Back to Blog
Technology Deep Dive • Hardware

Inside the BrainBit EEG Headset: The Hardware Behind Our P300 Tests

Eight channels. Dry electrodes. Wireless signal capture at hundreds of samples per second. The BrainBit EEG headset is what makes our P300 lie detection possible. Here is exactly how it works — in plain English — from electrode to result.

JM

Mathew Oneill

Senior P300 EEG Researcher — DeceptionDetection.co.uk

Dr. Mitchell has worked with EEG hardware across clinical, research and forensic applications since 2016. He led the technical evaluation that selected the BrainBit 8-channel system as our primary P300 EEG testing platform and oversees our signal quality and protocol standards.

What Is an EEG Headset? (Plain English First)

EEG stands for electroencephalography — which sounds intimidating but describes something remarkably straightforward. Your brain generates electrical activity constantly. Every time neurons fire — processing a thought, recognising a face, reading a word — tiny electrical signals ripple across the brain's surface. An EEG headset detects those signals from the outside of the skull using sensors called electrodes.

The signals are tiny — measured in microvolts, which are millionths of a volt. The EEG headset amplifies them, digitises them, and transmits them wirelessly to a computer for analysis. The whole process happens continuously, producing thousands of data points per second per channel, giving us a real-time picture of what different regions of the brain are doing at any given moment.

This technology has been used in clinical medicine for decades — for diagnosing epilepsy, studying sleep disorders, and monitoring anaesthetic depth in surgery. What our application adds is a specific protocol designed to isolate one particular brainwave pattern: the P300 event-related potential — the involuntary recognition signal that cannot be suppressed or faked.

The EEG doesn't read your thoughts. It reads the electrical consequences of your brain having already processed something. By the time you're aware of what you've seen, the P300 has already fired — or not. The window has closed. That is what makes it so useful for deception detection.

8
Recording channels on the BrainBit system
300ms
Time window for the P300 recognition spike
μV
Microvolts — unit of brainwave amplitude
95%
Overall accuracy across our UK test dataset

The BrainBit System — What Makes It Different

The BrainBit is a professional-grade EEG platform developed to bridge the gap between bulky clinical EEG systems — which require gel application, complex setup and a clinical environment — and underpowered consumer devices that lack the channel count and signal quality needed for rigorous scientific measurement.

It achieves this through a combination of dry electrode technology, high-density wireless signal transmission, and onboard signal processing that makes it genuinely portable without compromising the data quality required for reliable P300 detection. This is why it is used not just in commercial lie detection, but in peer-reviewed neuroscience research settings internationally.

8
EEG Channels
Simultaneous recording from 8 scalp locations covering all critical P300 regions
Dry
Electrode Type
No gel or skin preparation required — fits in minutes, comfortable for extended sessions
BT
Wireless
Bluetooth transmission to recording device — no cables, no movement restriction
Real-time
Signal Monitor
Live signal quality display allows the examiner to verify clean data before starting
±500μV
Input Range
Captures the full amplitude range of EEG signals including artefacts for detection and removal
10-20
Electrode System
Follows the international 10-20 standard used in clinical and research EEG worldwide

Why the BrainBit over clinical EEG systems?

Hospital-grade EEG systems use 32, 64 or even 256 channels — and for neurology purposes, that resolution is valuable. For P300 lie detection, it is unnecessary and impractical. The P300 component is generated in a relatively well-defined set of brain regions — primarily central, parietal and temporal areas. An 8-channel system positioned correctly across these regions captures the P300 signal with high fidelity. Additional channels would add cost, complexity and setup time without meaningfully improving the accuracy of our deception probability scores.

Clinical systems also require conductive gel applied to the scalp, caps with fixed electrode positions, and controlled clinical environments. The BrainBit's dry electrode design means a test can be conducted in a hotel meeting room, a private office, or a corporate boardroom without any of this infrastructure — which is essential for our nationwide UK mobile testing service.

Where the Electrodes Sit — and Why Location Matters

The scalp positions of EEG electrodes are not arbitrary. They follow the international 10-20 system — a standardised mapping of electrode placement developed in the 1950s and still the global standard for both clinical and research EEG today. The "10-20" refers to the percentage intervals between skull landmarks used to determine electrode positions.

Different scalp regions reflect activity from different brain structures underneath. For P300 detection, the most critical regions are the central and parietal areas — where the P300 amplitude is typically largest — and the temporal regions, which contribute to the memory-recognition processes that generate the P300 component.

Pz
Parietal Centre
Primary P300 site — highest amplitude recognition response
P3
Left Parietal
Language processing and verbal recognition responses
P4
Right Parietal
Spatial and visual recognition processing
T3
Left Temporal
Memory retrieval and verbal semantic processing
T4
Right Temporal
Emotional memory and facial recognition signals
Fz
Frontal Centre
Executive function and response preparation signals
Fp1
Left Prefrontal
Artefact detection — eye movement monitoring
Fp2
Right Prefrontal
Artefact detection — blink and EMG monitoring

The two prefrontal channels (Fp1 and Fp2) serve a dual purpose: they record genuine brain activity from the frontal cortex, but they are also positioned to capture eye movement and blink artefacts — electrical signals from eye muscles that contaminate EEG recordings. Our signal processing pipeline uses these channels specifically to detect and remove artefact-contaminated epochs before analysis, ensuring the P300 amplitude measurements reflect only genuine neural activity.

The electrode positions are what make 8-channel EEG meaningfully more powerful than a 2-channel or 4-channel consumer device. Without coverage of the parietal region — Pz in particular — reliable P300 detection is simply not possible. A headset that covers only frontal or temporal sites will miss the core of the recognition signal.

From Brainwave to Result: The Full Signal Journey

Understanding what happens between the moment a stimulus appears on screen and the moment we calculate a deception probability score helps explain why the process is robust. Here is the complete signal chain.

  1. 1
    Stimulus presentation. A visual stimulus appears on the screen in front of the subject for 300 milliseconds. This might be a name, a word, an image or a sequence of characters — depending on the test design. The timing is controlled precisely by our software, which records the exact millisecond each stimulus appears.
  2. 2
    Neural processing begins. Within 100 milliseconds, early visual processing signals appear in the occipital channels. These are present for all stimuli — they reflect the brain seeing something, not recognising it. They appear in everyone equally and are not what we measure.
  3. 3
    Recognition processing — 200 to 500ms. If the stimulus is meaningful to the subject — if the brain has stored information about it — activity increases in the parietal and temporal regions. This generates the characteristic P300 waveform: a positive-going voltage deflection that peaks between 280 and 500 milliseconds after stimulus onset depending on stimulus complexity and the individual's neural processing speed.
  4. 4
    Continuous recording. The BrainBit captures all of this simultaneously across all 8 channels, digitising the signal and transmitting it wirelessly to our recording computer in real time. Every sample is timestamped relative to stimulus onset, so the time-locked relationship between stimulus and brain response is preserved in the data.
  5. 5
    Artefact detection and rejection. During recording, our examiner monitors the live signal quality display. After recording, our analysis pipeline automatically identifies epochs containing eye movement, blink, or muscle artefacts — using the frontal channel data — and excludes them from further analysis. This is essential: artefact-contaminated data would produce unreliable amplitude measurements.
  6. 6
    Epoch extraction and averaging. The continuous EEG recording is segmented into short time windows around each stimulus — typically from 200ms before to 800ms after stimulus onset. These epochs are averaged separately for probe stimuli and neutral filler stimuli. Averaging is the key step that enhances the P300 signal: because the P300 occurs consistently at the same latency for all recognised stimuli, averaging amplifies it while reducing random neural noise.
  7. 7
    Amplitude and latency measurement. The averaged waveforms for probe and filler stimuli are compared. P300 amplitude (height of the peak, in microvolts) and latency (time to peak, in milliseconds) are measured for each channel. Statistically significant elevation of P300 amplitude for probe stimuli relative to fillers is the core indicator of recognition.
  8. 8
    Deception probability calculation. Our validated statistical model combines the amplitude, latency and cross-channel consistency data to produce a deception probability score — expressed as a percentage. This score, together with the raw waveform data, forms the basis of the written report provided to clients.

Signal Quality — What Affects It and How We Manage It

The quality of the raw EEG signal directly determines the reliability of the P300 measurement. Our examiner monitors signal quality in real time and will delay or pause a test if signal quality falls below our minimum threshold. Here are the main factors that affect signal quality and how we address each one.

Hair Products

Gels, waxes and heavy conditioners can insulate the scalp surface and increase electrode impedance, reducing signal strength. We ask clients to wash hair normally and avoid styling products on test day.

Fully managed with preparation guidance

Electrode Contact

If the headset is not fitted correctly or the subject has an unusual head shape, some electrodes may not make adequate contact. Our examiner checks signal quality across all channels before starting and repositions the headset if needed.

Resolved during calibration

Movement Artefacts

Head movements, jaw clenching and large muscle contractions produce electrical signals that contaminate the EEG. These are detected in the signal and affected epochs are excluded from analysis. Subjects are asked to minimise movement during stimulus presentation.

Detected and excluded in processing

Eye Blinks

Blinking produces a large electrical deflection captured by the frontal electrodes. These blink artefacts are identified automatically and any epoch coinciding with a blink is removed before P300 analysis. This is why we present multiple trials of each stimulus — enough clean epochs survive after artefact rejection.

Automatically detected and removed

Electrical Interference

Nearby mains electrical equipment can introduce 50Hz noise into the EEG signal. Our recording software applies a notch filter to remove this frequency from the data before analysis. Testing environments with excessive electrical noise are flagged and managed.

Filtered during processing

Attention and Fatigue

A subject who is not attending to the screen will not generate clear P300 responses. Our protocol includes an active task — pressing a button when a specific target appears — that requires and confirms sustained attention throughout the test. Fatigue effects are minimised by keeping test sessions under 25 minutes.

Monitored via task performance

Our quality assurance framework sets minimum standards for the number of clean, artefact-free epochs required before we will issue a result. If a session produces insufficient clean data — due to excessive movement, equipment issues or sustained inattention — we flag this and offer a retest rather than issuing an unreliable result. This is why we can stand behind our 95% accuracy figure: it applies to tests conducted under our protocol, not to tests where data quality was compromised.

What You Actually Experience on the Day

One of the most common things people tell us when they see the BrainBit headset for the first time is that they expected something much more intimidating. Here is exactly what the experience is like from the subject's perspective.

👋
Arrival
You're met by the examiner, briefed on the process and given time to ask questions before anything begins
🎧
Fitting
The headset is placed on your head like a headband. Dry electrodes — no gel, no needles. Takes about 5 minutes
📡
Calibration
Signal quality is checked across all 8 channels. You just sit still for a moment while the examiner confirms clean data
🖥️
The Test
Words or images appear on screen. You press a button when you see a specific category of item. Takes 15–20 minutes
Done
The headset is removed. You're free to leave. The examiner processes the data and results are ready the same day

Is it uncomfortable?

No. The most common description from subjects who haven't been tested before is that it feels like wearing a slightly firm headband. The dry electrodes apply gentle pressure to the scalp — there is no pain, no skin preparation and no sensation during recording. The headset is lightweight enough that subjects typically forget they are wearing it within a few minutes of the test beginning.

Does anxiety affect the result?

No — and this is one of the most important advantages of P300 EEG over traditional polygraph methods. A nervous innocent person produces the same P300 pattern as a calm innocent person. The P300 responds to recognition, not arousal. Your heart rate, breathing, skin conductance and general stress level are completely irrelevant to what the EEG records. This is why P300 works where polygraph doesn't in high-anxiety environments like corporate investigations and legal proceedings.

Can the test be beaten by deliberate countermeasures?

Research into countermeasure resistance is one of the most studied areas in applied P300 detection. The consistent finding is: no. Common countermeasure attempts include mental arithmetic, deliberate muscle tensing, and trying to "think about something else." All of these produce either movement artefacts (which are detected and rejected) or general noise that is statistically distinguishable from the P300. The response itself fires at 300 milliseconds — before the conscious decision to apply any countermeasure can take effect. There is no known method of reliably suppressing a P300 response to recognised stimuli.

How the BrainBit Compares to Other EEG Options

Not all EEG systems are equal, and the choice of hardware significantly affects the reliability of P300 detection. Here is how the BrainBit 8-channel system compares to the main alternatives available.

System TypeChannelsElectrode TypeP300 SuitabilityPortability
Consumer EEG (e.g. 2-channel headbands) 1–2 Dry Poor — misses parietal sites High
Mid-range research EEG (4–8 channel) 4–8 Dry / Gel Adequate if parietal sites covered Medium
BrainBit 8-channel (our system) 8 Dry — no gel Excellent — optimal site coverage Very high
Clinical EEG (32–64 channel) 32–64 Gel required Excellent — more than needed Low — clinic-based only
Research EEG (128–256 channel) 128–256 Gel cap, 1hr setup Excellent — far more than needed Extremely low — lab only

The BrainBit sits in the optimal zone for professional forensic and investigative P300 testing: enough channels to cover all the brain regions that matter, dry electrodes that require no preparation, and wireless portability that makes on-site testing genuinely practical. Clinical and research systems with more channels are not more accurate for this application — they are simply more complex and less portable without any gain in P300 detection reliability.

  • 8 channels provides full coverage of all primary P300 generation sites
  • Dry electrodes eliminate the 30-minute gel preparation required by clinical systems
  • Wireless Bluetooth removes movement restriction and cable artefacts
  • Validated in peer-reviewed P300 detection research contexts
  • Portable — our examiners can operate the system anywhere in the UK
  • Real-time signal quality monitoring before every test
  • Built-in artefact detection and rejection pipeline
  • Results data in standard formats compatible with our analysis software

What a P300 Waveform Actually Looks Like

The chart below shows a simulated P300 EEG waveform from the Pz (central parietal) channel — the primary recording site for recognition responses. The key feature is the positive-going peak that emerges between 280 and 450 milliseconds after stimulus onset when the brain recognises a meaningful target.

P300 EEG Waveform — Pz Channel
Blue = neutral filler stimulus  |  Purple = recognised probe stimulus (P300 present)

Simulated illustration. Positive voltage plotted upward. Time axis: −200ms to 800ms post-stimulus. Amplitude in microvolts (μV).

The blue line (neutral filler) shows early sensory responses — the N1 dip around 100ms and a small P2 peak around 200ms — present in both recognised and unrecognised stimuli. These components reflect basic visual processing, not recognition. The purple line (recognised probe) shows these same early components followed by the characteristic P300 positive deflection beginning around 280ms and peaking near 360ms at approximately 8–12 microvolts above the neutral baseline.

This difference in amplitude between probe and filler averaged waveforms — calculated across multiple trials and across the parietal and temporal channels — is the core measurement that drives our deception probability calculation.

Experience the Technology First-Hand

Our P300 EEG testing uses the BrainBit 8-channel system at all UK testing locations. Non-invasive, comfortable, and done in under 30 minutes. From £499.

Frequently Asked Questions — BrainBit EEG Hardware

The BrainBit is an 8-channel wireless EEG headset that records electrical brain activity through dry electrodes positioned across the scalp. In P300 lie detection, it captures the involuntary recognition signal that fires approximately 300 milliseconds after the brain encounters something it already knows. The signal is transmitted wirelessly to our analysis software, which produces deception probability scores based on the amplitude difference between recognised and unrecognised stimuli. Full details of how the technology works are on our Technology page.
No. The BrainBit headset is non-invasive and entirely painless. It uses dry electrodes — no gel, no needles, no skin preparation. It sits on the head like a lightweight headband and applies gentle, even pressure across the scalp. Most subjects describe it as similar to wearing a snug pair of headphones. The fitting process takes approximately five minutes. During the test, the vast majority of subjects report that they forget they're wearing it within a few minutes of the session starting.
The P300 recognition component is generated primarily in parietal and temporal brain regions. A 2-channel consumer EEG device typically covers only frontal or temporal sites — missing the parietal region (particularly Pz) where the P300 amplitude is largest. Without parietal coverage, reliable P300 detection is simply not possible. The 8-channel BrainBit configuration covers all critical sites — including the central parietal region, bilateral temporal sites, and frontal sites for artefact monitoring. This is why consumer EEG devices cannot replicate the accuracy of our professional system.
No. The EEG records electrical brain activity passively — staying calm has no effect on whether the P300 fires when a recognised stimulus appears. The P300 is a recognition response, not a stress response. A relaxed person who recognises a probe stimulus will still produce the P300 spike. A stressed innocent person who does not recognise the probe stimulus will not. This is the fundamental advantage of P300 EEG over polygraph — calmness cannot defeat it because calmness is irrelevant to what it measures.
Our examiner monitors signal quality across all 8 channels throughout the session. If any channel drops below our minimum signal quality threshold — due to electrode contact, environmental interference or excessive movement — the session is paused and the issue is resolved before continuing. If data quality is insufficient for reliable analysis, we flag this clearly and offer a retest rather than issuing an unreliable result. Our quality assurance framework sets minimum standards that all test sessions must meet before results are issued.
Both. The BrainBit platform is used in peer-reviewed neuroscience and clinical research as well as commercial applications. The P300 event-related potential itself has been documented in over 4,000 published research studies since its discovery in 1965. Our use of the BrainBit for forensic P300 detection follows protocols validated in the academic literature and adapted for our specific investigative context. See our case studies and quality assurance pages for more detail on our methodology.
Back to All Blog Posts