Can This Quest Add-On Really Read Your Mind for VR Avatars?

03 Juin 2026 •

I’ve been covering this space long enough to remember when “brain-computer interface” meant a lab coat, a cap full of wires, and a graduate student trying not to fall asleep. The promise was always huge: control your digital self with nothing but thought. The reality was usually a blinking cursor and a lot of patience.

So when I saw the headline from PiEEG — a Scotland-based startup claiming to turn your brain signals and facial micro-expressions into real-time avatar control on a Quest headset — I did what any self-respecting cynic would do. I rolled my eyes. Then I clicked.

Because here’s the thing. The idea of a brain-computer interface (BCI) that works with consumer VR hardware has been a tantalizing ghost for years. We’ve seen the demos at trade shows. The carefully controlled environments. The PR videos where someone thinks “open hand” and the avatar obliges. What we haven’t seen is something that actually works in your living room, with sweaty brows and dropped Wi-Fi signals and a cat walking across the keyboard.

PiEEG thinks they’ve cracked it. Or at least, they’ve cracked a version of it that’s cheap enough to sell to researchers, developers, and hobbyists — not just well-funded neuroscience labs. Their new product is a facial interface replacement for Quest headsets. It’s not a whole new headset. It’s an accessory. And that, in my view, is both its biggest strength and its most honest limitation.

The Brains Behind the Brain Reader

PiEEG was founded in 2022 by Dr. Ildar Rakhmatulin. That’s not a long time ago. In startup years, that’s barely out of diapers. But Rakhmatulin has been in the BCI trenches for a while, and his pitch has always been the same: make this stuff cheap enough that anyone can play with it.

Most consumer-grade BCIs are either toys with limited resolution or lab equipment that costs more than a used car. PiEEG wants to sit somewhere in the middle. Their previous products were open-source boards that let you hack together your own EEG setup. Think of it as the Arduino of brain reading. Now they’re trying to cram that same philosophy into a Quest face pad.

The accessory uses dry electrodes — no sticky gel required — to pick up electrical signals from your scalp and face. It’s not reading your thoughts in the sci-fi sense. It’s looking for patterns. Muscle twitches. Micro-expressions. The tiny voltage shifts that happen when you clench your jaw or raise an eyebrow or, yes, think about moving your hand.

But let’s be clear: this is not telepathy. Not yet. Not even close.

What It Actually Does (And Doesn’t)

Here’s where I have to put on my journalist hat and stop the hype train before it leaves the station. PiEEG’s accessory is designed to pick up two types of signals: brain activity (EEG) and facial muscle activity (EMG). The EEG part is what you’d expect — it can detect things like alpha waves, beta waves, and the P300 response (that’s the brain’s “oh, I noticed that” signal). The EMG part is more interesting for immediate avatar control, because facial muscles produce stronger, cleaner signals than brain waves.

What does that mean in practice? You could, theoretically, map a subtle eyebrow raise to a gesture. A clench of your jaw could trigger an emote. A quick blink could toggle something. It’s not mind control — it’s muscle control, plus a side of brain reading for flavor.

And that’s fine. Honestly, I think that’s smarter than promising full telepathic immersion. The BCI field has been burned by overpromising. Remember when everyone thought we’d be typing with our brains by 2020? Yeah. Me too.

What PiEEG seems to understand is that the path to practical BCI in VR is incremental. You don’t start with “think a sentence and the avatar speaks it.” You start with “clench your jaw and the avatar smiles.” That’s achievable. That’s sellable. That’s actually useful for social VR, where current avatars are often stiff, lifeless puppets waiting for a button press.

The Quest for Better Avatars

Let’s talk about why this matters. If you’ve spent any time in VR social apps like VRChat, Horizon Worlds, or Rec Room, you know the problem. Your avatar is a digital mannequin. Sure, you can wave your hands around, and some headsets track your eyes and mouth now. But the face? The subtle micro-movements that make human interaction human? Mostly missing.

Meta has been working on this with their own research. They’ve shown off prototypes that use cameras pointed at your face to infer expressions. It works, but it’s heavy on compute and requires good lighting. PiEEG’s approach is different. Instead of cameras, they’re reading the electrical signals that actually produce the expressions. It’s not visual. It’s physiological.

In my view, that’s a smarter approach for one big reason: it doesn’t care about lighting. You can be in the dark. You can be wearing a mask. You can have a beard that messes with camera tracking. The electrodes don’t care. They’re reading what your muscles are doing, not what they look like.

But there’s a catch. Actually, there are several.

The Reality Check

First, dry electrodes are notoriously finicky. They need good contact with the skin. Hair gets in the way. Movement creates noise. If you’re jumping around in a VR game, the signal quality is going to degrade. PiEEG knows this — they’ve spent years trying to solve it. But physics is a stubborn thing.

Second, the accessory is launching for researchers and developers, not consumers. That’s smart from a business perspective. You don’t want to sell a half-baked BCI to people who expect it to work perfectly out of the box. You sell it to tinkerers who can handle the noise, build their own software, and push the limits. But it also means we’re not getting a plug-and-play mind-reading headset anytime soon.

Third — and this is the big one — there’s the question of what you actually do with the data. Even if you can read brain signals reliably, what does that look like in a VR app? Do you map alpha waves to a color change? Do you use the P300 response to select menu items? These are open research questions. PiEEG is providing the tool. The killer app is still someone else’s job to build.

Where This Fits in the BCI Zoo

If you’ve been following BCI for a while, you know the landscape is a weird mix of ambitious startups and cautious academics. There’s Neuralink, which is aiming for brain implants and has all the hype (and controversy) that comes with Elon Musk’s involvement. There’s NextMind, which made a headband that let you control a computer by focusing on objects — until they were acquired by Apple. There’s OpenBCI, the open-source hardware folks who’ve been at it for years.

PiEEG is closer to OpenBCI than Neuralink. They’re not trying to drill into your skull. They’re not promising to cure paralysis (at least not directly). They’re saying: here’s a cheap, accessible EEG/EMG sensor that straps onto your Quest. Go figure out what to do with it.

I like that approach. It’s honest. It’s humble. And it might actually lead to real innovation, because the people who buy these things are the ones who will find creative uses for them. Researchers will use it to study presence and immersion. Hobbyists will build weird experiments. Indie developers might bake it into a game where your anxiety level changes the environment.

But let’s not pretend this is going to be in every VR headset next year. It won’t. The barrier to entry is still too high. You need to know how to handle noisy signals. You need to write your own software. You need to be comfortable with the fact that sometimes it just won’t work.

The Avatar Control Question

The headline promises “avatar control.” I want to dig into that, because I think it’s the most interesting — and most misleading — part of the pitch.

Can you control an avatar with your brain activity? Technically, yes. You could train a model to associate a certain brain state with a certain action. But it’s slow. It’s unreliable. It requires training and calibration. For real-time avatar control — the kind where you want your digital face to mirror your real face in milliseconds — the EMG signals are going to do the heavy lifting.

That’s not a failure. That’s just the current state of the art. Brain signals are noisy. They’re weak. They’re buried under muscle artifacts and electrical interference from the headset itself. Reading them in real time, on a mobile processor inside a Quest, is a monumental challenge. PiEEG is trying to do it with a few dry electrodes and a tiny board. I’m impressed they’re even attempting it.

But the demo videos will show you a person raising an eyebrow and the avatar doing the same. That’s EMG. That’s facial muscle, not brain. And that’s fine — it’s still cool. But let’s call it what it is: facial expression tracking through electrical sensing, not mind reading.

The Bigger Picture

What struck me here is the timing. We’re in a weird moment for VR. The Quest 3 is solid, but the hype cycle has moved on to AI. The metaverse is a punchline. Apple’s Vision Pro is beautiful and absurdly expensive. The industry needs something new to get excited about.

BCI could be that thing. But it has to be done right. It has to be integrated. It has to work without a PhD in signal processing. PiEEG’s accessory is a step in that direction, but it’s a step for early adopters, not the mainstream.

I think the real value here is in the data. If enough researchers and developers start using this thing, we’ll get a better understanding of what BCI can actually do in VR. We’ll find out which signals are reliable. Which applications are compelling. Where the technology breaks down. That knowledge is worth more than any single product.

So should you buy one? If you’re a researcher, a developer, or a hardcore hobbyist who loves hacking on EEG data, absolutely. If you’re a regular VR user hoping to control your avatar with your thoughts next week, maybe wait. The future of BCI in VR is coming — but it’s arriving in small, incremental steps. This is one of them.

And honestly? That’s okay. I’d rather have a hundred honest, imperfect steps than one more slick demo that disappears after the press cycle.