This piece comes from my weekly Neurofeedback and Chill livestream, where I run a neurofeedback session on my own head, walk through the mechanics, and answer questions live. This week I worked through several recent discoveries about the cells in your brain, then took audience questions on impulse control, seizures, tinnitus, vertigo, and alpha-theta training. I have anonymized everyone who asked a question.
When I was in school in the 1970s and 80s, the teaching was simple and wrong: you are born with all the brain cells you will ever have, and the count only goes down. We now know you generate new neurons throughout life. Even at 70, you are making roughly 600 to 700 new neurons a day. And neurons are far from the only cell type in your skull. A 2024 atlas indexed more than 3,000 distinct brain cell types. We are still scratching the surface of how they work.
What Are the Newly Discovered Brain Cells?
Ovoid cells: a memory cell for objects
You may already know about place cells. These sit in the hippocampus and fire when you are in a specific location. They are packed with the ability to produce BDNF, the main plasticity growth factor, and they go to work hard when you explore a new environment. That is one reason novelty is good for brain health. New environments light up place cells and drive information flow.
Researchers have now described a related cell with a different specialty: the ovoid cell, a small egg-shaped neuron that responds to physical objects rather than places. See a new object and the ovoids respond, then their firing patterns update as they learn that object. Place cells map where you are. Ovoid cells help map what you are looking at. Both feed the same hippocampal learning machinery I discuss in Biohacking Memory.
Glia-neuron hybrid cells that produce glutamate
After a serious brain injury, you get a flood of glutamate and blood into the torn tissue. Glutamate in excess and iron from blood both kill neurons fast. Microglia and astrocytes rush in, form physical scar tissue, and metabolize those toxic loads. We have long known that 8 to 10 years after a major injury, seizures sometimes appear in that same region. One proposed mechanism: the glial cells that arrived as cleanup crew start producing glutamate themselves, and that excess glutamate organizes into spike events.
A newer study suggests there may be an intermediate cell type sitting halfway between glia and neuron, neither fully one nor the other, and that this hybrid may be the glutamate source. The clinical relevance shows up in seizure work, where sensorimotor rhythm training can help the brain inhibit those spike events. More on that mechanism below.
A fourth meningeal layer (SLYM)
The classic teaching gives the brain three covering layers: the dura mater on the outside, the arachnoid in the middle, and the pia mater hugging the cortex. About five years ago we added the glymphatic system, a lymph network that drains the brain into the body and ties into immune signaling. Now there is a fourth meningeal layer described, the subarachnoid lymphatic-like membrane, or SLYM. It sits below the arachnoid and above the pia, carries lymphatic flow, and acts as both a physical and immune shield for the brain.
Spiral brain waves
The waves you see on an EEG read as oscillations moving up and down. Newer imaging shows wave patterns that spiral across the cortex rather than simply rising and falling, and these spirals appear to carry information. If you want the basics of what EEG bands mean and how I read them, start with QEEG Brain Mapping.
Neuron-AI organoid interfaces
You have heard of mini-brains, the small clusters of neurons grown in dishes for drug testing and studying information processing. These organoids, sometimes built from dozens of interacting cells, have now been wired to AI systems, with working information interfaces between the living neurons and the model. Pair living neural tissue with current large language models and reinforcement learning and you are looking at genuinely uncharted territory over the next few years.
The "mind-body" region that is really the precentral gyrus
One article framed a region as a brand-new discovery about the mind-body connection. The region is the precentral gyrus, the most posterior strip of the frontal lobe. Anyone who does neurofeedback knows this tissue well. The right precentral gyrus supervises whether your attention is actually engaged the way you intend. It sits over a descending pathway through the thalamus into the body and an ascending pathway into sensory cortex behind it. That central strip, C3 and C4 in EEG terms, is the control point between thought, feeling, action, and the body. The rest of the field is catching up to what neurofeedback practitioners have used clinically for decades.
What Was Dr. Hill Training on His Own Head?
I ran a single-channel montage, C4 referenced to the right ear (A2), to bias the training toward the right hemisphere. The protocol had three pieces:
- Inhibit 4 to 7 Hz (theta)
- Reward 11.5 to 14.5 Hz (sensorimotor rhythm, SMR)
- Inhibit 22 to 34 Hz (high beta)
In plain terms: make less theta, make a bit more SMR, and hold down high beta over the right sensorimotor strip. I am not training to a database average. You do not train any brain to the middle of the bell curve unless you are doing z-score training. I train in the direction a given brain and a given person want to go. For more on what SMR does, see SMR Neurofeedback.
How Does SMR Training Help Impulse Control and Sleep?
Your brain runs at the edge of criticality. Too ordered is functionally dead; too chaotic is a seizure. Living and thinking means balancing right at that edge, and your brain constantly pumps the brakes to keep overactivation from tipping into a coherent spike event. Sensorimotor rhythm is one of the main resources it uses to do that braking.
SMR only behaves like SMR on the sensorimotor strip. The same 13 to 15 Hz frequency elsewhere is ordinary beta processing. On the strip, SMR functions like a calming, regulatory rhythm despite its beta-range speed. The same thalamocortical circuits that generate waking SMR also generate sleep spindles (12 to 14 Hz bursts, called sigma during sleep) that hold you asleep. That shared circuitry explains why building SMR often improves daytime focus and nighttime sleep stability at once.
The flip side is theta. Tonic theta, the kind that sticks in a tissue and keeps it running on autopilot, shows up reliably on the front midline. Two very recognizable complaints map onto front-midline theta: songs stuck in your head all day, and compulsive nail-biting. Nail-biting belongs to a broader category of picking and grabbing behaviors, ticking and threading, that sit in the same neighborhood as OCD-type phenomena. Tuning down that extra theta gives you some control back over that tissue.
For ADHD specifically, the numbers from my clinical work are consistent: roughly one full standard deviation of improvement in impulsivity against an age-matched sample about every other month, around 25 sessions. Someone three standard deviations off the mean can walk that down into typical or above-average range over a course of training, with the change showing up on brain maps, on go/no-go testing, and in daily life. Impulsivity from a brain injury moves about half as fast as built-in patterns like ADHD, but it still moves. If you want the deeper dive on attention, see Does Neurofeedback Work for ADHD?.
Does Neurofeedback Help Seizures, Tics, and Tinnitus?
For seizures, the evidence is among the strongest in the field. Sterman's review work reported an average seizure reduction near 50%. In my own clinical experience, I have rarely seen a result that poor. Across the 30 to 40 people I have worked with whose seizure disorder was the primary complaint, almost all had strong reductions. In about half, the seizures crept back at a low level around a year later, and a second round of training tended to shut them down more permanently. Seizure work should sit near the top of the intervention list. The mechanism is the same SMR braking system described above, strengthening the brain's ability to inhibit the ictal spike.
Tourette's is a tic disorder, and I treat it much the way I treat OCD-type phenomena: a discharge problem, often involving the anterior cingulate coupling to speech and language areas. The flavor of the symptom tracks which region the anterior cingulate yokes to. Couple it to mid-temporal language tissue and you get vocal tics; couple it to the right temporoparietal junction running hot and you get environmentally triggered patterns like misophonia, claustrophobia, or agoraphobia. Tics and obsessions respond well to training, even if no single case is guaranteed. For the underlying circuit, see Biohacking OCD.
Tinnitus is more hit or miss. About half of clients see a change with neurofeedback. When it does start to move for someone, I am confident we can keep moving it toward a stable, lasting reduction. I am honestly not sure why we cannot get a larger effect for more people.
For vertigo and physical balance issues, after you have ruled out structural causes with imaging and tried the Epley maneuver, a neural component is worth checking. On a QEEG I look for slow brainwave clusters in the temporal lobes, or a lack of alpha paired with excess beta. Posterior temporal lobe protocols (T5, T6) tend to help with grounding and balance. If migraines are in the mix, I add HEG, hemoencephalography, to train vascular blood flow rather than only EEG amplitudes.
Does Alpha-Theta Training Do Something Special?
Alpha-theta re-educates the ability to enter nonlinear states, to reach flow, to access emotional material and talk about it. The key event is the crossover: alpha rises first, then drops, while theta rises and keeps rising. Even before the amplitudes literally cross, a rising theta against falling alpha is the crossover signature, because that theta rise is the release, the brakes coming off. More specifically, the 6.5 Hz theta burst is the moment of "aha," a brief spindle that releases information from wherever it was being held.
Excess beta interferes with reaching that soft state. I usually hold off on alpha-theta until someone has resolved excess alpha or theta and has settled their beta load to a manageable level, because brains in that calmer configuration respond far better to the protocol. It is a powerful protocol, healing for many people, and I only run it once I know a person's brain, their grounding points, and their history. If you want the broader picture on these states, see Biohacking Flow State and Mindfulness.
What Are the Side Effects of Neurofeedback?
Think of doing exercise wrong. Train completely the wrong thing a few sessions in and you will feel it: train beta when you needed alpha and you will probably feel anxious and a little off for twenty minutes to a couple of hours, then it wears off. Train too much beta on the left for focus and you might feel sharp during the day but wired at bedtime. These transient effects are useful data. They tell your provider to back off the time, drop the frequency, or add a stabilizing protocol such as SMR at the vertex alongside the focus training. The subtle effect a person feels before a change becomes permanent is exactly how good providers individualize protocols.
What This Means for You
The brain you were taught about in school was a fixed object that only declined. The brain we actually study generates new neurons daily, contains thousands of distinct cell types, runs information on spiraling waves, and is plastic enough to retrain at the scalp. If you want to see how your own brain is organized, a QEEG with an executive function assessment is the place to start, and from there you can train specific circuits over a course of sessions, remotely if needed. To learn what mapping shows, read Is Neurofeedback Legitimate? and Biohacking Plasticity, then book a brain map and watch your own data change.