Detection Dogs: Are Dogs and Handlers Biased? (Part 2)
Dogs can detect incredibly faint odors. But what if their human handlers give them the wrong cues? In Part 2 of this series, the Dog Trainer explores further complexities in the work of detection dogs.
Page 1 of 2
In Part 1 of this series on detection dogs, I talked about the alarming results of a 2011 study of drug-detecting and explosives-detecting dogs. The study, conducted by Lisa Lit, Julie Schweitzer, and Anita Oberbauer of the University of California at Davis, was called “Handler Beliefs Affect Scent Detection Dog Outcomes.”
The gist of it was that dogs’ handlers were led to believe that target scents were present in the test location. In fact, no such scents were present, so there should have been no alerts – but in 164 trials, the handlers reported that their dogs gave 225 alerts to target scents. An analysis of the results showed that the handlers’ expectations affected the dogs’ performance in this test version of a very high-stakes situation.
Predictably, detection dog handlers did not like Lit’s study. I remember the uproar on some trainer discussion groups, with handlers insisting the study was biased, artificial, unfair unfair unfair. An organization called the Scientific Working Group on Dog and Orthogonal Detector Guidelines published a letter detailing their objections. The ethologist Roger Abrantes and his colleagues Marco Costa Pinto, Miguel Rodrigues, and Tiago Dosta Pinto directly challenged Lit’s team with a 2012 article entitled “Handler Beliefs Do Not Affect Police Detection Dog Outcomes.”
This week we will explore their objections and attempt to answer the question: Was Lisa Lit’s study just blowing smoke?>
Were the Dogs in the Study Adequately Trained?
The Scientific Working Group’s first objection was that Lit and her colleagues hadn’t assessed the certification and maintenance training of the teams in their study, nor did they provide info about the handlers’ experience. Maybe so – the original article doesn’t say one way or the other – but isn’t that kind of beside the point? Lit studied active working teams, and the active working teams screwed up. If they screwed up because they were inadequately trained, then that tells you something about the quality of training done by working detection teams. It’s a little hard for me to see why the problem there is that Lit didn’t cherry-pick her subjects.
Was There Contamination?
Next up, the Working Group argued that there could have been scent contamination from samples Lit and her colleagues brought to the church where they did their experiment, even though the samples were triple-bagged and in a metal box and never entered the actual testing area. But if many of the dogs’ false alerts were near where the samples were stored, then the Working Group might have a point. Without that information, I’d call this one a toss-up.