Behavioral and physiological predictors of guide dog success

Abstract A range of mobility aids are available to assist people living with vision impairment, and of these, guide dog ownership offers them several unique benefits. However, training a dog to be a successful guide dog comes at a high cost (approximately AU$30,000). Therefore, the aim of this study was to determine whether temperament testing and kennel behavior measures could be used by Guide Dog Organizations for the early identification of dogs suitable for guiding work and thus to reduce production costs. Temperament tests (Passive and Noise, Sudden Appearance, and Dog Distraction Tests) and kennel behavior assessments (Activity Level, Salivary Immunoglobulin A Concentration, and Kennel Surveillance) were assessed in potential guide dogs (n = 25-113) at the Guide Dogs NSW/ACT Training Centre, Glossodia, New South Wales, Australia. Several significant predictors of guide dog success were identified. The presence of panting (P = 0.029) and licking (P = 0.005) when contrasted with baseline observations in the Dog Distraction Test, significantly reduced the probability of guide dog success. Success was also reduced with the latency for a dog to sit in the third Noise Test (P = 0.028), and when the time spent resting was reduced during the evening period (P = 0.018) in the Kennel Surveillance assessment. This study reports that 4 specific behavioral responses, which may reflect anxiety and restlessness, predict low suitability of dogs for guiding work. Through the identification of early predictors of guide dog success, resources can be more appropriately focused on dogs with a higher probability of success, whereas unsuitable dogs can be rehomed.