To best understand this article in the context of the behavior evaluation literature, please see National Canine Research Council’s complete analysis here.
Shabelansky, A., Dowling-Guyer, S., Quist, H., D’Arpino, S. S., & McCobb, E. (2015). Consistency of shelter dogs’ behavior toward a fake versus real stimulus dog during a behavior evaluation. Applied Animal Behaviour Science, 163, 158-166. doi: http://dx.doi.org/10.1016/j.applanim.2014.12.001
National Canine Research Council Summary and Analysis:
This 2015 study by Shabelansky, Dowling-Guyer, Quist, D’Arpino, and McCobb evaluated a specific subtest used in many common behavior evaluations (e.g., SAFER, Assess-A-Pet). Researchers compared shelter dogs’ behaviors towards both a live and a fake dog.
The purpose of this study was to determine whether a fake dog could reliably be used in lieu of a real dog during shelter evaluations. The authors underscore that they were not attempting to predict future behavior in real life situations, but rather wanted to evaluate consistency between the two scenarios. If the two stimuli yield similar responses this would justify the use of a fake dog instead of a live dog during behavior evaluations, which could potentially save resources and increase safety and/or comfort for shelter dogs and staff.
Forty-five dogs from two shelters in Massachusetts served as subjects in this experiment. The use of dogs from shelters is a strength of this design, as shelter dogs are the target population when these tests are applied. The study used a within-subjects design, with each dog experiencing both the live dog and fake dog trials. Order of presentation was counterbalanced across subjects and sex.
A certified pet dog trainer who was experienced with behavior evaluations recorded data for nearly 40 specific behaviors. A wide range of behaviors were coded including play bowing, licking, mounting, cowering, growling, approaching, tail wagging, and sniffing. Each behavior was marked as either present or absent and each occurrence was scored independently of the other behaviors. The presence, or absence, of behaviors was compared between the two conditions, which resulted in three scores: positive agreement, negative agreement, and Cohen’s Kappa. Percent agreement was impacted by low-occurring and high-occurring behaviors.
The researchers grouped several behaviors to form three behavior categories: friendly/playful, fearful, and aggressive and the overall finding was that the fake dog elicited behaviors that were consistent with those elicited by the real dog for friendly/playful, but not aggressive, behaviors. This is important because “aggressive” behaviors are what shelter staff are looking for when conducting these assessments; their purpose is to identify potentially aggressive dogs. Instead, the fake dog stimulus appears to elicit friendly behavior in dogs whose behavior was also friendly towards the real dog in the test, but little more.
This was a well conducted study with many strengths. For one, the authors provided a nuanced understanding of the data. The short time of each trial (~1 minute) may at first appear to be a weakness; more data could be collected with a longer trial period which could reduce the instances of negative occurrences. However, the trial duration mimics the limited time shelter staff has when conducting evaluations. It was enough time to observe several frequently-occurring behaviors. Of course, it was not without its weaknesses. Because the rater could see the stimuli, it is not possible to be blind to the condition. The authors note that the rater did not have access to the dogs’ behavioral histories, but that does not address the potential experimenter bias.
The data showed that with respect to measuring “aggression” on a behavior evaluation, a fake dog is not an adequate substitute for a real dog. Overall agreement between the conditions was high, but the authors acknowledge that this is due to a consistent lack of behavior, rather than a consistent presence. Moreover, behaviors that were consistent due to positive occurrences were friendly behaviors (tail wagging and sniffing, for example), not “aggressive behaviors.” Lastly, dogs responded with more “aggression” towards the fake dog than the real dog; if fake dogs are used for behavior evaluations, they are likely to induce false positives. For these reasons, fake dogs should not be used to predict how a dog might behave towards a real dog in the same evaluation, and evaluations that use fake dogs should be discounted as this method is not validated.
Abstract and Link to Purchase Full Text of the Original Article: