Decoding Neural Activity in Sulcal and White Matter Areas of the Brain to Accurately Predict Individual Finger Movement and Tactile Stimuli of the Human Hand

Try this: hold your hand perfectly still, and just think about moving your index finger.

Don’t actually move it. Just make the firm decision to do so.

In the milliseconds before you even realize you’ve made that choice, a microscopic storm erupts inside your head. Billions of neurons start firing across different regions of your brain, racing through complex neural pathways to carry information about your intention, your spatial awareness, and your motor control.

For decades, neuroscientists have been trying to eavesdrop on this exact conversation. The holy grail has always been a deceptively simple question: Can we read these electrical signals accurately enough to know exactly which finger a person wants to move — or which finger someone is touching?

Thanks to a massive leap in both neural recording technology and machine learning, the answer is starting to look like a resounding yes. But the most fascinating part isn’t just that we can decode these signals. It’s where we are finding them.

It turns out, we’ve been looking at the brain a bit too superficially. Literally.

Looking Beneath the Surface

When we picture brain activity, we usually imagine the cerebral cortex — the wrinkled, walnut-like outer layer of the brain. Naturally, this is where most traditional Brain-Computer Interface (BCI) systems have focused their attention. It’s right there on the surface, making it the easiest place to place sensors.

But the human brain isn’t a flat sheet of paper. It’s packed with deep, hidden crevices known as sulci, and under that lies a massive, tangled web of white matter connecting everything together.

Historically, these deep folds and pathways were placed in the too hard to reach or not important enough categories for neural recording. But recent breakthroughs are proving that these overlooked regions are actually goldmines of biological data.

Take the sulci, for example. The neural populations that control our hands aren’t just sitting neatly on the surface; they dive deep into these cortical folds. These hidden pockets are densely packed with neurons responsible for motor planning, finger-specific control, and processing touch. We are now realizing that trying to decode hand movement without looking into the sulci is like trying to listen to a symphony from the parking lot outside the concert hall. You’re missing the nuances.

The White Matter Revelation

Then there’s white matter. For the longest time, white matter was treated like the brain’s dumb plumbing — passive cables whose only job was to shuttle electrical signals from point A to point B.

That view is rapidly becoming obsolete.

Modern recording techniques are revealing that white matter actually contains rich, active neural signatures. It reflects the real-time communication happening between our sensory and motor networks. Instead of just being the wire, it’s a highway full of readable data about how distributed brain circuits coordinate something as complex as a hand gesture. Tapping into this highway opens up entirely new frontiers for neural interfaces.

Decoding the Millimeters

Understanding that someone wants to move their arm is one thing. Distinguishing whether they want to move their middle finger or their ring finger is a completely different nightmare.

From a neurological standpoint, the brain signals for moving adjacent fingers overlap massively. Yet, this is where machine learning steps in. By feeding these complex, noisy signals from the sulci and white matter into advanced AI models, algorithms are learning to spot the incredibly subtle, microscopic differences in brain activity.

We can now train algorithms to recognize the unique neural fingerprint of a thumb movement versus a pinky movement. In some recent studies, the prediction accuracy has reached levels that neuroscientists would have called science fiction just ten years ago.

The Two-Way Street of Touch

But human hands aren’t just tools for grabbing things; they are some of the most sophisticated sensory instruments on the planet. Movement is only half the equation.

Every time you brush your fingers against a coffee mug or type on a keyboard, distinct activation patterns fire back up to the brain. Researchers are now building systems capable of decoding this incoming traffic. They can identify exactly which finger was touched, when it happened, and even the characteristics of the texture.

This is a game-changer. It means the neural interfaces of the future won’t just be one-way streets where the brain barks orders at a computer. They will be closed loops capable of reading motor commands and understanding sensory feedback.

What This Actually Means for Our Future

This isn’t just an academic flex; it’s a lifeline.

For people living with paralysis, spinal cord injuries, or severe neurodegenerative diseases, the ability to translate a mere thought into a highly specific, actionable command is life-altering.

Imagine a robotic prosthetic that doesn’t just lazily open and close its fist, but can type on a keyboard because the BCI accurately reads the user’s intent for every individual digit. Now imagine that same prosthetic sending tactile feedback back to the brain so the user can actually feel what they are holding.

By digging deeper into the brain’s sulci and white matter, and letting AI translate the noise, we are making these systems infinitely more accurate. And in the world of prosthetics, accuracy equals natural movement, and natural movement equals independence.

The Road Ahead

Of course, we aren’t completely there yet. The brain is incredibly noisy, everyone’s anatomy is slightly different, and getting access to these deep regions still largely requires invasive surgery. On top of that, as our ability to read the brain improves, we are going to have to have some very uncomfortable, very necessary conversations about neural privacy and data ownership.

But the trajectory is undeniable.

The human brain has been broadcasting these incredibly complex, high-definition signals since the dawn of our species. The problem was never the transmitter; the problem was that we didn’t know how to build the right receiver.

Now, for the first time, we aren’t just hearing the noise. We are finally learning how to listen.