An innovative tool has been developed by US and UK researchers to help trainers identify dogs that are best suited to be assistance dogs for people with various impairments or disabilities, such as visual, hearing, motor, or psychological issues.
The tool is an artificial intelligence algorithm created by scientists at the University of East London and the University of Pennsylvania. It evaluates a dog’s personality based on data from nearly 8,000 Canine Behavior Assessment and Research Questionnaires (C-BARQ), which has been a standard test for evaluating potential working dogs for over 20 years.
The C-BARQ questionnaire, however, has some subjective questions that can bias the results. The artificial intelligence algorithm is able to adjust for these outlying responses, as explained by James Serpell, a professor of ethics and animal welfare emeritus at the UPenn School of Veterinary Medicine.
Professor Serpell and his colleagues’ algorithm groups C-BARQ results into five main categories, which include “excitable/attached”, “anxious/fearful”, “aloof/predatory”, “reactive/assertive”, and “calm/agreeable”, offering insight into a dog’s personality and their potential as professional working dogs.
Assistance dogs need to be patient, adaptable, and have good stress management skills to help their owners on a daily basis. Trainers prefer calm, docile dogs that are easier to train, as dogs with a strong character can be much more difficult to train.
The researchers believe in the potential of their canine personality test algorithm but acknowledge the need for further research. They suggest in a paper published in the journal Nature that the AI-based methodology could enhance the selection and training of dogs for specific roles in the future.
While the AI software is designed to aid in the selection of future assistance dogs, it could also be used in shelters to reduce the likelihood of adoptive families reconsidering their choice due to compatibility issues with their new canine companion. – AFP Relaxnews