The Brain Waves Too Slow for Anyone to Notice

Your Brain Has a Rhythm So Slow You Can't Feel It

Here's a thought experiment. Imagine you're standing on the beach, watching ocean waves. You can see the regular waves rolling in every few seconds. You can see the larger swells that come every 20 or 30 seconds. If you're patient, you might notice the tide shifting over the course of an hour.

Now imagine someone tells you: the most important wave in the entire ocean isn't any of the ones you can see. It's a wave so massive and so slow that you'd need to stand on that beach for ten minutes just to watch a single cycle complete. And this wave, they claim, is the one that organizes all the faster waves above it.

You'd be skeptical. Reasonably so.

That's roughly the situation in one corner of the neurofeedback world. A growing group of clinicians claims that the most powerful brainwaves your head produces aren't the ones neuroscientists typically study (the alpha, beta, theta, and gamma oscillations between 1 and 100 Hz). Instead, they point to fluctuations that unfold over seconds or minutes, oscillations below 0.1 Hz. Waves so slow that a single cycle takes 10 to 100 seconds to complete.

They call this infraslow neurofeedback. And the debate around it is one of the most interesting arguments in modern brain science.

First, a Refresher on What "Normal" Neurofeedback Trains

Before we go into the deep end of infraslow territory, let's make sure the foundation is solid.

Standard neurofeedback targets the EEG frequency bands you've probably heard about: delta (0.5 to 4 Hz), theta (4 to 8 Hz), alpha (8 to 13 Hz), beta (13 to 30 Hz), and gamma (30 Hz and above). These oscillations are produced by synchronized populations of cortical neurons firing in rhythmic patterns, and each band corresponds roughly to a different mental state. Theta correlates with drowsiness and creativity. Alpha reflects calm alertness. Beta tracks active thinking.

When you do traditional neurofeedback, an EEG device measures these oscillations in real time, a computer processes the signals, and you receive feedback (usually a video that plays smoothly or a tone that sounds) when your brain produces the target pattern. Over sessions, your brain learns to produce that pattern more reliably. It's operant conditioning applied to cortical electrical activity.

This approach has the strongest evidence base in neurofeedback for ADHD brain patterns, where training the ratio of theta to beta brainwaves over the frontal cortex has been studied in dozens of controlled trials and rated as a Level 1 intervention by the American Academy of Pediatrics.

All of this happens in the frequency range of roughly 1 to 40 Hz. The EEG machines used in most clinics and research labs are designed to capture this window. They apply high-pass filters, typically around 0.5 or 1 Hz, that deliberately block anything slower.

And that's where the story gets interesting. Because the brain doesn't stop oscillating below 1 Hz. It keeps going. Way, way down.

What Is the Secret Basement of Brain Activity?

In the early 2000s, neuroscientists started paying serious attention to something that had been lurking in their data for decades: extremely slow fluctuations in brain activity, oscillating at frequencies below 0.1 Hz. That means under one cycle per ten seconds.

These infraslow fluctuations (ISFs) had been noticed before, but they were mostly treated as noise. After all, when your EEG amplifier has a drift problem, the signal wanders slowly over time. How do you tell the difference between a genuine brain oscillation at 0.01 Hz and an equipment artifact at 0.01 Hz? For a long time, most researchers didn't bother trying. They filtered it out and moved on.

But a handful of researchers did bother. And what they found was striking.

Using DC-coupled EEG amplifiers (which don't filter out slow signals) and cross-referencing with fMRI data, they demonstrated that infraslow oscillations are real neural phenomena, not artifacts. A landmark 2006 paper by Vanhatalo and colleagues published in the Proceedings of the National Academy of Sciences showed that infraslow EEG fluctuations correlate strongly with the slow fluctuations visible in fMRI blood-oxygen-level-dependent (BOLD) signals. The same slow rhythm showed up in both measurement modalities, which is hard to explain if it's just noise.

Here's the "I had no idea" moment: these infraslow oscillations appear to modulate everything above them. The amplitude of your alpha brainwaves, your beta waves, even your gamma waves, rises and falls in sync with these ultra-slow fluctuations underneath. Think of it like this: if your normal brainwave bands are musicians in an orchestra, infraslow oscillations might be the conductor. They don't play any notes themselves, but they determine when the instruments get louder and when they get softer.

Research from Palva and Palva (2012), published in Trends in Neurosciences, proposed that infraslow oscillations create temporal windows of excitability across the cortex, essentially controlling when faster oscillations can fire and how strongly they fire. This is called cross-frequency coupling, and it suggests that the slowest brain rhythms sit at the top of a hierarchical organization of brain timing.

The Othmer Method: Training What Others Filter Out

The clinical application of infraslow neurofeedback is most closely associated with Siegfried and Sue Othmer, two figures who have been central (and controversial) in the neurofeedback community for decades.

The Othmers' story is personal. Their son Brian suffered from severe epilepsy and behavioral problems in the 1980s. Conventional treatment wasn't working. They encountered neurofeedback through the early work of Barry Sterman, whose research on SMR (sensorimotor rhythm) training in cats had led to preliminary human applications for seizure control. The results they saw with Brian led them to devote their careers to neurofeedback development.

By the mid-2000s, the Othmers had shifted their focus to infraslow frequencies. They developed the Cygnet neurofeedback system and a clinical method that specifically targets oscillations below 0.1 Hz, typically in the range of 0.001 to 0.1 Hz.

Here's how their protocol works in practice.

The Training Setup

The clinician places EEG electrodes on the client's scalp, typically at one or two sites based on the presenting symptoms. The EEG amplifier is DC-coupled, meaning it doesn't filter out the slow frequencies that conventional equipment blocks. The system records the infraslow fluctuations in cortical electrical potential.

The Feedback Mechanism

Instead of a video that pauses and plays (as in traditional neurofeedback), ISF training typically uses continuous auditory feedback. The client hears a tone or ambient sound whose properties (pitch, volume, or spatial positioning) shift in real time based on the infraslow signal. The relationship is continuous rather than binary: the sound morphs fluidly as the brain's slow potential shifts.

This is a key difference from classical protocols. In traditional neurofeedback, the feedback is reward-based: you either hit the threshold or you don't. In ISF training, the feedback mirrors the brain's activity continuously, giving the brain constant information about its own slow fluctuations.

The Optimal Frequency Search

Here's the part that skeptics find most problematic. The Othmer method involves finding each client's "optimal frequency" within the infraslow range. The clinician starts at a particular frequency (typically around 0.01 Hz) and adjusts up or down in small increments, observing the client's reactions. If the client reports feeling calmer, more alert, or more settled at a particular frequency setting, the clinician continues training there.

The optimal frequency varies from person to person and even from session to session. It's determined largely by clinical observation and client self-report rather than by an objective biomarker. This reliance on subjective tuning is one reason the method draws criticism from researchers who favor standardized, manualized protocols.

What Conditions Do Practitioners Target?

ISF neurofeedback practitioners report applying the method across a remarkably wide range of conditions. This breadth is itself a source of both enthusiasm and suspicion. When something claims to help everything, you should ask questions.

The pattern you'll notice: the conditions where ISF is most commonly applied tend to involve dysregulated cortical excitability or disrupted baseline brain state regulation. This is consistent with the theoretical model that infraslow oscillations govern the brain's overall excitability set-point.

But "consistent with the theoretical model" is not the same thing as "proven." Let's look at the evidence honestly.

The Evidence: What We Know and What We Don't

This is the section where you need your skeptic hat on. The evidence for infraslow neurofeedback is genuinely interesting but genuinely limited.

What the Research Shows About Infraslow Oscillations Themselves

The neuroscience of infraslow oscillations is on solid ground. Multiple labs using different methodologies have confirmed that:

• Infraslow fluctuations (under 0.1 Hz) are real neural signals, not artifacts
• They correlate with fMRI BOLD fluctuations and resting-state network dynamics
• They modulate the amplitude and phase of faster oscillations through cross-frequency coupling
• They relate to cortical excitability states
• They are disrupted in several neurological and psychiatric conditions

This research comes from respected labs and has been published in top journals including PNAS, The Journal of Neuroscience, and Trends in Neurosciences. The existence and functional importance of infraslow oscillations is not seriously disputed.

What the Research Shows About Training Them

This is where things get thinner. The clinical ISF neurofeedback literature consists primarily of:

• Practitioner reports and case series (many from the Othmers' own training network)
• A handful of small controlled studies
• Review articles that cite the basic science on infraslow oscillations alongside the clinical case reports

What's missing is the gold standard: large, randomized, double-blind, sham-controlled trials. The kind of studies that established traditional neurofeedback's evidence base for ADHD took decades of work by multiple independent research groups. ISF neurofeedback hasn't undergone that level of scrutiny yet.

A 2013 study by Smith, Collura, and colleagues published in Journal of Neurotherapy reported that ISF training improved symptoms in a group of patients with various conditions, but the study design lacked a proper sham control. A 2016 case series by the Othmer group documented improvements in migraine frequency and severity, but again without a randomized control condition.

Why the Gap?

There are legitimate reasons the evidence base is thin, and they're worth understanding rather than dismissing.

First, the equipment requirements limit who can do ISF research. You need DC-coupled amplifiers, which are more expensive and technically demanding than standard EEG systems. This means fewer labs have the hardware to investigate.

Second, the individualized nature of the Othmer method makes it difficult to standardize for clinical trials. When the optimal training frequency varies by person, designing a clean sham condition is harder than in traditional neurofeedback, where you can deliver feedback based on someone else's EEG recording.

Third, and this is the blunt reality: the neurofeedback community is small, and the ISF sub-community within it is smaller still. Large randomized controlled trials are expensive. They require funding, institutional support, and research infrastructure that this niche field hasn't yet attracted at scale.

None of these reasons excuse the evidence gap. They explain it. Explaining why evidence is thin is not the same as saying it isn't needed.

The Crown captures brainwave data at 256Hz across 8 channels. All processing happens on-device. Build with JavaScript or Python SDKs.

Explore the Crown

The Controversy: Why Scientists and Clinicians Can't Agree

The debate around ISF neurofeedback isn't just about data. It's about epistemology. How much evidence do you need before you adopt a clinical technique?

The skeptic's position goes something like this: the history of medicine is littered with plausible-sounding interventions that turned out not to work once properly tested. The brain is complicated. Slow oscillations are real, but that doesn't mean feeding them back through headphones trains anything useful. The lack of controlled trials isn't a detail to be addressed later. It's the whole point. Until ISF passes the same evidentiary bar that traditional neurofeedback protocols have passed, it should be labeled experimental.

The practitioner's position goes something like this: we have hundreds of clinical cases showing consistent improvements across multiple conditions. The neuroscience of infraslow oscillations supports the mechanism. Waiting for perfect evidence means withholding something that helps real people with real suffering. Traditional neurofeedback itself was dismissed by mainstream neuroscience for decades before the RCTs caught up with the clinical reality.

Both sides have a point. The trick is holding both truths simultaneously without collapsing into either blind faith or reflexive dismissal.

What would settle the debate is straightforward: multi-site, randomized, sham-controlled trials with standardized ISF protocols. These studies are feasible. They just haven't been done at the necessary scale. Until they are, ISF neurofeedback occupies an awkward middle ground: too much clinical signal to ignore, too little controlled evidence to fully endorse.

The Hardware Question: Why Your EEG Amplifier Matters

If you're interested in ISF neurofeedback, either as a practitioner or as someone curious about trying it, the hardware situation deserves a frank conversation.

Infraslow oscillations live below the frequency range that most EEG devices are designed to capture. Here's why.

Standard EEG amplifiers are AC-coupled. This means they include a high-pass filter (typically at 0.5 Hz or higher) that blocks very slow signals. This filter exists for good reasons: electrode drift, skin-electrode impedance changes, and movement artifacts all produce slow voltage shifts that can overwhelm the signal you're trying to measure. The high-pass filter cleans all of that out. But it also cleans out the infraslow brain oscillations.

To record ISF signals, you need a DC-coupled amplifier with a very long time constant. These amplifiers let through the slow fluctuations, but they also let through all the noise and drift that the high-pass filter would normally remove. This means ISF recording demands more careful electrode preparation, more stable electrode-skin interfaces, and more sophisticated artifact handling.

Where the Neurosity Crown Fits

Let's be honest about this, because honesty builds more trust than marketing ever could.

The Neurosity Crown is an 8-channel EEG device sampling at 256Hz with on-device processing via the N3 chipset. It's designed for real-time brain monitoring and neurofeedback development in the classic frequency bands. Its standard signal processing pipeline includes filtering that focuses on the conventional EEG range.

For infraslow neurofeedback in the Othmer sense, you'd need a DC-coupled system specifically built for that purpose. The Crown isn't designed for ISF training.

What the Crown does exceptionally well is everything above the infraslow range: capturing the delta, theta, alpha, beta, and gamma oscillations that decades of research have validated as trainable markers of attention, calm, focus, and cognitive state. Its 8-channel coverage across all cortical lobes, 256Hz sample rate, and open SDKs in JavaScript and Python make it one of the most capable consumer EEG platforms for building and experimenting with neurofeedback applications in these well-established bands.

And here's something worth noting: if infraslow oscillations truly modulate everything above them, then tracking the faster bands might give you an indirect window into what's happening at the infraslow level. Changes in alpha power, beta variability, and cross-frequency dynamics that the Crown can detect may reflect shifts in the underlying infraslow architecture. This is speculative, but it's the kind of thing that open SDK access and a creative developer community could explore.

What the Infraslow Story Tells Us About Brain Complexity

Step back from the clinical debate for a moment and consider what infraslow oscillations reveal about the brain as a system.

Your brain operates on timescales spanning several orders of magnitude. Gamma oscillations cycle at 40 to 100 times per second, organizing local neural computations on the scale of milliseconds. Delta waves roll through during deep sleep at 1 to 4 cycles per second. And infraslow fluctuations unfold over 10 to 100 seconds, potentially orchestrating the dynamics of everything faster.

This is like discovering that beneath the weather (which changes hour to hour) and the seasons (which change month to month), there are climate cycles operating over decades and centuries that shape everything above them. The brain's timescale hierarchy suggests a far more layered regulatory architecture than we typically appreciate.

It also suggests something humbling: we've been studying brain dynamics through a narrow temporal window. The conventional EEG range of 1 to 100 Hz captures an important slice of brain activity, but it's not the whole story. Infraslow research is a reminder that the brain is always more complicated than our instruments are designed to reveal.

Where ISF Neurofeedback Goes from Here

The future of infraslow neurofeedback depends on one thing more than anything else: rigorous controlled research.

The basic science is in place. Infraslow oscillations are real, functionally important, and disrupted in clinical conditions. The clinical reports are suggestive. The theoretical model is plausible. What's needed is the bridge between those two: carefully designed trials that test whether ISF training produces effects beyond placebo, beyond the therapeutic relationship, and beyond the general relaxation that comes from sitting quietly with headphones on for 30 minutes.

This research is beginning to happen, slowly. As EEG hardware continues to improve and as DC-coupled amplifiers become more affordable and accessible, the technical barriers to ISF research are lowering. Academic interest in infraslow oscillations as a research target has grown substantially since the mid-2000s, and it's plausible that clinical trials will follow.

In the meantime, if you're considering ISF neurofeedback as a client, here's a reasonable approach: seek a practitioner who is transparent about what the evidence does and doesn't show, who monitors your progress with objective measures (not just subjective reports), and who treats ISF as part of a comprehensive approach rather than a standalone miracle.

And if you're a researcher, developer, or neuroscience enthusiast interested in brain oscillations and neurofeedback more broadly, the tools available today let you explore more than any previous generation could. Eight channels of real-time EEG data, open APIs, and integration with AI tools open up possibilities for investigating brain dynamics that didn't exist even five years ago.

The Slowest Waves Might Carry the Biggest Signals

There's a pattern in science that repeats across disciplines. Researchers develop instruments tuned to a specific range. They study what those instruments can see. They filter out what they can't. And then, decades later, someone widens the aperture and discovers that the stuff they were filtering out was important all along.

Radio astronomers filtered out low-frequency signals for years before discovering the cosmic microwave background radiation. Geneticists called 98% of the genome "junk DNA" before discovering it was full of regulatory elements controlling the other 2%.

Infraslow oscillations may follow the same arc. For decades, EEG researchers filtered them out as noise. Now a growing body of evidence suggests they're a fundamental organizing layer of brain dynamics. Whether training these oscillations with neurofeedback produces the clinical results practitioners describe is still an open question, one that deserves better studies, not just louder opinions.

What isn't open to question is that your brain is doing more than you can see through any single measurement window. The signals are there. The hierarchy is real. And the more we learn about how the brain coordinates its own activity across timescales, the closer we get to something that matters deeply: understanding how the most complex system in the known universe manages to produce a coherent you, moment by moment, from the millisecond flutter of gamma to the slow, deep roll of the infraslow.

That's a question worth staying curious about.