Estimated reading time: 13 minutes
Key Takeaways
- Stress physically changes what officers see, hear, and remember. During high-stress incidents like an officer-involved shooting, physiological responses narrow attention, distort perception of time and distance, and filter out peripheral details.
- The more stressful the incident, the fewer details officers will recall accurately especially verbal exchanges, spatial positioning, and peripheral information. A review of 20 peer-reviewed studies of police and military officers confirmed this pattern consistently.
- Immediate, independent documentation is the most evidence-supported first step. Officers should record their own written or verbal account as soon as possible after an incident.
It starts like any other Tuesday. A patrol officer is midway through a routine shift, responding to a fender-bender report, a noise complaint, the usual. Then a disturbance call drops. Strip mall on the east side. He’s close, so he takes it.
Nothing about the call sounds like much. But when he arrives, the scene takes a dark turn. The subject is agitated, moving erratically, and then there’s a weapon. In the span of a few seconds, seconds that will later feel simultaneously like a fraction of a moment and an eternity, the officer makes a decision. Shots are fired. The subject goes down. The officer is uninjured.
And then it’s quiet.
The adrenaline is still surging as he calls it in, hands steadier than he would have expected, brain running on pure autonomic pilot. Relief settles in somewhere underneath all of it. He’s okay, the scene is contained, but it doesn’t last long. Because standing there in the parking lot, waiting for the cavalry to arrive, something else starts to creep in. The weight of what comes next.
He knows the process. He’s been through the training. But knowing the process and living it are two different things. There will be an administrative investigation. Possibly a criminal review. Attorneys, investigators, command staff, and maybe eventually a jury will pick apart every second of those few moments. His account of what happened isn’t just a report, it’s the foundation of everything that follows. His career, his freedom, his reputation all hinge, in part, on how clearly and accurately he can describe what his brain experienced under one of the most extreme physiological conditions a human being can face.
Over the years, agencies, unions, and legal advisors have operated under different assumptions about the best way to handle that account. Collect testimony immediately while it’s fresh. Wait 48 to 72 hours to let the officer stabilize. The advice has shifted, sometimes conflicting, and often not grounded in much beyond professional intuition. So, what does the evidence actually say about how and when to collect post-incident accounts from officers? That’s exactly what this research set out to answer.
The Study
In a systematic review published in Policing: An International Journal, researchers Paula M. Di Nota, Bryce E. Stoliker, Adam D. Vaughan, Judith P. Andersen, and Gregory S. Anderson synthesized two decades of empirical research on how stress affects police memory.
Their process was rigorous. Starting with 969 studies, they applied strict criteria to land on 20 peer-reviewed experimental investigations of memory for stressful events among law enforcement officers and military officers. The studies included both controlled simulations and real-world incidents. The authors excluded eyewitness studies, PTSD-focused research, and non-peer-reviewed literature. They were specifically focused on police, because police are not average eyewitnesses, and what happens to memory under the conditions officers regularly face deserves its own body of evidence.
Their core questions: How does stress affect what officers encode, consolidate, and recall? And what does that mean for the way agencies collect post-incident testimony?
How Stress Affects Police Memory: The Basics
Before getting into the findings, it helps to understand the basics of how memory is formed, because what the research shows makes a lot more sense once you understand what the brain is doing during and after a critical incident.
Memory has three stages: encoding (taking in information), consolidation (stabilizing and storing it), and retrieval (accessing it later). Stress doesn’t just affect one of these. It reaches into all three, and not always in the ways you’d expect.
Encoding: What the brain decides to record
Think of encoding as the brain’s intake process, the sensory and perceptual data captured in the moment an event is happening. When a threat appears, the body’s autonomic nervous system activates fast. Blood flow shifts to the large muscle groups. Fine motor skills deteriorate. And attention narrows sharply, almost like a spotlight, onto the source of the threat.
This is the system working exactly as designed. The brain is making a rapid triage decision: this is what matters right now. Focus here. Everything else can wait.
The consequence is that an officer in a high-stress encounter may encode the weapon, the subject’s movements, and the immediate threat zone with precision, while capturing almost nothing about verbal exchanges, bystander positions, or what was happening in the periphery. Not because he wasn’t paying attention because his brain was paying maximum attention to what it determined mattered most. There is also a set of well-documented perceptual distortions that come with extreme stress: auditory exclusion (sounds that seem muffled or absent), tunnel vision, a warped sense of time, and distorted perception of distance. These are not embellishments after the fact. They are predictable physiological effects of extreme arousal on the senses.
Consolidation: What actually sticks
After an event, the brain works to stabilize and store what was encoded, a process called consolidation. This is where short-term impressions either get locked into longer-term memory or start to fade. For high-stress events, this process can take longer than it does for routine experiences, unfolding across hours and multiple sleep cycles. REM sleep plays a real role here: the brain uses it to process and stabilize emotionally significant memories.
Retrieval: What comes out
Retrieval is the final stage: accessing stored information and bringing it back to conscious awareness. Here’s a detail that surprises people: stress doesn’t just affect what gets recorded, it also affects what can be recalled. During and immediately after a high-stress event, the prefrontal cortex (the part of the brain that handles complex reasoning, verbal articulation, and organized thought) has reduced blood flow and is operating at diminished capacity. The brain redirected those resources to the parts handling survival.
What the Research Found About Officer Memory Under Stress
Physical stress before an incident impairs what gets encoded
When officers experienced strenuous physical exertion immediately before a critical incident simulation, recall accuracy dropped, including recognition of target individuals and retention of briefing information delivered before the exertion began. Applied to our officer at the strip mall: if he ran a foot pursuit before the confrontation, his brain’s encoding capacity was already compromised when the critical moment arrived. The chase that preceded the shooting matters for the memory of the shooting.
High-stress scenarios produce fewer accurate details, especially verbal and peripheral information
Across multiple studies, officers consistently recalled fewer details and produced more errors in high-stress scenarios than in lower-stress ones. One study found that roughly half of officers involved in a lethal force simulation did not notice the suspect’s concealed hand (the one that eventually produced a weapon and fired at them). They encoded location details and spatial positioning. They missed verbal exchanges. Separately, research found that only 43% of officers in a use-of-force scenario accurately described their own shot placement, and only 57% could correctly identify the moment the situation justified lethal force. This is the brain’s attentional filtering process at work: threat-relevant information gets encoded, everything else gets crowded out.
Active role officers are not more accurate than observing officers
One study compared officers who took the primary role in a critical incident to those in secondary roles. There was no difference in accuracy between the groups. But primary role officers had significantly fewer correct details overall and the size of that gap was directly tied to how strong their physiological stress response was. The more stressed the officer, the larger the memory gap. Being the one who made the decision doesn’t give you a better memory of it. It may make it worse.
Stress affects every officer differently which is why accounts don’t always match
There is what researchers call an inverted-U curve when it comes to stress and cognitive performance. Moderate stress can actually sharpen encoding. Extreme stress degrades it. But the threshold for where any individual officer tips from adaptive to maladaptive stress is highly personal. Two officers standing at the same scene, experiencing the same incident, can have dramatically different physiological responses and dramatically different memory outcomes. This may help account for why officers at the same scene sometimes give accounts that don’t align. It is not necessarily that someone is being dishonest.
Delays in recall do not help and likely hurt
Here is where the research collides most directly with common practice. Some agencies, and some legal and union guidelines, operate on the assumption that giving an officer 48 to 72 hours before a post-incident interview allows them to recover and produce better testimony. The research does not support this. Delayed recall is associated with poorer accuracy and fewer details. Immediate recall after an independent account improves retention over time.
The concept of “critical incident amnesia”, the idea that stress-induced memory gaps will fill in over time, is not well-supported by the evidence. What actually happens is the opposite: details fade, and the gaps get filled with reconstructed memories and colleague accounts.
Colleagues and BWC both carry contamination risk
Officers who conferred with colleagues before completing an independent account showed increased confidence in their memories, but no improvement in accuracy. Worse, they incorporated false details from colleagues into their own accounts, but only when they had not already completed an independent written account. The independent account acts as a kind of protective anchor.
As for body-worn cameras: one study allowed officers to correct their post-incident notes after reviewing their BWC footage and found that important omissions, things like the presence of drugs, weapons, or individuals, went uncorrected. Verbal commands and subject behaviors were corrected. This pattern raises real questions about what is and is not being captured by the correction process.
Training-induced stress improves memory outcomes
This is the constructive finding in an otherwise sobering set of results. Studies on psychological performance training that incorporates realistic physiological stress showed improvements in situational awareness, lethal force decision-making, and recall. One study found that cadets who practiced tactical breathing during a high-stress exercise recalled more items from the exercise space compared to controls. Stress inoculation, when done right, appears to encode skills more robustly against later retrieval interference.
Post-Incident Interview Best Practices: What Your Agency Should Do
The researchers offered four concrete recommendations grounded in the evidence. Here’s how each one plays out using our officer from the strip mall.
1. Get an independent account down immediately before anything else.
As soon as the scene is stable and the officer is medically cleared, he should record an account of his own internal and external perceptions: what he saw, heard, and did. Written or verbal, it does not matter. What matters is that it happens before he talks to colleagues or reviews any footage.
This is not about building a legal defense. It is about anchoring his memory before contamination can occur.
2. Scale the timing of the formal interview to the magnitude of the stress response.
For lower-stress incidents, an officer in a secondary role who did not apply force testimony should be gathered relatively quickly after the independent account. For an officer who discharged a firearm and is showing signs of acute stress, the research supports allowing one night of REM sleep before the formal interview, to allow the officer to recover. But the formal interview should follow promptly after that. Some agencies operate under policies of waiting 48 hours or more. The evidence does not support this as a memory protection measure. It is more likely to work against accurate recall than in favor of it.
Officers experiencing significant distress should have access to medical and mental health support before sitting down for a formal interview. An officer trying to manage acute stress alone is an officer whose retrieval is under additional strain.
3. Treat the post-incident interview as a high-stakes, high-stress event in itself.
The officer sitting across from an investigator knows what is on the line: their career, their freedom, their reputation. That stress is real and it affects retrieval. Those conducting post-incident interviews should be trained on trauma-informed interview techniques or accompanied by a mental health professional.
Interviewers who understand trauma and mental health ask better questions. They do not expect a perfect linear account. They probe gently for sensory details rather than demanding a play-by-play narrative. Cognitive interview techniques, developed specifically to improve memory retrieval accuracy, are a strong fit here.
4. Make sure officers know their rights and resources before they need them.
The research notes that officers are more likely to give accurate accounts when the post-incident process is less adversarial and when they have access to legal representation and union support. If your agency has policy on legal representation during post-incident questioning, officers need to know how to access it before an incident happens, not in the parking lot after one.
Mental health resources should be proactively offered, not quietly available. An officer who is trying to manage acute stress alone is an officer whose memory is under additional strain.
For a comprehensive guide to structuring the administrative and investigative side of officer-involved shootings, the COPS Office guide is free and directly applicable: Officer-Involved Shootings: A Guide for Law Enforcement Leaders
For guidelines specifically on mental health support and agency preparation before and after an OIS, the IACP’s Officer-Involved Shooting Guidelines are available here: IACP Officer-Involved Shooting Guidelines
Final Thoughts
Back to Tuesday afternoon.
The officer in that parking lot is about to go through a process that will scrutinize a few seconds of his life in exhaustive detail. The people running that process, the investigators, administrators, legal teams, want the same thing he does: an accurate account of what actually happened. The research covered here offers something useful for all of them: a clearer picture of what memory does under extreme stress, and what conditions give it the best chance of being preserved accurately.
The sequence matters. Getting an independent account down before anything else. Scaling the interview timing to the stress response, not to a default policy. Conducting the interview in conditions that minimize additional stress rather than compound it. These are not radical departures from standard practice. For many agencies, they are refinements. But they are refinements grounded in two decades of evidence, and they exist to serve a goal everyone in the room shares. An account that reflects, as accurately as possible, what the officer actually experienced in those few seconds.
Enjoyed this post?
Get more research translated into plain language delivered straight to your inbox.
Subscribe below and let me know which topics you want to hear more about. It only takes a minute.
If you found this helpful, share it with your team or on your socials. It helps more people make sense of research that matters.
English