This piece comes from my Monday night livestream, where I run a live neurofeedback session and then teach through a topic with audience questions. This week the topic was insomnia. I have anonymized the audience questions and kept the substance.
Why does your brain fight you when you want to sleep?
You lie down exhausted. Then your brain ramps up. It starts replaying a conversation from the day, or a song loops, or you cycle through tomorrow's to-do list. You are tired and your mind will not downshift.
That experience is real, and the cause is not a character flaw or poor discipline. Sleep is an active state the brain has to build and protect. When the circuits that build and protect it are underpowered or pointed the wrong way, sleep does not arrive cleanly, and it does not hold.
Insomnia covers multiple distinct mechanisms. Two people can describe the exact same complaint, hard to fall asleep, hard to stay asleep, waking up tired, and have completely different causes underneath. Before any intervention helps, you want to know which mechanism you are working with.
What is SMR and why does it matter for sleep?
SMR stands for sensorimotor rhythm. It is a low-frequency beta spindle, usually 12 to 15 Hz, that shows up on the vertex of the head at Cz, the central zenith. I trained mine slightly under that range during the session, around 11.5 to 14.5 Hz, because undershooting your native SMR tends to produce a stronger physical relaxation, a body-level calm.
If you have watched a cat on a windowsill tracking a bird, you have seen SMR in action. That liquid stillness with laser focus is a state of motor inhibition. Predators rest deeply while staying locked in because you can act fastest from deep relaxation. SMR is the brain's tone for physical stillness, for holding attention steady, and for falling asleep.
SMR and sleep spindles are the same thing. Neurologists also call this Sigma. When you are awake, this thalamocortical activity lets your body sit still and your mind settle. When you are asleep, the identical circuit fires as a sleep spindle. A vertex sharp wave registers a piece of outside sensory information, and the spindle that follows is the brain noticing the stimulus and refusing to wake to it. That is how you stay asleep through a car door or a creaking house.
When SMR tone is poor, you do not cycle through sleep stages cleanly. The architecture gets ragged. There is also evidence that sleep spindles help trigger memory consolidation during sleep, which is why SMR training tends to improve both nighttime sleep and daytime focus. The same thalamocortical loops do both jobs. I have written a full article on the mechanism if you want the waveform detail: see SMR Neurofeedback: Train Sleep, Focus, and Self-Control and Biohacking Sleep.
The evidence here sits at high confidence for the SMR-spindle equivalence and moderate confidence for the training effect. The Hoedlmoser work showed that roughly 10 sessions of 12 to 15 Hz SMR training at Cz enhanced spindle activity and declarative memory (Hoedlmoser et al., 2008).
How does neurofeedback actually train sleep?
During the livestream I set up a one-channel SMR protocol at Cz. The montage matters, so write it out fully when you take notes: Cz minus A1, with A1 (the left ear) as reference and the right ear as ground. A neurofeedback amplifier is bipolar. It creates the signal by subtracting one location from another, so Cz-A1 and Cz-A2 are different protocols measuring different signals. Do not write it as a division with a slash. It is a subtraction.
The protocol rewarded SMR (11.5 to 14.5 Hz at Cz) and inhibited two bands: theta at 4 to 7 Hz and a faster beta at 20 to 32 Hz. Reward means the game and the tones keep running when my brain spends about half a second producing more SMR and less theta and less high beta. Inhibit means the feedback dims or stops when those unwanted bands climb.
You do not do this with conscious effort. Someone on the stream asked how I could talk and train at the same time. The answer is that neurofeedback works through operant conditioning below awareness. My mind was not driving the feedback. My brain gradually learned, through thousands of tiny half-second rewards, to favor the target pattern. Neurofeedback does not feel like much the first few sessions. After that, each protocol leaves a subtle after-effect, and you can read how a protocol is landing by how your body feels afterward. The fuller mechanism is in Is Neurofeedback Legitimate? A Research Overview.
Cz SMR is body-based. I layer it into many sleep approaches, many anxiety approaches, and sometimes pain and stress work.
Why do sleep meds and many neurofeedback studies fail?
A review paper came out in the fall doing a replication and concluding that neurofeedback does not help insomnia. The problem is the protocol they replicated. Both the original study and the replication ran SMR on the left side at C3, the left precentral gyrus.
Left-side SMR maintains vigilance and focus rather than supporting sleep. The left precentral gyrus produces beta to stay locked into a mode whether awake or asleep. Cz SMR helps sleep onset. C4 SMR (right side) helps broad sleep architecture, including sleep inertia and sleep paralysis. C3 SMR does the wrong job. Both groups in that study reported some relaxation, which is what left SMR produces, so the study did not disprove anything while drawing a strong negative conclusion.
This is the same pattern I see in the ADHD literature. The meta-analyses concluding neurofeedback does not work for ADHD pool studies that do not run neurofeedback the way it is actually done. Real neurofeedback iterates and individualizes, like personal training. A single fixed protocol across a group for 10 or 20 sessions is not how the work functions. Nobody has run the study that would settle it, because doing 50 to 100 heavily individualized sessions across dozens of people, with proper sham control, would cost somewhere between one and five million dollars. In the meantime there are over a thousand published studies, mostly small samples with small effect sizes, pointing in a consistent direction. Read them yourself rather than trusting a headline. For ADHD specifically, see Does Neurofeedback Work for ADHD?. One note on SMR and ADHD: the cat-on-the-windowsill state of high SMR tone, physical inhibition plus steady attention, is the functional opposite of ADHD, which is why SMR shows up in both the sleep and the ADHD literature.
What are the different types of insomnia?
Map your complaint to a mechanism before you pick a tool.
Primary sleep-architecture insomnia
Trouble with sleep depth or maintenance maps to the central strip. For depth and maintenance issues you might train C3 in the 15 Hz beta range and above. For sleep onset, Cz SMR helps. For sleep inertia in the morning or sleep paralysis, right-side SMR at C4 tends to work better.
The busy, perseverative mind
You are tired and want to sleep, but your brain ramps up, gets stuck, loops songs and conversations. This is a front-midline feature with an executive and obsessive flavor, closer to OCD-style perseveration than to a primary sleep-architecture problem. You target the front midline directly, with neurofeedback or with compounds like N-acetylcysteine. See Biohacking OCD: Targeting the Cortico-Striatal Circuit and Biohacking Anxiety.
Threat-sensitivity insomnia
If you were assaulted at night or traumatized around sleep, a threat response can fire as you try to drop off. That is an anxiety and trauma mechanism, and it calls for different tools than a sleep-architecture problem.
Behavioral and circadian insomnia
If you are ADHD-flavored, staying up late, eating, watching TV, never quite dragging your behavior toward bed, that is a different problem from the busy-mind type, and it responds to circadian and behavioral changes.
Years of heavy alcohol can also erase your downshift lever, leaving you unable to find the gear that drops you into sleep. Same surface complaint, different cause.
What circadian strategies actually fix sleep?
Three behaviors carry most of the weight, and they work across nearly every type of insomnia.
Stop eating before bed. Let your blood sugar drop. The drop allows a strong growth hormone release once you are asleep, which deepens sleep. Eating late shifts your sleep cycle in the wrong direction.
Get up early, and get morning light. Morning light is the flavor that drives circadian entrainment, not evening light. Spend less worry on blue blockers and evening screens and more on getting bright light into your eyes early, ideally seven days a week. The morning is the most important time for fixing sleep. You fix sleep by working counter to the cycle, not by sleeping in. Oscillatory systems need negative feedback with a delay to stay stable, and getting up early is the negative feedback that times the whole rhythm. For the full morning protocol, see Biohacking Your Morning.
Move when you wake. After the bathroom, go for a walk, do some yoga, wake the machine up before you become sedentary, stressed, or start eating.
Which supplements support sleep, and how?
These compounds work by timing the circadian rhythm and supporting the hormonal shift into early sleep, not by sedating you. After several days of consistent use the sleep urge starts asserting itself more strongly on its own.
A starting stack: a microdose of melatonin, some L-theanine, and some GABA. From there you can add ashwagandha or phosphatidylserine, both aimed at lowering cortisol and stabilizing the hormonal and GABAergic metabolism that kicks in during early sleep stages.
ZMA (zinc magnesium aspartate) helps many people sleep and lowers cortisol. Two cautions from my own experience. ZMA can produce intense, even angry dreams in sensitive people. When I first started it I had a run of vivid, stressful dreams about chasing a stolen bicycle barefoot across campus night after night. If your dreams turn weird on ZMA, that is likely the cause. Many ZMA formulas also contain B6, which can drive vivid dreaming on its own, so check your label. Zinc taken in the evening can produce the same nightmare effect in sensitive people.
The environmental basics still matter: keep the room dark and the bed cold.
How do you know which insomnia you have?
The right tool depends entirely on the mechanism, and you cannot guess the mechanism from the surface complaint. A quantitative EEG, or QEEG brain map, shows you where the obstacle sits. Is it sleep onset, sleep depth, a busy front-midline mind, a threat response, a missing sleep urge, morning inertia, sleep paralysis? Each one points to a different protocol and a different set of behavioral and supplement levers.
The brain map is more than the blueprint for a neurofeedback plan. It teaches you how your own brain works, and that knowledge gives you agency, the ability to choose interventions and understand why they matter.
The anxiety around not sleeping often becomes worse than the not sleeping itself. The useful move is to find out which mechanism is in your way, then apply pressure in the right place.
If you want to look at your own sleep architecture, get a QEEG at any Peak Brain office or partner location, or watch the earlier livestreams on SMR and sleep for more depth. Start by mapping the obstacle, then pick the tool that matches it.