As many as 70 percent of dogs that start a service dog training program are let go before graduation. Given that it can cost up to $50,000 to develop one of these valuable pups, organizations that raise these dogs are always looking for better ways to predict who will be up to the task.
You may remember we wrote about a study last year where Emory University neuroscientists looked at dogs' preference for praise over treats. Their lab was the first to conduct functional magnetic resonance imaging (fMRI) experiments on awake, unrestrained pups to understand canine cognition and inter-species communication. Now they're using this technology to help solve the classic service dog dilemma--finding more accurate ways to eliminate unsuitable dogs earlier in the process.
Their study looked at 43 dogs who underwent service training at Canine Companions for Independence (CCI) in Santa Rosa, California. All of these pups passed CCI's standard behavioral tests, which selects dogs with a calm temperament to start the formal training program.
Scientists used fMRI to look for higher activity in the amygdala, an area of the brain associated with excitability. They found that dogs showing higher activity here were more likely to fail the training program.
"Data from fMRI provided a modest, but significant, improvement in the ability to identify dogs that were poor candidates," explains research lead Gregory Berns. "What the brain imaging tells us is not just which dogs are more likely to fail, but why."
The team believes that the fMRI would boost the ability to distinguish pups that would ultimately not pass from 47 to 67 percent.
This technology is expensive, so it wouldn't be practical for individual trainers, but could be utilized by larger organizations such as CCI. There's also an additional training component since the dogs must learn to remain still while undergoing the fMRI.
The second part of the study built on Emory's original treat research. In these experiments, the dogs were taught hand signals for "treat" and "no treat," which were shown while the pups were in the fMRI. They found a correlation between training program success and the caudate, a region of the brain associated with rewards.
In response to the treat signal, those who had more activity in the caudate were more likely to complete the service dog training program. In contrast, those with more activity in the amygdala were more likely to fail.
"The ideal service dog is one that is highly motivated, but also doesn't get excessively excited or nervous," explained Gregory. "The two neural regions that we focused on--the caudate and the amygdala--seem to distinguish those two traits. Our findings suggest that we may be able to pick up variations in these internal mental states before they get to the level of overt behaviors."
Gregory's team hopes to refine this evaluation technique and apply it to a broader range of working dogs, such as military and police pups.