Good Cop, Better Cop
Is it possible to somehow distill the very essence of expert decision-making under stress, quantify it, and put it to use in a training program? Ward says that's exactly what he is trying to do.
Paul Ward and Joel Suss
Paul Ward, right, adjusts the eye-tracking device on graduate student Joel Suss as he runs through a law enforcement simulation. Photo by Mariusz Nowak
“Some tend to use them as video games and just tell the recruits to do the best they can. Frankly, you don’t learn much from that.”

For more information:

http://web.me.com/pw70/ACE_lab

Good Cop, Better Cop

by Frank Stephenson

It’s a life-or-death world out there, and experience can make all the difference. Can we teach street smarts?


Three days after Christmas 2010, 24-year-old rookie police officer Jillian Smith of the Arlington, Texas, Police Department responded to a call from a distraught woman claiming to have just been sexually assaulted by her ex-boyfriend and wanting to file a report. Smith, only two weeks out of field training, showed up alone at the apartment shared by the 38-year-old woman and her 11-year-old daughter.

Within minutes, the woman’s ex-boyfriend—a registered sex offender—burst into the apartment with a handgun and killed Smith with a shot to the head. He then shot and killed his ex-girlfriend before turning the gun on himself. In the mayhem, the terrified 11-year-old girl escaped to a nearby friend’s apartment.

In an average year, more than fifty police officers in the US are killed in the line of duty. Most are in their prime (in 2010 the average age was 38) and most are experienced officers with an average of at least ten years on the force. Every death is tragic, yet the death of a rookie officer somehow always seems more so. When young cops die, inevitably the same questions are raised: Did the officer make a “rookie” mistake? Did he or she suddenly confront a situation that their training failed to address—either adequately or at all?

Statistics on the circumstances surrounding novice police officers killed or wounded in the line of duty aren’t easy to come by, nor is there much evidence that the ways recruits are trained in any of the country’s hundreds of law enforcement academies each year are fundamentally flawed. But implicit in an almost universal requirement by police agencies that rookies be partnered with veteran officers suggests a well-understood maxim in law enforcement administration, namely that no amount of classroom instruction or field training—regardless of its quality or duration—can match on-the-street experience for creating an expert cop.

Whatever it is about real-world experience in any field that requires quick and accurate decision-making—from law enforcement to professional sports—is endlessly fascinating to cognitive psychologists, scientists who study how we think. Michigan Tech’s Paul Ward is one such researcher who has made a career pursuing the elusive elements that make up the thought processes of people who are remarkably good at making correct, split-second decisions—and doing it consistently.

Ward is an associate professor in the University’s Department of Cognitive and Learning Sciences. His latest work uses a series of highly controlled video simulations designed to contrast the performance of rookie policemen with that of their fellow officers with years of experience. He, along with his grad student Joel Suss and other colleagues, recently published findings that they believe offer some of the best insights yet uncovered in the study of expert decision-making under stress.

“From the start, our No. 1 goal in this research has been to develop a comprehensive understanding of skilled decision-making in complex law enforcement situations,” Ward said. “We believe we can use this knowledge to test and improve training methods [in law enforcement] that can help save lives.”

As it stands now, most law enforcement recruits go through about twelve weeks of training that incorporates classroom instruction, practice on the firing range, and exposure to a series of highly realistic video simulation exercises. Ward said that such simulations—which are commercially produced and can cost up to $100,000 apiece—can offer a powerful way to test and improve recruits’ decision-making skills. These interactive programs present novices with experiences that mimic real-world situations closer than any other training technique available. Most training centers have them as part of their standard curricula, but how these sophisticated tools are used is spotty, Ward said.

“Some centers use these tools extensively, but others just don’t have the manpower to put all their students through these exercises,” he said. “Some tend to use them as video games and just tell the recruits to do the best they can. Frankly, you don’t learn much from that. So, I think it’s fair to say that the training centers aren’t always getting the kind of feedback from these simulations that they could be.”

Ward’s team has designed techniques for using these training videos as a means of getting inside participants’ heads. They use what they call a “think out loud” protocol, in which participants tell what they’re thinking at every step as they progress through simulations that often involve the use of lethal force. When the video ends, the officers immediately report on what just happened and try as best they can to explain exactly how they reacted and why.

“We’re getting about as close as we can to the stream of thought of these officers without having a video recorder in their minds’ eye,” Ward said. “In these tests, it’s truly astounding to see the differences between skilled SWAT officers and rookies.”

For his latest study, twenty-eight police officers volunteered for Ward’s experiments run in his campus lab. Half the officers were rookies, fresh from basic training in the academy. Their average age was 24.7 years. The other fourteen were highly skilled tactical officers, who averaged 38.7 years in age with an average of 15.4 years of service. All officers were run through a series of twenty video scenarios, projected on a 9-by-12-foot screen, which included situations ranging from simple traffic stops to dangerous hostage stand-offs. Eleven of the videos ended in confrontations that forced the officers to fire their weapons or risk getting “killed.”

As Ward predicted, in every run-through the skilled officers significantly out-performed their younger, less-experienced counterparts. These officers not only reacted faster (on average, by about 1.4 seconds) but also used their weapons more effectively, hitting the right targets more often than did the novices. How were they doing that?

Through their careful interviewing techniques and detailed analysis of the simulated scenarios, Ward and his colleagues were able to prove that skilled officers are much better at picking up on environmental cues and faster at analyzing them than the rookies. In test after test, the veteran officers showed a keener ability to assess situations, develop and analyze their options, and successfully intervene—e.g., either shoot the perpetrator or otherwise diffuse the situation peacefully.

In parsing the officers’ do-or-die decision-making process—which typically spans only a fraction of a second—the researchers found that the more experienced officers generated a higher number of critical options before taking action than did their counterparts. This lightning-fast mental calculus suggests that over the years, officers acquire skills that enable them to encode critical information in long-term memory and in a way that is instantly available to them on the fly when it’s most needed. Rookies simply don’t have this amazing retrieval system and as a result risk injury to themselves or others before they acquire it the only way they can—through on-the-beat trial and error.

Is it possible to somehow distill the very essence of expert decision-making under stress, quantify it, and put it to use in a training program? Ward says that’s exactly what he is trying to do. He envisions dramatically improved training in any profession that demands rapid and accurate life-or-death decisions, from law enforcement and the military to health care.

“The ultimate goal for us is to improve the performance of those who make really critical decisions on a daily basis,” Ward said. “We see it as a way to build a better, safer world.”