Human-directed agression in shelter dogs: how to test for better prediction of outcomes

Abstract

In the last few decades many test series were developed and used to measure behavior in shelter dogs to identify individuals with aggressive tendencies who may not suitable for rehoming. However, relatively few studies evaluated the effectiveness of these test procedures. According to a survey based on questionnaire data, 40.9% of the dogs that had passed a temperament test showed aggressive behavior in their new home within 1 year of the adoption. The aim of the present study was to explain this high error-percentage by finding the possible shortcomings of the procedure. According to our hypothesis, dogs’ behavior is suppressed in many respects when kept in a shelter, thus tests carried out during the first few days might have less predictive value. Human-directed aggression was investigated in 25 shelter dogs. We first tested the dogs 20–40 hours after intake and a second time 2 weeks later. The same test series was applied on the 2 occasions and we analyzed the difference between the behavior shown in the 2 tests. Based on our experiences from a previous study on pet dogs, the following subtests were used: friendly greeting, taking away a bone, threatening approach, and dominant approach. The dogs were tethered to 2 trees in a V-shape with two 3-meter long chains, so that they could not do semicircular movements. The test procedure was carried out by 2 women experimenters. We coded the level of aggression in all the 4 subtests. We found significant differences in the dogs’ aggressive behavior between the first and the second test. The number of dogs showing aggression in the taking away a bone test and also the level of aggression observed increased remarkably for the second test date. In contrast to results in pet dogs tested with their owners, the threatening or dominant approach did not evoke aggressive behaviors from shelter dogs. We suggest that timing is a crucial factor of the testing procedure to gain more predictive results in tests trying to predict human-directed aggression in shelter dogs. Moreover, we propose that threatening approaches from humans in a shelter setting tend to mainly evoke avoidance behavior in shelter dogs (partly because of the owner’s absence), whereas the same stimuli might facilitate (fear-related) aggression in pet dogs (in the presence of the owner).