Headbands, EEG headphones and at-home stimulation headsets are landing in benefits portals, right next to the gym discount and the meditation app. 🧠 Some of these products have real clinical evidence behind them. Others have a focus score and a nice slide deck. HR copy rarely says which is which.
This piece is a field guide. I lean on the best data I can find, I flag what is vendor-reported, and I keep three categories apart: medical, consumer wellness, and workplace monitoring. They are not interchangeable, and treating them that way is how a perk goes wrong. ⚖️ I find no trial in the sources I searched that tests a neurotech perk the way the best workplace wellness research tests ordinary ones, so some of what follows is my read on what good evidence would look like.
Three kinds of neurotech hide behind one word
The medical end is the easiest to pin down. Flow Neuroscience has FDA premarket approval (PMA P230024) for its FL-100 headset, a prescription tDCS device for adults in a current moderate-to-severe depressive episode who are not treatment-refractory. It launched in the US in September 2026, backed by the 10-week, sham-controlled EMPOWER trial. 💊 IEEE Spectrum reports that the company also sells a headset under the Halo brand, which it calls identical tDCS hardware, as a wellness device with no prescription. Same electrodes, very different claims.
The consumer end runs on EEG. The Muse band uses a four-channel EEG sensor to feed meditation feedback. Neurable’s newest headphones put 12 dry EEG channels in the ear pads to estimate focus. Neither is a treatment. The third category, workplace monitoring, is where the data may travel to someone other than you. 👀
Medical: FL-100 is FDA-approved, prescription-only, and limited to one indication.
Consumer wellness: Muse and Neurable One report scores, not diagnoses, and I find no approval for treating a condition.
Workplace monitoring: fatigue and focus trackers, where the employer may see the output.
Your first question: which of these three is the company actually paying for?
Five signs the perk is working
1. The claim matches the regulatory status
If the perk is FL-100, the legitimate route is a prescription and a clinician evaluation, not a promo code. If it is an EEG band, the claims should sound like relaxation and attention training, not treatment. Mismatch is the loudest red flag in this market. ✅
Approval is still not the end of the argument. Fast Company reports that a 2024 study led by Andre Brunoni found at-home use of the Flow device no better than a placebo of internet browsing, and it describes the approval trial itself as mixed. A good vendor tells you that up front. A mediocre one hopes you never read it.
2. Nobody on the payroll can see your brain data
This is the sign I care about most. Neural data is a new legal category, and the law is patchy. Colorado was first to cover neural data in April 2024, and California added it to its sensitive-data list the same year. A California bill, AB 1883, would restrict employer surveillance tools that collect neural data, and Illinois has HB2984 pending. The Neurorights Foundation reviewed 30 consumer neurotech companies and found that nearly all appeared to have unrestricted access to users’ neural data. 🔒
So a good program shows employers aggregate numbers only, with individual data staying on your device or with a third party. Nita Farahany also points out, per the Center for Democracy and Technology, that Colorado’s law applies only when the data is collected for identification purposes. That is a gap, not a safeguard.
3. Someone ran a control group
Wellness perks look great in observational data and fall apart in randomized trials. The Illinois Workplace Wellness Study randomly assigned programs among 4,834 employees at a large employer and found no significant effects on 40 of 42 first-year outcomes, including medical spending, productivity, health behaviors and self-reported health. It also found that healthier people were the ones who signed up. 📊
That study did not test neurotech. I think it sets the bar anyway: if a vendor cannot describe a comparison group, the success story is probably selection.
Good sign: a waitlist or sham arm, with outcomes measured by someone outside the vendor.
Weak sign: before-and-after scores from people who chose to join.
Ask for: the trial registration number.
4. The score tracks something outside the app
A focus score is a model output. It earns trust when it moves with behavior. Neurable reports that a US Air Force Research Lab collaboration compared its headphones against research-grade gel EEG and that the score dropped when participants were interrupted during a Stroop test, which is a reasonable design for a validation. The company’s own research page also describes a study of 43 participants over 114 hours of office work. Both are vendor-reported, so I treat them as promising, not settled. 🎧 IEEE Spectrum’s reviewer also said getting a good signal was sometimes a struggle, which is a fair reminder that dry sensors have limits.
5. It targets a narrow job, like fatigue
The best-supported workplace use is not “optimizing” everyone. It is detecting drowsiness in roles where a lapse hurts someone. Verdict describes SmartCap’s LifeBand, an EEG headband that scores fatigue from one (hyperalert) to five (involuntary sleep), and IEEE Spectrum notes that a Neurable researcher helped run a 2022 study on measuring drowsiness in truck drivers. A narrow target means a narrow, checkable claim. 🚛
Three signs it is not working
1. Managers see individual dashboards
Some systems send concentration dips to the wearer and the manager. Welcome to the Jungle describes earbuds that notify users and their managers through a workplace wellness app. An Australian legal forum asked whether anyone can freely consent to wearing a device that is a condition of employment, and warned about discrimination against neurodivergent or older workers. A Frontiers paper on the EU position notes the technology can let employers predict performance or behavior from brain data. My reaction is simple: that is surveillance with a wellness label, and people stop trusting the perk. 😬
2. The proof is a tiny vendor study
IEEE Spectrum reports a company blog claim that 67 percent of users showed a focus increase in a 20-participant experiment. That is a marketing number. Without a control group, a registered protocol or independent analysis, it cannot carry a procurement decision. 🧪
3. Success means sign-ups and satisfaction
A randomized trial of 53 novice meditators, published in Mindfulness, compared an app alone with the app plus a Muse headband for a month. Most people meditated three or four times a week and were happy with their device (84% in the app group, 74% in the Muse group). Yet Muse’s EEG scores did not improve, and the app-only group scored worse at follow-up. The authors conclude those scores are not a proxy for mindfulness or mental health. People liked it. That is not the same as it working. 📉
Red flag: satisfaction surveys are the headline result.
Red flag: a dashboard score is the only outcome.
Red flag: the vendor cannot say who can access raw EEG.
What to ask before the next renewal
You do not need a neuroscience degree. You need five questions and the patience to ask them twice. 💡
Which category is this: medical, consumer wellness, or monitoring?
Who can see individual data, and can the vendor put that in writing?
Is there a comparison group, and who measured the outcome?
What does the score correlate with outside the app?
What happens to raw EEG if the contract ends?
If you follow this area of neurotech professionally, NeurotechMag Pro gives you the structured intelligence feed to go deeper than any single article can. For everyone else, the habit that pays off fastest is checking a vendor’s claim against its regulatory status and its trial registration before you wear anything. 🔍
The Illinois result bothers me most because it is the best data we have and it is not about neurotech at all. So here is the question for your next benefits meeting: if the program were cancelled tomorrow, what measurable thing would change?