The Four Maps Your Brain Cycles Through Every Second
Your Brain Has a Clock Speed. It's About 10 Ticks Per Second.
Here's something that will change how you think about your own mind.
Right now, as you read this sentence, your brain isn't doing one continuous thing. It's doing a rapid series of discrete things. Snap, snap, snap, snap. About ten times every second, the entire electrical landscape across your scalp reorganizes into a new configuration, holds stable for roughly a tenth of a second, then snaps into the next one.
Not a gradual fade. A snap. Like channel surfing on a television, except the channels are states of consciousness and the remote is being operated by something we don't fully understand yet.
These brief, stable configurations are called EEG microstates. And here's the part that really bends your mind: there aren't thousands of them. There aren't even dozens. Across decades of research, involving thousands of participants and millions of data points, scientists have found that the vast majority of your waking brain activity can be described by just four topographic maps. Four. Labeled, with the creativity typical of scientists, A, B, C, and D.
Four maps, cycling in rapid succession, somehow producing the entire richness of human conscious experience. The neuroscientist who discovered them, Dietrich Lehmann, called them the atoms of thought. And that name, once you understand what it means, is one of the most profound ideas in modern neuroscience.
Before Microstates: The Problem With Treating EEG Like a Radio
To understand why microstates matter, you need to understand the limitation they solve.
For most of EEG's history, scientists analyzed brainwaves the way you'd analyze music. Break the signal down by frequency. How much alpha? How much beta? How much theta? This is called spectral analysis, and it's genuinely useful. It tells you a lot about the brain's general state: whether someone is alert, drowsy, meditating, or in deep sleep.
But spectral analysis has a blind spot. It treats EEG like a single signal, or at best, a collection of independent signals from different electrodes. It doesn't ask: what does the pattern across all electrodes look like at any given instant?
Think about it this way. Imagine you have eight microphones recording a conversation in a room. Spectral analysis is like analyzing each microphone's audio separately, checking the frequency content of each one. You'd learn a lot about the sounds in the room. But you'd miss something critical: the spatial pattern. Which direction is the sound coming from? How does the configuration of sound across all eight microphones change over time?
That spatial, whole-head perspective is exactly what microstate analysis provides. Instead of looking at individual channels or frequency bands, you look at the topographic map of voltage across the entire scalp at each moment in time. And when you do that, something astonishing falls out of the data.
The Discovery: Lehmann's Radical Observation
In 1987, Dietrich Lehmann at the University of Zurich did something that seems obvious in hindsight but was genuinely radical at the time. He took high-density EEG recordings and, instead of analyzing frequencies, he looked at the global topography of the voltage field at each individual time point.
What he expected to see was a continuously changing landscape. After all, the brain is constantly active, constantly shifting. The topographic map should morph smoothly and endlessly, like a weather system.
What he actually saw was completely different.
The topography would stabilize. For about 80 to 100 milliseconds, the pattern of positive and negative voltages across the scalp would hold roughly constant. Then, abruptly, it would flip to a different configuration. That new configuration would hold for another 80 to 100 milliseconds. Then flip again.
The EEG wasn't a smooth river. It was a series of steps.
And the configurations it stepped through weren't random. When Lehmann and later researchers applied clustering algorithms to identify the most common topographies, the same four maps kept appearing. Study after study after study. Different participants, different labs, different countries. Four maps. Always the same four.
This was the birth of EEG microstate analysis. And it raised a question that neuroscience is still grappling with: why four?
The Four Canonical Microstates: A Field Guide to Your Brain's Building Blocks
Let's meet each one. The standard labeling (A, B, C, D) is based on the orientation of the electrical field across the scalp. Each microstate corresponds to a distinct spatial configuration, and each one has been linked to specific cognitive networks through simultaneous EEG-fMRI studies.
| Microstate | Topography | Associated Network | Cognitive Function | Typical Duration |
|---|---|---|---|---|
| A | Right-anterior to left-posterior | Auditory / temporal networks | Phonological processing, verbal thinking | ~65-80 ms |
| B | Left-anterior to right-posterior | Visual / occipital networks | Visual processing, imagery | ~80-100 ms |
| C | Symmetric anterior-posterior | Salience network, default mode | Interoception, self-referential thought | ~100-120 ms |
| D | Frontal midline dominant | Dorsal attention network | Focused attention, executive control | ~80-100 ms |
Microstate A: The Voice in Your Head
Microstate A shows a diagonal voltage pattern running from right-anterior to left-posterior. In fMRI studies, its occurrence correlates with activation of temporal cortex regions, specifically areas involved in auditory and speech processing.
In plain English: microstate A appears to be what your brain looks like when it's doing verbal, phonological processing. When you're reading silently and "hearing" the words in your head. When you're composing a sentence before you speak it. When you're replaying a conversation from earlier today. The electrical signature of language-related thinking has a specific topographic fingerprint, and that fingerprint is microstate A.
Research from Britz and colleagues (2010) published in NeuroImage confirmed this link by recording EEG and fMRI simultaneously. Every time microstate A appeared in the EEG, auditory cortex regions lit up in the fMRI. The correlation was striking.
Microstate B: The Mind's Eye
Microstate B is essentially the mirror image of A, running from left-anterior to right-posterior. Its fMRI correlate is activation in visual processing regions, particularly the occipital cortex and areas associated with visual imagery.
This is your brain in "seeing" mode. Not just processing visual input from your eyes, but generating internal visual representations. When you imagine what your apartment looks like from above. When you picture a friend's face. When you mentally rotate an object. The topographic map snaps to microstate B.
Here's what's interesting about A and B being near-mirror images of each other: it suggests that verbal and visual processing, two of the most fundamental modes of human cognition, are organized as complementary spatial patterns across the brain. Your brain literally flips its electrical orientation depending on whether you're thinking in words or pictures.
Microstate C: The Self Network
Microstate C has a symmetric, anterior-to-posterior orientation. It's the one that correlates with activation of the default mode network and the salience network, two of the most studied large-scale brain networks.
The default mode network is what fires up when you're not focused on the outside world. It's the network of daydreaming, self-reflection, and mind-wandering. The salience network decides what deserves your attention by monitoring both external stimuli and internal body signals (a process called interoception).
Microstate C, then, is your brain turned inward. It tends to have the longest duration of the four microstates, typically 100 to 120 milliseconds, which makes intuitive sense. Self-referential processing, the feeling of being a continuous "self" moving through time, seems like the kind of thing that would take a bit longer per cycle.
Studies have found that microstate C increases during rest and decreases during focused task performance. When you're lost in thought, zoning out, or ruminating, your brain spends more time in the microstate C configuration.
Your sense of having a continuous, unified self might actually be built from discrete 100-millisecond snapshots. Microstate C, which correlates with self-referential brain networks, fires roughly 8 to 10 times per second. Between those pulses, your brain is busy doing other things (processing language, handling visual information, directing attention). The subjective experience of being "you" isn't a smooth stream. It's a flickering signal that your brain stitches together into the illusion of continuity. Consciousness, in a very real sense, has a frame rate.
Microstate D: The Spotlight
Microstate D shows a frontal midline topography and correlates with the dorsal attention network, the system responsible for voluntary, goal-directed attention. This is your brain's spotlight, the configuration it snaps into when you're deliberately focusing on something.
When you read a difficult paragraph and really bear down on it. When you're scanning a crowded room for a friend's face. When you're debugging code and holding a complex logical structure in working memory. Your brain's topography flips to microstate D.
Microstate D is particularly interesting for anyone who cares about focus and cognitive performance, because its parameters (how long it lasts, how often it occurs, how smoothly it transitions to other states) appear to be direct markers of attentional capacity. More on that shortly.
How Microstates Tile Your Consciousness
Now here's where it gets really interesting. These four maps don't just occur in isolation. They form sequences. And the properties of those sequences, the grammar of how your brain strings microstates together, carry as much information as the microstates themselves.
Researchers measure several key parameters:
Duration: How long each microstate lasts before the brain transitions to another one. In healthy adults at rest with eyes closed, the average is about 80 to 100 milliseconds, though this varies by class. Longer duration suggests more stable processing in that particular mode.
Occurrence: How frequently each microstate class appears per second. Typical values are 2 to 4 occurrences per second per class, meaning the brain cycles through each map a few times every second.
Coverage: What percentage of total recording time each microstate class occupies. In healthy adults, the four classes typically cover roughly equal territory, though microstate C often has slightly higher coverage due to its longer duration.
Transition probabilities: The likelihood of transitioning from one microstate to another. This is where the "grammar" metaphor comes in. The transitions aren't random. Microstate A is more likely to transition to microstate C than to microstate B, for example. These transition preferences are consistent across individuals and appear to reflect the brain's underlying functional architecture.
Think of the four microstates as letters in a four-letter alphabet. Your brain writes "sentences" using these letters, producing sequences like A-C-D-B-A-D-C-B-D-A. The meaning isn't just in the individual letters. It's in the patterns. A healthy brain produces sequences with specific statistical properties: particular transition probabilities, characteristic durations, and predictable coverage ratios. When these properties change, it often signals a change in brain state or, in clinical contexts, a neurological or psychiatric condition.
Why Clinicians Care: Microstates as Biomarkers
The real power of microstate analysis hit the neuroscience community when researchers started comparing microstate parameters across clinical populations. What they found was remarkable: different brain disorders produce different, measurable disruptions to the microstate landscape.
Schizophrenia: The Broken Spotlight
The strongest finding in clinical microstate research involves schizophrenia. Across multiple studies and meta-analyses, patients with schizophrenia consistently show shortened duration of microstate D, the attention-related map.
Think about what this means. The topographic configuration associated with focused, goal-directed attention is less stable in schizophrenia. It flickers on and off more rapidly, spending less time in each episode. This aligns precisely with one of the core cognitive symptoms of the disorder: difficulty maintaining sustained attention and a fragmented quality to conscious experience.
But it's not just microstate D. Studies have also found increased occurrence and coverage of microstate C (the self-referential map) in schizophrenia, which dovetails with the excessive self-monitoring and aberrant salience detection characteristic of the condition.
Depression and Anxiety
Depression alters the microstate landscape differently. Several studies have reported changes in microstate C parameters in major depressive disorder, specifically increased duration and coverage. If microstate C represents self-referential processing and rumination, this finding maps perfectly onto the clinical experience of depression: an excessive, sticky focus on the self that's hard to break free from.
Anxiety disorders show their own signature, with some research indicating disrupted transition probabilities, the brain getting "stuck" in certain microstate sequences rather than flowing freely between them.
ADHD brain patterns and Aging
ADHD research has found disrupted microstate sequencing, with altered transition probabilities suggesting a less organized temporal structure to brain state dynamics. The brain still produces all four microstates, but the way it strings them together is different, less predictable, more chaotic.
Aging produces a gradual shift in microstate parameters as well. Older adults tend to show shorter overall microstate durations and altered coverage ratios, potentially reflecting the well-documented changes in large-scale brain network connectivity that accompany aging.
The Atoms of Thought: What Microstates Tell Us About Consciousness
Here's where we zoom all the way out.
The existence of EEG microstates raises one of the most provocative questions in neuroscience: is consciousness discrete?
We experience the world as a smooth, continuous flow. One moment blends into the next. There are no visible "frames" in the movie of your life. But the microstate data suggests something different. It suggests that your brain processes information in chunks, brief packets of about 100 milliseconds each, and that what you experience as continuous awareness is actually your brain's remarkably smooth editing job, stitching these packets together so smoothly that the seams disappear.
This idea isn't as wild as it sounds. Other evidence points the same way. Psychological research on temporal binding shows that events occurring within roughly 100 milliseconds of each other are perceived as simultaneous. The "specious present," the philosophical concept of the minimum duration of subjective experience, has been estimated at somewhere between 50 and 200 milliseconds.
The microstate duration of 60 to 120 milliseconds sits right in the middle of that range.
Lehmann's "atoms of thought" metaphor becomes powerful when you think about what atoms do. They're the basic building blocks. They combine into molecules. Molecules combine into complex structures. If microstates are the atoms, then maybe sequences of microstates are the molecules, and patterns of sequences are the tissues and organs of thought. The richness of human experience wouldn't need to emerge from a single complex process. It could emerge from the combinatorial possibilities of a small set of simple elements rapidly sequencing.
This is, to be clear, still a hypothesis. But it's a hypothesis with increasing empirical support, and it connects microstate research to some of the deepest questions about the nature of mind.
Meditation, Flow, and the Microstate Fingerprint
If you're interested in mental performance rather than clinical diagnosis, microstates have something to say to you too.
Meditation studies have found that experienced meditators show altered microstate parameters compared to non-meditators. Specifically, focused-attention meditation tends to increase microstate D duration (the attention map stays stable longer) and decrease microstate C coverage (less mind-wandering, less self-referential chatter). Open-monitoring meditation, by contrast, can increase microstate C coverage, reflecting the broad, accepting awareness of internal states that characterizes the practice.
These aren't just interesting findings. They're evidence that you can change your brain's microstate dynamics through training. The temporal structure of your consciousness is not fixed. It's malleable.
Flow state research, while still early for microstates specifically, points in a promising direction. The subjective characteristics of flow, intense focus, loss of self-awareness, altered time perception, map neatly onto what you'd predict from microstate theory: increased microstate D (the attention state), decreased microstate C (the self state), and possibly increased temporal stability overall (each state lasting longer, transitions becoming smoother).
Research has identified characteristic microstate signatures for various mental states. At rest with eyes closed, all four microstates appear in roughly balanced proportions. During focused tasks, microstate D increases while C decreases. During meditation, the pattern depends on the style of meditation practiced. In sleep, the microstate landscape transforms dramatically, with slower temporal dynamics and different spatial configurations emerging during each sleep stage. These state-dependent changes make microstates a powerful tool for objectively characterizing what the brain is doing at any moment.
How Does Microstate Analysis Actually Work?
For the technically curious, here's what happens under the hood.
The standard microstate analysis pipeline starts with clean, artifact-free EEG data from multiple channels. The more channels, the better the spatial resolution of the topographic maps, though research has shown that meaningful results can be obtained from reduced montages.
Step 1: Compute global field power (GFP). GFP is a measure of the overall strength of the electric field at each time point, calculated as the standard deviation of voltage across all channels. GFP forms a continuous curve with peaks and troughs.
Step 2: Extract topographies at GFP peaks. The key insight is that at GFP peaks, the signal-to-noise ratio is highest and the topographic maps are most clearly defined. You extract the voltage pattern across all electrodes at each GFP peak.
Step 3: Cluster the topographies. Apply a clustering algorithm (typically modified k-means or atomize-agglomerate hierarchical clustering) to group similar topographies together. This is where the four canonical maps emerge from the data.
Step 4: Back-fit the maps. Assign every time point in the original EEG to the cluster map it most closely resembles. This produces a continuous sequence of microstate labels: A, B, C, D, A, D, C, A, B, and so on.
Step 5: Compute statistics. Calculate duration, occurrence, coverage, and transition probabilities for each class and for the overall sequence.
The elegance of this approach is that the maps aren't imposed on the data. They emerge from it. You let the clustering algorithm discover the most common topographic configurations, and it keeps finding the same four. That's not a design choice. That's what the brain does.
Classic microstate research uses 64 or 128 channel EEG systems. But a 2019 study by Khanna and colleagues found that microstate analysis with as few as 8 well-placed electrodes can recover the canonical microstate classes with reasonable accuracy. The key is spatial coverage, having electrodes that sample frontal, central, and posterior regions, rather than sheer channel count. This finding opens the door to microstate analysis with consumer-grade EEG devices that cover the right scalp regions.
From Research Labs to Your Living Room
Microstate analysis started as an esoteric tool in academic EEG labs. For decades, it required expensive high-density EEG systems, specialized analysis software, and deep expertise. But two trends are converging to change that.
First, the computational tools have become accessible. Open-source packages like the Cartool software from Lehmann's own group, plus Python libraries and EEGLAB plugins, have democratized microstate analysis. If you have multi-channel EEG data, you can run microstate analysis on a laptop.
Second, consumer EEG hardware has reached the point where meaningful topographic analysis is possible. You don't need 128 channels to capture the fundamental spatial structure of microstates. You need enough channels, placed in the right locations, to distinguish between the four canonical topographic orientations.
The Neurosity Crown's 8 channels sit at positions CP3, C3, F5, PO3, PO4, F6, C4, and CP4. That's coverage of frontal regions (F5, F6), central regions (C3, C4), centroparietal regions (CP3, CP4), and parietal-occipital regions (PO3, PO4). This distribution spans all four quadrants of the scalp, which is precisely what you need to differentiate the orientations of microstates A through D.
With the Crown's raw EEG access at 256Hz and open SDKs in JavaScript and Python, researchers and developers can build microstate analysis pipelines that run on real-time data. Combined with tools like BrainFlow and Lab Streaming Layer (LSL), the Crown becomes a platform for exploring brain state dynamics outside the confines of a clinical lab.
This isn't a replacement for high-density research systems. Those systems provide spatial resolution that 8 channels can't match. But for tracking your own microstate dynamics during work, meditation, sleep, or focus training, for building the feedback loops that could one day let you train your brain's state-switching in real time, 8 well-placed channels are where the practical frontier begins.
The Road Ahead: Where Microstate Research Is Going
Microstate analysis is experiencing a renaissance. After decades as a niche method used primarily by a few European labs, it's now spreading rapidly through the neuroscience community. Several developments are driving this.
Machine learning approaches are being applied to microstate analysis, moving beyond the traditional four-class model to discover more nuanced, data-driven brain state taxonomies. Some recent studies have identified six or seven reliable microstate classes, adding maps E, F, and G to the original four. What these additional states represent is still being worked out.
Real-time microstate classification is becoming feasible, opening the door to neurofeedback paradigms based on microstate parameters. Imagine a system that detects when your brain is spending too much time in microstate C (rumination) and nudges you back toward microstate D (focused attention). That's the kind of closed-loop system that could transform how we think about cognitive training.
Microstate-based biomarkers are moving toward clinical application. The consistency of microstate abnormalities in conditions like schizophrenia has led several research groups to propose microstate parameters as diagnostic or prognostic tools. A simple EEG recording, analyzed for microstate dynamics, could potentially flag early signs of neurological conditions before behavioral symptoms appear.
Cross-modal studies are linking microstate dynamics to behavior, genetics, and other neuroimaging modalities. The goal is to build a complete picture of how these 100-millisecond brain state atoms relate to everything from personality traits to genetic polymorphisms to the structural wiring of the brain.
Four Maps. Ten Switches Per Second. One Consciousness.
Here's what stays with me about EEG microstates.
We spend so much time thinking about consciousness as something continuous, unified, and indivisible. The "stream of consciousness" that William James described over a century ago. And subjectively, that's what it feels like. An unbroken flow.
But the microstate data tells a different story. It tells us that consciousness might be more like a mosaic than a stream. Tiny tiles, each only a tenth of a second long, each representing a fundamentally different configuration of brain activity, assembled so rapidly and so smoothly that the seams vanish.
Four tiles. That's all the brain seems to need. Four basic spatial patterns of electrical activity, cycling at roughly 10 Hz, their sequences and durations shifting to match whatever cognitive demand the moment requires. Language, vision, self-awareness, focused attention. The atoms of thought, combining and recombining in real time to produce everything you've ever experienced.
The technology to observe this process is no longer locked away in university labs. With multi-channel EEG and the right software, anyone can watch their own brain snap between states, measure the dynamics of their attention, and begin to understand the temporal architecture underlying their moment-to-moment experience.
That architecture has been running since before you were born. It will run until you draw your last breath. And until very recently, no one even knew it was there.
Now you do.