Getting Past the Performance in Law Enforcement Interviews
I've spent fifteen years doing pre-interview preparation for cases where suspects have legal representation and media training. The people who hire me usually assume the problem is extraction. It isn't. The problem is that trained interviewees know how to give you what you asked for while saying nothing you can use. This happens in federal contexts constantly, and the people I work with are often surprised when their three-hour session produced six usable paragraphs. There's a specific workflow I rely on when dealing with prepared subjects. It comes from watching too many interrogations fall apart because the interviewer was chasing the wrong metric. You want the Larry Hall Real Life Interview method because it addresses exactly this, but you need to understand how it actually functions before applying it. Most people get the setup wrong and then blame the technique.
Why Standard Interview Models Fail Against Trained Subjects
When someone has been through investigative interview training, they've been given a script. It's not elaborate. It's usually three moves: establish credibility early, deflect with cooperative language, and pivot to what they control. The interviewer, reacting to surface compliance, ends up building a timeline the subject approved. That timeline is fiction. It's the fiction the subject wanted you to record. I encountered this in 2019 on a witness intimidation case. The subject had been a union steward for twenty-two years. He knew procedure. He knew how to make a statement that looked complete while containing zero evidentiary value. My partner was three weeks into a timeline reconstruction that collapsed at deposition. The subject had answered every question honestly within his own definitional framework, which happened to exclude anything that mattered. We lost two months and a key prosecution theory because we accepted cooperative language as substantive language.
The Actual Mechanism Behind Structured Unscripted Inquiry
The approach works because it removes the subject's ability to control the conversational frame. Standard forensic interview models give the subject boundaries. They say "ask open-ended questions" and then the subject defines what "open" means. The structured variant removes the boundary-setting opportunity by establishing the frame before the subject enters the room, then anchoring every response to that frame rather than allowing the subject to reframe. You anchor by establishing operational definitions at the outset. Not legal definitions. Operational ones. If you're investigating alleged obstruction, you define what you mean by "attempted communication" before asking whether any communication occurred. The subject can't later claim they misunderstood the term because you told them what you meant. This cuts the deflection path that trained interviewees exploit. The timing matters more than most practitioners realize. Most people set the anchor too late, after the subject has already established a narrative position. I typically do it within the first four minutes, before small talk completes. The subject hasn't formed their defensive posture yet, so they accept the frame without noticing they've accepted it. By the time they'd normally start managing impressions, they've already agreed to the terms of engagement.
Get the Full Details
What This Actually Looks Like In Practice
A practical example from last quarter. I was preparing for a case involving financial records and a CFO who had been interviewed by internal audit twice before. Both times, the internal team got detailed explanations of why the transactions were normal. The explanations were internally consistent and completely inaccurate. The CFO had never been asked to justify the transactions against the operative definition of "routine." She'd only been asked to describe them. My opening anchor established that "routine transaction" meant something very specific: standard billing cycle, documented approval, no unusual timing, no counterparties outside the approved vendor list. Then I asked whether any transactions during the period failed to meet all four criteria. She could have deflected by talking about process. Instead, she had to confront whether her transactions met the defined standard. Four of seven flagged transactions couldn't meet all four criteria. The other three required clarification on the timing criterion, which exposed the anomaly. This took forty-seven minutes. The previous internal interviews had taken three days each. The difference wasn't the questions. It was what the questions were measuring.
Where This Approach Breaks Down
I need to be honest about the limitations because people who recommend this technique rarely do. It requires the interviewer to understand the substantive domain better than the subject expects. If you're investigating technical fraud and you don't understand the technical controls being discussed, your anchor will be wrong, and the subject will exploit it. I've seen this happen with cryptocurrency cases where the interviewer anchored on "exchange" terminology that didn't match the actual platform architecture. The second limitation is admissibility. Some jurisdictions require explicit Miranda warnings before any structured inquiry, and establishing an operational frame can be construed as the functional equivalent of questioning if done aggressively. I always run my anchors past counsel before application. About twelve percent of the time, counsel identifies issues that require modification. There's also a subject population where this doesn't help. Neurodivergent subjects, particularly those with certain communication processing differences, may not navigate the frame-acceptance mechanism the same way. I had a case in 2021 where the anchor caused the subject to disengage entirely rather than comply with the frame. The subject wasn't being difficult. They were processing differently. We switched to a completely different approach and recovered the necessary information, but it took longer and produced less detailed testimony than I would have preferred.
Alternative Approaches When This Method Fails
If you're dealing with a subject who has extreme media training, sometimes the better path is the chronological reconstruction model. Instead of anchoring definitions, you ask the subject to walk through their day hour by hour for the relevant period. Trained subjects often have pre-written narratives that break down under minute-by-minute pressure because the narrative contains gaps that don't align with temporal reality. This approach is slower, maybe two to three times the duration, but it catches inconsistencies that definitional anchoring misses. For subjects with communication differences or cognitive processing issues, the cognitive interview variant remains more reliable. It's designed for exactly this population. The framework assumes good faith and uses recall-enhancement techniques rather than frame control. It's less efficient against prepared deceptive subjects but far more reliable across diverse populations. The key insight most people miss is that no single model handles all interview scenarios. The Larry Hall Real Life Interview methodology is valuable specifically because it targets prepared, compliant-but-evasive subjects who exploit standard models. It's not universal. Understanding when it fails tells you as much as understanding when it works.
Practical Application Notes
If you're implementing this, start with the anchor timing. Four minutes in, before the subject establishes narrative control, is when I've found it most effective. Early enough that the subject hasn't formed their defensive posture. Late enough that you've done basic rapport establishment, which still matters even when you're controlling the frame. Document your operational definitions in writing before the interview. Not for the subject. For yourself and for any subsequent proceedings. When the defense challenges your methodology, having written definitions prevents the "you moved the goalposts" argument. I've had this happen twice in my career, and both times the written record saved the prosecution's position. The approach also works differently depending on whether the subject has legal counsel present. With counsel, the frame acceptance becomes more deliberate. The subject's lawyer will often object to definitional anchoring as suggestive. In those cases, I shift to the chronological model or use the cognitive interview variant. The underlying principle remains: control what you're measuring. The mechanism just changes.
I still use this workflow on roughly sixty percent of my cases. The remaining forty percent require alternative approaches, usually because of the subject population or jurisdictional constraints. Understanding which category your interview falls into takes practice, but it's the kind of practice that prevents case collapse at deposition.
