A few days ago, I heard a story on the radio about police dogs and their handlers. The reporter was talking to a retired police dog handler who now trains dogs and works as an expert witness. What he said was disturbing for anyone who gets stopped by a police officer-and-dog team, but, to anyone with dog training experience, sounds plausible.
What he was talking about was how common it is for the K9 handlers to miscue their dogs.
Sometimes it is conscious and intentional; the officer wants to do a search so he claims that the dog alerted. That gives him probable cause so the search is legal. That’s the idea, anyhow. This reporter said that “cops even joke about dogs being probable cause on four legs.” The trainer backs this up.
Sometimes, he says, the miscuing is unintentional — subconscious. The officer believes that there will be something to find, and he unconsciously signals this to the dog. A study done at the University of California, Davis, in 2011 tested teams on searchers where there was no contraband. The study was actually looking at the handlers’ behavior. Dogs are so attuned to their handlers — and often so eager to do what they think the person wants them to do — that they’ll signal. And the researchers found that, over and over, handlers led their dogs to alert to … nothing.
I’m taking a (very beginning) scent work class with Cali. At this stage, I always know where the target scent is. And I do have to watch myself so that I don’t give her hints. It’s so easy to look or walk toward the hiding place without even realizing I am doing it. I don’t yet know what it will be like to let Cali go and search when I have no idea where the scent is. I can certainly see where she’d take a cue from my behavior, though, and alert to something I was looking at or walking toward.
K9 handlers are often experienced police officers, but generally they are not dog trainers. They get trained to work with their already trained K9 partners, but it’s often hard for even the most experienced dog professionals to fully trust their dogs all the time.
Professional scent-team trainers are getting smarter about how they test. They’ll set up multiple rooms and randomly assign teams to test sites. No one in the room — not the handler and not the evaluator — knows where or even whether there is anything to find. It’s as much a test of whether the handler trusts the dog as it is a check on the dog’s skill. Some handlers are so sure there is something to find that they have the dog go around the search area or vehicle over and over; in these cases, the dogs often false alert. In the test arena, these teams fail, but in real life …
A K9 officer tells the reporter that “the dogs are mainly used to confirm what we already suspect.
Suspect. Not “know.”
The officer goes on to say that “about 99 percent of the time, we get an alert.”
Any dog trainer will tell you that the hard part isn’t teaching skills to the dog. The human handler is always the weak spot in the team.
The stakes aren’t high in the training games that I play with Cali. If she alerts on an empty container, I shrug and send her back to look. I’d like to think that, if I didn’t know whether there was anything to find, that I’d trust her. Well, not yet, but once she becomes more reliable. But who knows? It’s so easy to miscue and — intentionally or not — lead the dog to do what she thinks you want.