The Washington PostDemocracy Dies in Darkness

Police usually wait days before interviewing officers in shootings. A new study says they shouldn’t.

Alabama State Police officers investigate the scene after a fatal officer-involved shooting in Decatur in November 2015. Detectives typically want to interview shooting suspects right away, but police officials generally insist on a delay of a few days when officers are involved. (Gary Cosby Jr./Decatur Daily)

In the hours after a police officer shoots someone, here’s something investigators usually don’t do: interview the officer.

For years, the accepted wisdom in the police community has been that the officer should be given time to calm down from the traumatic event and that full sleep cycles once or twice before an interview will enhance his or her recall of the episode. Some jurisdictions have rules or laws mandating a waiting period, and some departments have reached similar agreements with police unions.

“Officers should have some recovery time before providing a full formal statement,” the International Association of Chiefs of Police states in its “Officer-Involved Shooting Guidelines.” The guidelines say, “An officer’s memory will often benefit from at least one sleep cycle prior to being interviewed leading to more coherent and accurate statements.”

But a new study of 87 veteran police officers, some interviewed immediately after active-shooter training and some two days after the training, could start to change that thinking. “We did not find any evidence,” wrote criminologists Geoff Alpert, Louise Porter and Justin Ready, “that delay improves either recall or cognitive capability that could indicate enhanced ability to respond to questioning.”

Even though officers who were questioned shortly after emerging from a high-stress scenario “felt heightened anxiety, and reported less confidence in their cognitive ability,” the study found that those factors “did not seem to impede officers’ ability to recognize details of the scenario, or form new memories and perform in the cognitive tasks.”

The study added a touch of common sense, noting, “While our results suggest immediate questioning is better for memory, this must be considered alongside other priorities, such as protecting and optimizing officer well-being,” a component recommended in every police shooting case.

The study also built on a commonly held perception about memory. “One of the most fundamental laws of memory,” said Stanford professor Elizabeth Loftus, a renowned expert on human memory, “is that it fades over time.” She noted that memory “not only fades, but is also vulnerable to post-event contamination, or eroding.” Loftus said that “we couldn’t see any good reason for this kind of long delay” in conducting police interviews.

Bill Lewinski of the Force Science Institute is regarded by many police officials as the definitive expert on the subject of interviewing officers after shootings. He is a behavioral scientist who has studied police use-of-force issues since 1975. “Delay enhances an officer’s ability to more accurately and completely respond to questions,” he wrote in one widely cited paper, arguing for “a recovery period of at least 48 hours before being interviewed in depth.”

He was critical of the Alpert group’s paper, which was first previewed at a national police chiefs’ conference and is being reviewed for publication. Lewinski said that he had interviewed more than 2,000 officers who had been involved in shootings and that “their reactions are all over the place.” But he said that experts “will tell you, no question, if you need to improve memory, sleep on it.” Lewinski said there were “serious flaws in the current research. The best study has not been done.”

The new study took place in Queensland, Australia, in 2016, following a live active-shooter simulation in an abandoned building. The average age of the 87 officers who participated was 42, with an average of 15.7 years of police experience. The officers were divided into two groups: One group was interviewed immediately after firing at the shooter in the simulation, and the other was interviewed two days later. The first group was also interviewed two days later, to see how their memory performed in a second interview.

The officers were given 19 multiple-choice questions related to different parts of the event, separated into nine threat-related questions and 10 questions about nonthreatening details. The threat questions asked about visual items such as the weapons, auditory items such as verbal threats and spatial items such as the proximity of the shooter to the officers. The nonthreat questions included queries about the clothing of those present, who fired first and the number of shots, and the rooms searched before encountering the shooter.

“Officers’ general cognition (their ability to respond on the tests),” Alpert’s group wrote, “did not seem to be directly affected by how recently they had experienced the scenario, and no significant improvement was evident after two days either between or within groups.” They did note that “recall of non-threat related information was significantly better in the immediate condition compared with the delayed condition,” which Alpert attributed more to a focus on weapons and threats than “the color of the wall.” The study concluded that “delayed questioning negatively impacted officers’ recognition . . . only for details not directly associated with threat stimuli,” but officers who were questioned immediately “did not experience the same memory decay over time, showing that early questioning can aid memory retention.”

Alpert, a University of South Carolina professor who studies high-risk police behavior, said the policy of waiting to question officers was “based on philosophy, not data.” He said delaying creates an issue of “Do you remember because you were told or because it happened?”

Lewinski criticized the use of multiple-choice questions in the study to gauge accuracy of memory and levels of stress, which could be better done with heart and pulse monitors. He said using trained cognitive interviewers would be a more realistic way to replicate how an officer would respond in a police interview. He also noted that the study did not consider that many officers have worked extremely long hours by the time of a shooting or post-event interview, which harms their cognitive abilities.

Another issue with interviewing officers after a shooting is that they are entitled to the same protections as any citizen, namely the Fifth Amendment right not to incriminate oneself. Although officers are required to speak with internal-affairs investigators, those interviews are usually conducted much later and cannot be used against an officer in a criminal case. Officers typically consult a lawyer soon after a shooting, and those lawyers typically advise the officers not to talk.

But officers can give statements to homicide investigators if they wish, and often do. Attorney Edward J. Nuttall, who has represented dozens of officers involved in shootings, said: “Based on my experience, the memory is no better or worse immediately after, whether it’s three or six days after the incident. What is much different is their physical self. Their blood pressure, the adrenaline, the ability to articulate what they did or what they observed. Dr. Lewinski’s the authority on this. It’s common knowledge that after three or four sleep cycles, your adrenaline will go down, your blood pressure will go down, and you’re better able to articulate what you’ve observed.”

Nuttall noted that many officers are traumatized by the event. “Shooting someone is something no one ever wants to do,” he said. “Unfortunately it happens, they have to deal with it, they have to live with it. It’s trauma. Many face PTSD. And that PTSD starts the day of the incident. I’ve seen men and women still dealing with it 20 years later.” He added that “99.9 percent of the time they’ve acted appropriately and in accordance with the law.”