This is adapted from one of my weekly Neurofeedback & Chill livestreams, where I usually run practical neurofeedback and walk through a biohacking topic. This session I focused on photobiomodulation, or red light therapy, and demonstrated a brain-form-factor device live. What follows is the teaching content cleaned up for reading, with the audience Q&A folded in anonymously.
I have watched red light therapy come into the consumer biohacking scene for more than a decade. It arrived as a bit of a wild west: helmets, panels, individual spot lights, lasers, red light, green light, near-infrared. Plenty of anecdotal benefit, not a lot of clean technique. I started working with it seriously about a year ago, mostly because one device let me program light by quadrant and pulse, which mapped onto how I already think about brains from a QEEG.
What is photobiomodulation, and how does light affect brain tissue?
Photobiomodulation is exogenous stimulation with light. Photons hit your tissue and drive cellular changes. Two mechanisms carry most of the effect.
The first is mitochondrial. In roughly the 600 to 800 nanometer range, light energy activates cytochrome c oxidase, an enzyme in the mitochondrial electron transport chain (Hamblin, 2016). You get a boost in ATP production. Remember high school biology: adenosine diphosphate takes on another phosphate to become adenosine triphosphate, a coiled spring of stored energy that cells then burn. Red light helps the mitochondria load that spring.
The second is a broader cascade of secondary signaling. Photobiomodulation triggers nitric oxide release, shifts in cellular calcium, and changes in ion channels, especially as you move up into the faster near-infrared wavelengths (Hamblin, 2016). There is also good evidence that mitochondria coordinate across cells, organizing their energy patterns across tissues. They are far more central to life and energy than the textbook "co-opted bacteria" story suggests. When you stimulate with light, you are nudging the cell's primary energy machinery.
Which wavelengths reach the brain, and which only reach skin?
Wavelength determines penetration depth, and penetration depth determines whether you are treating skin or brain.
Red light at 630 to 660 nanometers reaches about a centimeter or less. That is fine for surface wounds, skin, anti-aging, and possibly hair, but it barely touches brain tissue. Several papers show that once you account for absorption through the scalp and skull, 600-range light does not get meaningfully into the brain (Jagdeo et al., 2012). If you see panels you hang on a wall, or red light hair-regrowth helmets, they are usually in this lower range. Good for the body. Insufficient for cognition.
Cross into near-infrared around 800 nanometers and penetration jumps to roughly three to four centimeters. The cortex is a scrunched-up sheet of paper, heavily folded but not deep as a whole structure. Three to four centimeters of penetration is enough to reach it (Jagdeo et al., 2012). As you climb toward 1070 to 1090 nanometers, the light penetrates deeper and broader still, because infrared scatters and gets absorbed less, and reflects less heat. That is why the serious brain photobiomodulation devices target the near-infrared band rather than visible red.
The general rule: slower, lower-nanometer light handles skin, tissue repair, and surface healing. Faster, higher-nanometer near-infrared reaches the brain. Laser light, the original low-level laser therapy, likely produces effects equivalent to near-infrared LEDs because coherent light penetrates better with less scatter and absorption. The advantage is in penetration, not in some special property of the laser.
What does brain photobiomodulation actually help with?
This is a young field, so read the studies carefully. Methodology and definitions vary even more than they do in neurofeedback, and pooling messy studies produces messy conclusions.
The reported effects cluster around memory and recall, focus and attention, and a general sense of mental clarity; people describe running "smoother." There is a recovery phenomenon, both physical and cognitive, that seems boosted by red light. Studies show up across cardiac surgery recovery, general aging, and athletic recovery. The plausible through-line is blood flow and energy metabolism. Neurovascular coupling, the matching of blood supply to neural demand, is implicated in several diseases of aging and cognition, and a lot of post-viral and post-concussion problems have a metabolic component.
One wavelength deserves a separate mention. Stimulation at 40 Hz, in the gamma range, has animal-model evidence for clearing amyloid plaques and tau tangles (Iaccarino et al., 2016), and there is human work suggesting an amyloid-clearing effect as well. That is why a "focus" style protocol often runs a few minutes of 40 Hz flicker.
A note on entrainment: light flicker can produce a modest entrainment effect in neurons, and the literature supports that. Auditory entrainment, binaural beats and that category, does not produce reliable brain change. I do not use those and I would not lean on them.
How do I use red light therapy with neurofeedback?
I rarely use photobiomodulation as a standalone training. I use it the way I use a metabolic support like hyperbaric oxygen (HBOT) or hyperventilation training: prime or feed the system, then let the EEG training shape the change.
Photobiomodulation is outside-in stimulation, a metabolic primer rather than a training process. You bring up the energy with light, then neurofeedback steers the direction. My working theory is that the light lubricates and feeds the brain's processes, and the neurofeedback then lands more smoothly and more powerfully because the brain has more fuel to work with. In concussion, brain fog, post-viral fatigue, old mold exposures, or quiet post-Lyme states without active infection, adding metabolic support like HBOT or HEG to EEG neurofeedback seems to add to the impact and may help the change stick faster, though this is my coaching observation rather than a controlled finding. Photobiomodulation seems to operate in a similar lane, though I cannot say it matches the magnitude of HBOT.
Where this matters most is a brain that is genuinely struggling. A concussed brain may take a long time to change unless you bring all the tools. Metabolic support, in my experience, can turn slow work into faster progress.
What does a starting protocol look like?
This is not medical advice; it is how I structure it in my Peak Brain programs.
I start most people on a solid red glow, ten minutes a day, three days in a row. Then a couple of days off. Then five days in a row of the same. If you tolerate that without feeling over-trained or taxed, you are reasonably safe to start pushing into other protocols without an over-response.
The device I demonstrated runs four presets I use most:
- Glow is solid red light, the least taxing option. Many people feel little; some feel calm or focused. A good place to begin.
- Peace is a 10 Hz flicker, in the alpha band. The intent is to push an alpha effect and, in a brain carrying excess beta, to help break that beta up.
- Focus is roughly six minutes of 40 Hz gamma stimulation, the amyloid-clearing wavelength.
- Energize is a short beta protocol, activating.
Once you have a couple of weeks under you, twice a day, three to five days a week is reasonable. I often run a priming protocol before another intervention and a recovery protocol at the end of the day or after something taxing.
What side effects should you watch for?
Photobiomodulation is easier to overdo than neurofeedback. Neurofeedback side effects tend to be subtle; you adjust and they fade. Red light overtraining is a fatigue event.
Some adjustment effects are expected and fine: brief tiredness that wears off, slightly altered sleep that normalizes in a couple of days, a short-lived "up and on" feeling. Those are metabolic-boost signatures.
Stop the session if you notice dizziness, headache, fatigue, irritability, anger or agitation, or a stuffy, "cold coming on" congestion. That congestion is my best guess at a glymphatic or lymphatic drainage response as the body reacts to the metabolic movement. If these show up within a couple of hours after a session, note them and reduce dose before the next one.
When I ran a glow protocol live on myself, the effect was very subtle. My head flushed visibly red, clear evidence of blood flow change, like a gentle workout. I felt slightly calmer and slightly slower in processing, an energy effect rather than better inhibition.
When should you go gently, and when should you push?
If you carry a lot of metabolic load, heavy Lyme history, co-infections, deep illness-related fatigue, move slowly. The brain is remarkably plastic and can take a lot, but when the metabolism is inflamed and on a hair trigger, it is easy to push too hard and trigger adverse responses with any brain-targeting biohack.
Aggressive brain biohacking belongs weeks after recovery, once the brain is stable and has resources. During an acute event you target what is actually happening, case by case, rather than trying to drive change.
How do you decide what your brain actually needs?
I am hot on the brain map for a reason. There are 30 to 40 fairly discrete circuits that tend to dysregulate and show up on the EEG. These patterns sometimes diverge from tissue anatomy, but they consistently reflect different failure modes in different regions, which is part of how I think about EEG phenotypes.
Failure modes are region-specific. Left frontal areas tend to fail in the slow bands, excess theta and alpha, with delta marking tissue problems. The right frontal corner fails more in beta and less in alpha. The anterior cingulate and front midline can fail in slows, alphas, and betas, each meaning something different. So the decision is region-specific. If you have low alpha in a region, you probably do not want to add metabolic boosting. If you have a big beta blob on the front midline and you add HEG, you may discover exactly how OCD you can be, which is a cortico-striatal pattern lighting up. If you have a low-and-slow frontal variant, metabolic support can be excellent.
The peace protocol made sense on the map I showed live, a brain with excessive alpha and beta running two and a half to three standard deviations high, heavy front-midline excess. A 10 Hz push can help break up that beta. With light I think you can stay a bit more linear than with neurofeedback, using alpha to break up beta or to bring up low alpha power.
This is why I will not start neurofeedback without a map. I am not a clinician and I am not running a therapeutic transference container; I am acting as a trainer and coach, helping you understand your brain, generate ideas, and iterate. Without a clinical relationship I rely more heavily on the QEEG, the executive-function testing, and your day-to-day responses on sleep, stress, and mood to know how a protocol is landing. Look at your raw data. Look at your maps. Then do things to your brain.
How HEG and HBOT compare, and how to stack modalities
A few questions came up about related tools.
HEG (hemoencephalography) is blood-flow biofeedback. The passive infrared version reads summed heat coming off the brain, a gross metabolic signal. Dr. Jeff Carmen's near-infrared HEG instead sends in light that interacts with oxygen and measures the change in red-light level as a proxy for local oxygenation, a more focal signal that also mildly stimulates and tends to cause more headaches. Passive infrared HEG is my preferred blood-flow tool. It is striking for classic migraine with aura: a migraine is essentially spreading cortical depression, a metabolic cramp with a blood flow collapse, and pumping the vasculature with HEG over sessions appears to restore perfusion and stabilize metabolic tone, which lines up with reports of reduced severity and then incidence (Carmen, 2004).
For stacking, the research on human transformation from the 1970s and 80s holds up: piling on too many changes kills compliance, but there is a sweet spot. Combining a couple of modalities of regulatory change tends to make durable transformation more likely than any single change alone, and adding a third often produces a larger jump. I pick three core areas of regulatory change with each person, mixing brain goals with behavioral ones like sleep, substance use, or a morning practice.
One ordering rule: with most modalities, dive last. I do not have people do HBOT before neurofeedback in the same day. HBOT hyper-normalizes the brain. Excess delta and theta, slowed alpha, and brain fog all drop sharply on a QEEG for hours after a chamber. Train in that state and two things can go wrong: the brain has far more flexibility and energy than it is used to and can swing too hard, and any technique that needs you to see an outlier (z-score training, for example) loses the outlier you were trying to train. I have watched the overreaction firsthand, a strong positive shift that then exhausts and cramps in a place the brain is not used to being. The known exception is infra-slow and infra-low neurofeedback, which is deeply metabolic and works fine, even better, after HBOT.
Is the brain device worth it?
The brain-targeted near-infrared helmets run a few thousand dollars, and a question came up about cheaper alternatives. I have looked, and I have not found a flexible panel I would trust for the brain. Studies from competing manufacturers show that moving red light away from the head or changing the angle produces a large drop in absorption and penetration. A form-fitting device that captures and contains the light, in the 1070 to 1080 nanometer range, is what gets light into brain tissue.
A lower-cost 810-nanometer panel can be useful for general body work, anti-inflammatory and anti-aging, and there is a follow-on benefit for the brain whenever you heal or exercise the body. Weightlifting helps the brain. For the brain specifically, the form factor and wavelength matter.
For most of the people I work with, I can imagine a use case for metabolic support like photobiomodulation, HEG, or HBOT in maybe half of them. The real need, the acute metabolic struggle, is more like 10 to 20 percent. Those are the people climbing into chambers and strapping on these devices, and they are the ones showing accelerated change.
Where to start
Start with your data. If you are near one of our offices in New York City, Los Angeles, Orange County, or St. Louis, or you do a fully remote program, the brain-mapping special gives you a year of unlimited maps and consults so you can track changes from any intervention you stack. Once you understand what your brain is doing, the question stops being "what is the protocol" and becomes "what is my brain doing, and what do I want to gently push." That narrows your options and eliminates a large portion of what could go wrong before you ever turn a device on.
I wrote a companion article on brain biohacking with photobiomodulation if you want the longer-form version, and I plan a deeper review once the literature settles. For now: get a map, start gentle with glow, watch for the fatigue and congestion side effects, and use the light to feed the system before you train it.