, who may suggest an anticonvulsant. An anticonvulsant for depression sounds backward until you look at the map. That is the point of mapping: it tells you how this brain is producing the experience.
Price versus cost: how should families think about money?
There are two numbers. The price is what you pay for the service. The cost is what you pay for doing nothing.
Neurofeedback is often not cheap, and it is predominantly out-of-pocket. Many insurers classify it as investigational or not medically necessary, and Medicare reclassified it in 2024 from experimental to not medically necessary while still not covering it. But untreated brain regulation problems carry hard, documented costs. ADHD in particular is associated with lost income, higher rates of divorce, and higher rates of accidents and other adverse outcomes (Barkley et al., 2008). Research puts the global economic burden of mental health disorders in the trillions of dollars. For an individual, untreated dysregulation can mean losing a large share of lifetime earning potential.
The flip side matters too. This is not only a disease model. I work with traders, golfers, and other high performers. Give someone a slightly less reactive mode and more sustained focus, and they stop chasing losses and make better decisions. For more on the performance side, see biohacking flow state and biohacking intelligence. On the practical money question, I have written about neurofeedback cost in 2026 and whether neurofeedback is covered by insurance.
Which is the best neurofeedback method: surface, z-score, or LORETA?
There is no single best form, and anyone who tells you otherwise does not have the data to back it. We have essentially no head-to-head research answering this question. It is an active conversation among experienced practitioners, and we do not yet have answers.
Think of it like carpentry. Carpentry works. The best tool depends on what you are building. A good carpenter carries a full belt and is skilled with each tool. A skilled practitioner works well with whatever tools are in front of them, the way a strong musician makes a cheap guitar sound good. When you choose a provider, ask about their experience with your specific concern, what tools they use, and who trained and supervised them.
One word on credentials. A clinical license or a board certification is not required to do good neurofeedback, and I do not think it should be. No research links specific neurofeedback credentials to better outcomes, and some of the best practitioners built their skill through years of practice rather than a certification pathway. Credentials are at most a baseline-training signal, never a guarantee. What actually predicts good work: the practitioner genuinely understands the technology and the neuroscience, they individualize from QEEG data instead of running one-size-fits-all protocols, they track outcomes with objective measures and adjust, they re-map every 20 to 25 sessions, and they are honest about non-response, which runs 15 to 30 percent and is normal.
One real caution: there are people selling things branded as neurofeedback that are not neurofeedback. A headband and a marketing claim is not the same as real training and validated equipment. Ask hard questions about the technology and the person's experience before you commit.
Which ADHD medication matches which brain pattern?
ADHD medication does work for some people. The honest problem is that medication is usually prescribed to a diagnostic category rather than matched to the brain pattern it actually fits. When you look at the EEG, the matching question gets much more specific. What follows is mechanistic reasoning about what the research shows, not a prescribing recommendation, and any medication decision belongs with a prescriber who examines the data.
A few of the patterns the research and the maps point to:
- Methylphenidate (Ritalin) is a dopamine reuptake inhibitor. It tends to work when there is fronto-central theta.
- Amphetamines (Adderall, Vyvanse) act differently. They do not target that fronto-central theta well. They tend to fit slow posterior alpha, where the system needs an agonist push.
- Atomoxetine (Strattera), a reuptake inhibitor, can fit milder slowing where you do not need a strong agonist.
- Modafinil (Provigil) fits a primary disorder of vigilance, people who simply cannot stay awake, closer to narcolepsy without cataplexy. That is a different problem from needing a stimulant. The first off-label work on modafinil in ADHD showed higher dropout, largely from its histaminergic side effects. That is what happens when you medicate the category instead of the right subset.
- Beta spindles point toward guanfacine or clonidine, or for a more severe spindle, something like gabapentin, which edges into anticonvulsant territory.
- Epileptiform discharges appear in a meaningful share of the ADHD population and may call for an anticonvulsant. That same fact means you have to be careful handing out antipsychotics based on ADHD behavior alone, since several antipsychotics lower the seizure threshold.
- Frontal alpha suggests an antidepressant response: an SSRI for standard frontal alpha, an SNRI or a tetracyclic depending on whether that alpha is slow or fast.
The principle is simple. Psychiatry treats an organ it rarely examines. If it looked at the EEG, the research suggests we would get better outcomes, fewer side effects, and less difficult withdrawal. Find the pattern, match the compound to the pattern.
Does ADHD medication stop working over time?
There is a real limit here that families should know about. Many of these medications have a useful window of about five to ten years. A child started at seven can reach college and find the medication no longer doing much, having never learned the underlying skills in the meantime. As the brain stops registering the drug as novel, people tend to need higher doses and collect more side effects.
Neurofeedback offers a different path because it teaches a skill the brain keeps. Vince Monastra's school study is the model here: children with high theta-beta ratios diagnosed with ADHD were all given methylphenidate, and half were randomly assigned to also receive neurofeedback. When the medication was withdrawn, the medication-only group reverted to their ADHD behavior, and the neurofeedback group held their gains (Monastra, Monastra & George, 2002).
I am not against short-term medication. If a child will be expelled in ten days unless their behavior changes, the responsible move is to get medication on board through a pediatrician now, then build the durable skill with neurofeedback and titrate the medication down carefully later. Long-term, for ADHD specifically, the goal is for the trained brain to take over.
A serious warning on stopping medication: do not do it abruptly. During neurofeedback, people sometimes feel so much better that they stop their antidepressant on their own. SSRI withdrawal can be severe, and stopping cold turkey often drives people right back onto the medication because they cannot tolerate the discontinuation. Any taper belongs with the prescribing physician.
Can neurofeedback give you an off switch that medication can't?
I work with a man who credits his success as an attorney to his methylphenidate. He may be right that it helps his focus. He has also paid for it. His relationships are in poor shape because he cannot disconnect from work into life. Methylphenidate keeps you on. It does not give you a switch.
Neurofeedback can train that switch. Flow involves being able to turn high focus on and off when you need it. That capacity for self-regulation is a skill the medication does not teach.
Is psilocybin good for the brain? What about plasticity?
It depends entirely on what you are asking for. The dose makes the poison.
If you ignore everything else, psilocybin has positive effects on many brains, though not all. People with epileptiform content in their EEG, or large beta spindles, warrant caution with entheogens. The likelihood of a bad outcome with psilocybin, ketamine, DMT, or ayahuasca rises with that same epileptiform content and those beta spindles. Those are the two features that most predict medication response, so it is no surprise they also predict bad reactions to psychedelics.
There is a deeper assumption worth correcting. People treat plasticity as something you should maximize. You are making plenty of new connections and reorganizing them every day already. The task is to shape plasticity, not to flood the system with more of it. A lot of my biohacker clients arrive with side effects from doing every plasticity hack at once. I have seen people overdo lion's mane, drive BDNF sky-high, and end up with blunted affect, because plasticity raised without shaping lets the system reorganize into a less functional state.
You see something similar in elite performers. A pianist who has played four hours a day for her whole life can hit a rough stretch and lose the skill, because unshaped plasticity can disrupt a deeply learned pattern. Plasticity is rarely the binding constraint unless you are dealing with something acute.
For most people, psilocybin's better use is the occasional intentional session, not a daily microdose to boost plasticity. If you want to shape your brain reliably, meditation, a walk, or neurofeedback will do it without sourcing risky psychedelics every day. My pieces on biohacking plasticity and biohacking meditation go deeper on safer ways to do this.
How do I actually decide between medication and neurofeedback?
Frame it the way you would frame building muscle. Weightlifting takes time and earns the result. Steroids are faster and carry a different risk profile down the road. Some people genuinely need medication, and for some it is life-saving. Many others reach for it because neurofeedback takes time and sustained work, and the medication looks like the shortcut.
Both can be reasonable. The medication may stop serving you in a few years and may demand higher doses and more side effects over time. Neurofeedback builds a skill that stays. For many families the strongest plan uses both in sequence: stabilize with medication if the situation is urgent, build the durable regulation with training, then taper carefully under a prescriber.
What I'm really training when people reach out
About 80 percent of my work is remote, and most people find me because of a label: ADHD, seizures, anxiety, trauma. When we talk, I reframe the work into brain resources rather than diagnostics. There are six to eight regulatory features I keep returning to: attention rather than ADHD, the stress response rather than anxiety, and within those the perseverative, ruminative, social, sensory, and hypervigilant patterns, plus sleep onset and maintenance.
When someone describes trauma, I validate the experience and the physiology. On the map I might see the posterior cingulate lit up in beta, the brain's lifeguard running in high gear because the cost of missing danger felt so high that it learned the world was not predictable. I am not telling them anything about whether they have PTSD. I am declining to get locked inside the tight DSM frame, and I am pointing at a resource that can be trained.
Working at the level of resources, with a brain map open, supports agency. People stop thinking in symptoms and start reporting day-to-day progress toward their own goals. The same eight regulatory features keep showing up across labels, which is exactly why mapping the brain matters more than naming the disorder. If you want a starting point, look at the QEEG brain mapping guide and the overview of remote neurofeedback.
References
- Sterman (1972). Suppression of seizures in an epileptic following sensorimotor EEG feedback training. doi:10.1016/0013-4694(72)90028-4
- Arns (2009). Efficacy of Neurofeedback Treatment in ADHD: The Effects on Inattention, Impulsivity and Hyperactivity: A Meta-Analysis. doi:10.1177/155005940904000311
- Ghaziri (2013). Neurofeedback Training Induces Changes in White and Gray Matter. doi:10.1177/1550059413476031
- Johnstone (2005). Clinical Database Development: Characterization of EEG Phenotypes. doi:10.1177/155005940503600209
- Barkley (2008). Challenges in diagnosing adults with ADHD. doi:10.4088/jcp.1208e36