Back to Blog
Field Report • London • April 2026

Three Cases, One Day in London

The corporate case was booked first — the other two came in the week before. By the time the train pulled into Euston at 7:43am, I had a data theft investigation in the City, an infidelity dispute in Islington, and an estate accusation in Southwark to get through before I could go home. This is how the day went.

JM

Dr. James Mitchell

Senior P300 EEG Researcher — DeceptionDetection.co.uk

Dr. Mitchell travels to London regularly for corporate and private P300 EEG testing sessions. All cases in this report are anonymised. Names, identifying details and some minor contextual information have been changed to protect client confidentiality.

9:15am
📍 The City, EC2
Corporate — data breach, 2 subjects
1:30pm
📍 Islington, N1
Relationship — infidelity accusation, 1 subject
4:45pm
📍 Southwark, SE1
Estate — 2 siblings, financial dispute
1

The City: A Data Breach That HR Couldn't Resolve

9:15am — Financial services firm, EC2 — Two subjects tested

The meeting room was on the fourteenth floor. Floor-to-ceiling glass, the dome of St Paul's in the middle distance, the kind of office where everything is designed to project confidence. The HR director who let me in was not projecting confidence. She had been dealing with this investigation for eleven weeks.

A significant volume of client data had been accessed and, the firm believed, exfiltrated from their systems over a period of three months. The IT security investigation had identified two employees — both senior, both long-serving — whose access credentials had been used during the relevant periods. Both denied any involvement. The internal investigation had hit a wall. Both employees knew they were under suspicion. Both had retained legal counsel. Formal disciplinary proceedings had stalled because there was not enough evidence to proceed and not enough confidence to drop it.

The P300 EEG test had been proposed by the firm's external legal advisors as a voluntary step. Both employees had agreed — in writing, with their solicitors' knowledge — to participate. The agreement took three weeks to negotiate. By the time I arrived, everyone in the room had been carrying this situation for long enough that the atmosphere had that particular quality of exhausted professionalism where people are very polite and very tense simultaneously.

The test design

The probe stimuli were built around specific technical details of the exfiltration method — the particular software tool used, the destination file path structure, a specific naming convention found in the extracted files, and the email domain of the external recipient identified in the IT logs. None of these details had been shared publicly within the firm. The IT security team confirmed they had been disclosed only within the core investigation group.

A person who had not been involved in the data exfiltration had no means of knowing these details. A person who had been involved would have stored them.

The results

⚠️ Subject A — Deception Indicated — 91.4%
✅ Subject B — No Deception Indicated — 7.2%

Subject A produced a strong, unambiguous P300 recognition response to the software tool name, the naming convention, and the external email domain. Subject B produced no meaningful recognition response to any probe stimulus across any of the eight recording channels. The amplitude difference between the two profiles was among the clearest I have recorded in a corporate setting — 14.1 microvolts at Pz for Subject A's probe average versus 2.3 microvolts for Subject B.

I delivered the verbal results to the HR director and her legal advisor. The HR director said very little. The legal advisor made two notes. Then we discussed the written report timeline and I was back in the lift by 11:50am.

Field note

The thing about corporate investigations is that a clear result — someone who is not guilty — is often more immediately useful than a deception-indicated one. Subject B had been living under this suspicion for eleven weeks alongside Subject A. Now they had a clean neurological record and a signed examiner's report to prove it. That matters.

2

Islington: The Long Accusation

1:30pm — Private appointment, N1 — One subject tested

The second appointment was in a serviced office suite near Upper Street. The client had found us online three months ago and had been working up to booking ever since. I know this because she mentioned it in the first two minutes — almost apologetically, as though she needed to explain why it had taken this long to get here.

She and her partner had been together for eight years. The suspicion had started about fourteen months ago — a specific name, appearing more often than she felt comfortable with, in his phone, in conversations, in small ways she struggled to articulate separately but couldn't stop adding up together. He denied anything beyond friendship. She believed him and then didn't and then believed him again. The cycle had been going on long enough that she wasn't sure anymore which version of reality she was trying to confirm.

He had agreed to the test. He was not at the appointment. This was her choice — she wanted to understand the process before he came in, though in practice the pre-test work had already been done in a call the previous week and this was the test itself. He arrived twenty minutes after I did, shook my hand, sat down, and said: "Let's get this done."

The test design

Relationship tests are the most sensitive to design quality. The probe stimuli were built around specific, verifiable details — not general suspicion. The specific name, embedded among other plausible names. Two locations she believed were significant, embedded among similar locations. A specific time pattern she had documented. These were things that only someone with relevant knowledge would encode distinctly.

The result

✅ No Deception Indicated — 6.8%

No meaningful P300 recognition response on any probe stimulus. The amplitude profiles for probe and filler stimuli were statistically indistinguishable. His brain did not recognise the specific details that would have indicated relevant stored knowledge.

I delivered the result to them both together. She asked me to explain the probability score again. I did. Then she asked a third question, which was not really a question about the result — it was the question she had probably been carrying for fourteen months about how she had got here. I don't have a test for that one. I gave her our post-test support resources and the number for Relate, and left them to talk.

Field note

A clear result in a relationship test doesn't fix the fourteen months. It removes the factual question at the centre of them. That's not nothing — it's actually quite significant. But the space it creates still has to be filled with something, and that's not our job. We can only remove the uncertainty. What happens in the absence of it is theirs.

3

Southwark: Two Siblings and Their Mother's Estate

4:45pm — Private appointment, SE1 — Two subjects tested

The third appointment was the one I had been thinking about on the tube between Islington and London Bridge. Estate disputes are different from the other case types. In a corporate investigation, the relationship between the parties is professional. In a relationship test, the parties usually want to stay together. In an estate dispute, you are dealing with people who are grieving and angry at the same time, who are fighting over something that will never compensate them for the thing they've actually lost, and who may not speak to each other again regardless of what the result shows.

The two siblings — a brother and sister, late forties — had not spoken directly since their mother's funeral four months ago. Their solicitors had been communicating. The dispute centred on approximately £40,000 that had been transferred from their mother's account in the eight weeks before her death. The daughter alleged the son had used his power of attorney to make the transfers to benefit himself. The son denied this, claiming the transfers were at their mother's explicit instruction for legitimate reasons he was not at liberty to discuss in detail due to her expressed wishes regarding privacy.

They had agreed — again, through solicitors — to attend the same appointment but be tested separately. They arrived at different times. They did not share the waiting area. The solicitors had been clear that the results would be shared with both parties and could be used in the civil proceedings that were pending.

The test design

This was the most technically careful design of the three cases. The probe stimuli were built around specific details of the transfers — the receiving account type, a specific reference code used in two of the transactions, the name of the third-party individual who had received one portion of the funds, and the specific date of the largest single transfer. These details had come from the bank records held by the solicitor — information that was not in the public domain and had not been shared between the two parties in correspondence.

The results

⚠️ The son — Deception Indicated — 88.7%
✅ The daughter — No Deception Indicated — 9.1%

The son produced a clear P300 recognition response to the receiving account type, the reference code and the third-party name. The daughter produced no recognition response to any of these details — consistent with someone who had no knowledge of the specific mechanics of the transfers.

I delivered results separately, as agreed. The daughter's solicitor received the written report summary by phone within the hour. The son had left the building before I had finished packing the equipment.

I do not know what happened next in the legal proceedings. These cases typically take months. What I know is that the evidential picture was significantly clearer at the end of the day than it was at the beginning of it — for the solicitors, for the daughter, and in a different way, for the son.

Field note

Estate disputes tend to be about more than money. The money is the thing that can be pointed at and argued over. What is actually being argued about is usually a version of: who did our parent love most, who did they trust most, who are they choosing even now in the way their death has been handled. P300 can address the factual question. It cannot address the one underneath it.

Reflections on a London Day

The 18:32 from Euston was quiet. I had a seat at a table and the EEG case in the overhead rack and about three hours to think about the day.

London produces a particular density of testing situations. Three cases of the kind we handled today — corporate, relationship, estate — in three different parts of one city, all booked independently, all arriving at roughly the same point in their respective situations. Each one different in every surface respect and similar in the fundamental one: someone needed to know something that they couldn't find out any other way.

The London testing market is unlike any other UK city we operate in. The corporate work is denser and often more complex — larger organisations, higher stakes, more legal involvement from the start. The private and relationship cases tend to arrive later in the crisis cycle than they do in smaller cities, probably because London generates more alternative support structures and people exhaust more of them first. The estate cases are proportionally more common — London's property wealth and the complexity of family structures in a city where people have moved from all over the world means contested estates are frequent.

We cover London from central locations across all zones. Same-day appointments are available most days. The logistics of a day like today — three locations, three setups, three very different emotional registers — are manageable with good planning and a packed equipment case. The harder thing to manage is not the logistics. It is carrying each situation appropriately: present enough to do the work well, separate enough not to take it with you on the train.

I'm getting better at that. Not perfect. But better.

Need P300 EEG Testing in London?

Corporate investigations, relationship testing, estate disputes. Same-day appointments available across London. From £499.

Back to All Blog Posts