What Is the BRAIN Initiative?
In 2013, a President Bet $6 Billion That We Don't Understand Our Own Brains
On April 2, 2013, President Barack Obama stood in the East Room of the White House and announced what he called "the next great American project." Not a mission to Mars. Not a new particle accelerator. A mission to understand the three pounds of tissue sitting between your ears.
"As humans, we can identify galaxies light years away," Obama said. "We can study particles smaller than an atom. But we still haven't unlocked the mystery of the three pounds of matter that sits between our ears."
He wasn't exaggerating. In 2013, neuroscience could tell you a lot about individual neurons. It could tell you a lot about what large brain regions did. But the middle ground, how millions of neurons work together in circuits to produce thought, perception, and behavior, was largely a black box. We had maps of the parts and maps of the whole, but almost nothing in between.
The BRAIN Initiative was created to fill that gap. And the story of what's happened since is one of the most consequential, and least talked about, science stories of the century.
The Gap in the Map
To understand why the BRAIN Initiative matters, you need to understand what neuroscience couldn't do before it existed.
By 2013, neuroscience had two powerful but limited perspectives on the brain. At the micro level, researchers could study individual neurons in extraordinary detail. They could record the electrical activity of one neuron, or maybe a few dozen, using fine electrodes inserted into brain tissue. At the macro level, brain imaging tools like fMRI could show which large brain regions activated during different tasks, with spatial resolution of a few millimeters.
But the brain doesn't work at either of those scales alone. Cognition, behavior, consciousness: these emerge from the coordinated activity of millions of neurons working together in circuits. And in 2013, there was no technology that could record from millions of neurons simultaneously while an animal (let alone a human) was behaving naturally.
Think about it this way. Imagine trying to understand how the internet works by reading individual emails one at a time, or by looking at satellite photos of server farms from space. Both perspectives give you real information. Neither one tells you how the network actually functions. For that, you'd need to observe the traffic in real time, across the entire network, at the resolution of individual packets.
That's what the BRAIN Initiative set out to build for the brain. New tools. New technologies. New ways of watching neural circuits in action.
The Name Is a Clue: It Was Always About Tools
BRAIN stands for Brain Research through Advancing Notable Neurotechnologies. That last word is the key. The initiative's founding premise wasn't to study the brain directly. It was to build the instruments that would allow the brain to be studied in ways that were previously impossible.
This was a deliberate strategic choice, and a brilliant one. The planners, led by a committee of top neuroscientists including Cori Bargmann and William Newsome, looked at the history of science and noticed a pattern. The biggest breakthroughs tend to follow the invention of new tools, not new theories. The telescope made modern astronomy possible. The microscope made biology possible. PCR (polymerase chain reaction) made modern genetics possible. The sequencing technologies behind the Human Genome Project didn't just accelerate genomics. They created it.
The BRAIN Initiative bet that the same thing was true for neuroscience. Build better tools for observing the brain, and the discoveries will follow.
Over a decade later, that bet has paid off spectacularly.
The Toolbox: What the BRAIN Initiative Built
The technologies that have come out of this initiative are genuinely remarkable. Here are the ones that have changed the game.
Neuropixels: 10,000 Neurons at Once
Before the BRAIN Initiative, the standard tool for recording individual neurons in animals was a simple electrode that could pick up activity from maybe 1 to 10 neurons at a time. If you wanted to record 100 neurons, you needed 100 electrodes, each one painstakingly positioned by hand.
Enter Neuropixels, a silicon probe about the width of a human hair that contains nearly 1,000 recording sites along its length. Insert one of these probes into an animal's brain, and you can simultaneously record from hundreds of neurons across multiple brain regions. Insert several probes, and you can record from 10,000 or more neurons at once.
In 2005, a typical neuroscience recording experiment might capture data from 10 to 50 neurons. By 2025, thanks to Neuropixels and similar technologies funded by the BRAIN Initiative, researchers routinely record from 10,000 neurons simultaneously in behaving animals. That's a 200-fold to 1,000-fold increase in just 20 years.
Neuropixels were developed with BRAIN Initiative funding at the Howard Hughes Medical Institute's Janelia Research Campus and are now manufactured by the company IMEC. They cost a fraction of what previous recording systems cost, and they've been adopted by hundreds of labs worldwide. The data flowing out of Neuropixels experiments is rewriting our understanding of how neural circuits work.
Brain Cell Atlases: Cataloging Every Cell Type
The BRAIN Initiative Cell Census Network (BICCN) set out to do something that sounds deceptively simple: catalog every type of cell in the brain. It turned out to be one of the most ambitious biological surveys ever attempted.
The human brain contains roughly 86 billion neurons, but those neurons aren't all the same. They differ in their gene expression, their shape, their electrical properties, and their connections. How many distinct types are there? Before the BRAIN Initiative, estimates ranged from a few hundred to maybe a thousand. Nobody really knew.
The BICCN used single-cell RNA sequencing, a technique that reads the genes active in individual cells, to classify brain cells by their molecular identity. The results, published in a landmark series of papers in 2023, identified over 3,000 distinct cell types in the human brain. More than three thousand. Each with its own molecular signature, its own role in the circuitry, its own potential relevance to disease.
This atlas is to neuroscience what the periodic table is to chemistry. It's the foundation everything else gets built on.
Advanced Optogenetics and Voltage Indicators
Optogenetics, the technique of using light to control genetically modified neurons, existed before the BRAIN Initiative. But the initiative funded its rapid expansion and refinement. New opsins (light-sensitive proteins) were developed that respond to different wavelengths, activate faster, inhibit rather than excite, and can be targeted to specific cell types with surgical precision.
Even more importantly, the BRAIN Initiative funded the development of genetically encoded voltage indicators (GEVIs), proteins that flash when a neuron fires. Combined with advanced microscopy, these tools let researchers watch thousands of individual neurons fire in real time in a living brain, without any electrodes at all.
The Fruit Fly Connectome
In 2024, a consortium funded partly by the BRAIN Initiative published the complete connectome of the Drosophila (fruit fly) brain. Every one of its approximately 139,000 neurons. Every one of its roughly 50 million synaptic connections. It was the first complete wiring diagram of an insect brain, and it took years of automated electron microscopy and AI-assisted tracing to complete.
Why does a fly brain matter? Because the fly brain, despite having 600,000 times fewer neurons than the human brain, uses many of the same fundamental circuit motifs. Feedback loops, lateral inhibition, parallel processing pathways: they're all there. The fly connectome is a Rosetta Stone for understanding how neural circuits compute.
The Number That Explains Everything: $6 Billion
| Funding Source | Approximate Investment | Focus Areas |
|---|---|---|
| NIH | Over $4 billion | Cell atlases, connectomics, new recording tools, clinical translation |
| DARPA | Over $500 million | Brain-computer interfaces, non-invasive neurotechnology (N3 program) |
| NSF | Over $200 million | Fundamental neuroscience, computational models, data infrastructure |
| Private Foundations | Over $1 billion | Allen Institute, HHMI, Kavli Foundation, Simons Foundation |
| Industry Partners | Varies | Technology development, commercialization, open data platforms |
This scale of investment has created a flywheel effect. More funding means better tools. Better tools mean bigger datasets. Bigger datasets attract more researchers. More researchers generate more discoveries. More discoveries justify more funding. The BRAIN Initiative has become a self-reinforcing engine of neuroscience progress.
The Part Nobody Expected: It Changed How Science Gets Done
Here's the thing about the BRAIN Initiative that doesn't get enough attention.
Beyond the specific technologies and discoveries, the initiative has fundamentally changed the culture of neuroscience. And that cultural shift might matter more than any individual tool.
Open Data as Default
The BRAIN Initiative adopted a radical (for neuroscience) stance on data sharing: all data funded by the initiative must be made publicly available. This wasn't the norm before. Historically, neuroscience labs would generate data, publish a paper about it, and then the data would sit on a hard drive in someone's office forever.
Now, terabytes of neural recording data, brain imaging data, cell atlas data, and connectomics data are freely available through platforms like the DANDI Archive and the Brain Cell Data Center. Any researcher, anywhere in the world, can download and analyze data that cost millions of dollars to collect. Graduate students in Lagos can work with the same datasets as professors at MIT.
Team Science
The BRAIN Initiative has also pushed neuroscience toward larger, more collaborative teams. Many of its flagship projects, like the cell atlas and the connectomics efforts, involve dozens of labs working together, sharing protocols, standardizing methods, and cross-checking results. This is common in physics (particle physics experiments routinely have hundreds of authors) but was unusual in neuroscience, where the single-PI lab has been the standard unit of science for generations.
Computational Infrastructure
Perhaps most importantly, the BRAIN Initiative recognized early that the data revolution it was creating would require new computational tools. It has funded the development of open-source software for analyzing neural data, cloud platforms for storing and sharing datasets, and standards for describing brain data that make it possible to combine results across studies.
This infrastructure is invisible but essential. It's the difference between having a mountain of data and having usable knowledge.
The Global Brain Race
The BRAIN Initiative didn't happen in a vacuum. It triggered (or accelerated) a global wave of brain research investments that, collectively, represent the largest coordinated effort to understand any single organ in the history of science.
The European Union launched the Human Brain Project in 2013 (the same year as the BRAIN Initiative) with 1 billion euros in funding, focused on computational brain simulation. Japan launched Brain/MINDS in 2014, focusing on the marmoset brain as a primate model. China announced the China Brain Project in 2016, a massive effort covering basic neuroscience, brain-inspired AI, and brain disease treatment. South Korea, Australia, Canada, and Israel have all launched their own national brain research programs.
The total global investment in these brain initiatives now exceeds $10 billion. We are living through the largest investment in brain science in human history, and most people have no idea it's happening.
Why This Matters Outside the Lab
You might be reading this and thinking: that's impressive, but what does it mean for me? I don't have a Neuropixels probe. I'm not mapping fly brains.
The answer is that the BRAIN Initiative is building the scientific foundation that makes consumer brain technology possible. Every advance in our understanding of neural circuits, signal processing, electrode materials, and brain data analysis ripples outward from the research lab into the real world.
Consider the journey of EEG. In the 1920s, it was a laboratory curiosity. In the 1960s, it became a clinical diagnostic tool. In the 2000s, it began to shrink into consumer devices. Each step along that journey was enabled by fundamental research, the kind of research the BRAIN Initiative funds at unprecedented scale.
The initiative's investments in miniaturized neural recording technology are directly relevant to the consumer EEG devices that exist today. DARPA's N3 program, funded as part of the BRAIN Initiative, explicitly aims to develop non-invasive BCIs with the performance of implanted devices. The signal processing algorithms that clean EEG data, separate brain signals from noise, and extract meaningful cognitive states are being refined by researchers whose work is funded by this initiative.
The Neurosity Crown, with its 8 EEG channels at positions CP3, C3, F5, PO3, PO4, F6, C4, and CP4, sampling at 256Hz with on-device processing via the N3 chipset, represents the downstream end of this research pipeline. The neuroscience that tells us which electrode positions capture the most informative signals, the signal processing that separates brain activity from artifacts, the machine learning that translates raw EEG into focus scores and calm scores: all of this builds on decades of fundamental research that the BRAIN Initiative is now accelerating dramatically.
The path from federal research to consumer brain technology follows a clear trajectory:
- Fundamental neuroscience reveals which brain signals matter (funded by NIH through the BRAIN Initiative)
- Sensor technology gets smaller, cheaper, and more sensitive (funded by DARPA and NSF)
- Signal processing algorithms improve artifact rejection and feature extraction (developed in BRAIN Initiative-funded computational neuroscience labs)
- Open data and standards allow developers to build on research-grade science (mandated by BRAIN Initiative data sharing policies)
- Consumer devices bring these advances to anyone who wants to understand their own brain
The Next Decade: What's Coming
The BRAIN Initiative isn't done. The plan extends through at least 2035, with several major milestones on the horizon.
A complete mouse connectome at synaptic resolution is expected by the late 2020s. This will be the first complete wiring diagram of a mammalian brain, providing an unprecedented reference for understanding how complex neural circuits are organized.
Real-time, whole-brain imaging in behaving animals is getting closer. New microscopy techniques funded by the initiative can already image tens of thousands of neurons simultaneously. The goal is to scale this to millions, capturing the activity of an entire brain as an animal navigates, decides, and learns.
Precision psychiatry based on brain circuit data is moving from concept to clinical trials. The idea is to diagnose and treat mental illness based on objective measurements of brain circuit function, rather than subjective symptom checklists. The BRAIN Initiative's cell atlases and connectivity maps are providing the baseline data that makes this possible.
Next-generation non-invasive BCIs are a stated goal of the initiative. DARPA's N3 program has funded several teams working on non-invasive neural interfaces with dramatically higher resolution than current EEG, potentially bringing us closer to the science fiction vision of high-bandwidth brain-computer communication without surgery.
The Most Important Thing the BRAIN Initiative Has Taught Us
After more than a decade and billions of dollars, what's the single most important thing we've learned?
It might be this: the brain is more organized than we thought, and more dynamic than we imagined.
The cell atlases revealed that the brain's cellular diversity follows precise, reproducible patterns. The connectomics data shows that neural wiring isn't random. It's structured into motifs and modules that repeat across species. The large-scale recording technologies show that neural activity flows through circuits in highly organized patterns that change on timescales of milliseconds.
But within that organization, there's breathtaking dynamism. The functional connectivity between brain regions shifts moment to moment. Neural populations can rapidly reorganize which information they encode. The brain isn't a fixed machine. It's more like a jazz ensemble: structured enough to be coherent, flexible enough to improvise.
Understanding that balance between structure and flexibility is, in a sense, the core mission of the BRAIN Initiative. And it's the core challenge for anyone building technology that interfaces with the brain.
We are in the early chapters of the story of understanding the human brain. The BRAIN Initiative didn't finish that story. What it did is something arguably more important. It built the tools that will let us write the next chapters. And for the first time, those tools aren't locked away in billion-dollar laboratories. They're becoming small enough, affordable enough, and open enough that the next breakthrough might come from anyone with curiosity and a connection to their own neurons.