Can AI Assess Canine Personality?

Can AI Assess Canine Personality?

Artificial intelligence is creeping into many aspects of our human lives. Now, it’s even being used to help map and assess canine personalities and behaviour traits. Holidays4Dogs takes a deeper dive into the subject of artificial intelligence and asks, can AI assess canine personality?

Artificial intelligence has been used widely in the past, for predicting human personality outcomes. However, such an approach has not previously been used to determine canine personality. Before now, behaviour trait collection has relied on questionnaires of the type derived from the C-BARQ project.

What is ‘personality’ in the canine sense?

In human psychology terms, personality refers to the way a person consistently behaves, acts, thinks and feels. Likewise, personality in dogs could be thought of as patterns of behaviour traits. These might include prey-drive, trainability, curiosity and aggression. Behavioural issues that we might think of as problems are simply natural behaviour for dogs. However, often they have to deal with these natural instincts in unnatural settings and environments. Canine personality traits can also impact on public health and safety i.e. attacks on people.

New ground breaking studies using AI to predict canine personality.

A ground breaking study by the University of East London (UEL) is using artificial intelligence (AI) to understand and predict which puppies are more likely to be successful as support dogs.  This could offer a significantly useful new tool to help trainers. Particularly for establishing which puppies are going to respond better to training. This could not only reduce costs, but benefit service users practically and financially.

The University worked closely with support dog organisations, by using data from behaviour surveys collected by puppy raisers at two stages of development – 6 and 12 months. This data was then applied to AI machine learning in in order to understand the mysteries of canine personality.

The findings show promise for use in determining pet dog personality and matching to the right homes. One of the most common reasons for people relinquishing their dog to a rescue centre is the dog’s behaviour. With more detailed mapping tools, predicting a dog’s behaviour could provide more information to trainers and dog carers. In turn, this could prevent dogs ending up in unsuitable homes. In addition, establishing the type of personality a dog has can inform any specialised training and care required – thereby, raising welfare standards.

5 main personality types in dogs.

The study discovered five main personality types amongst domestic dogs. These were  – Excitable/Hyper-attached, Anxious/Fearful, Aloof/Predatory, Reactive/Assertive and Calm/Agreeable. However, it is the predictive modelling that holds promise for predicting outcomes in young assistance dogs.

What are the current challenges in raising support dogs?

One of the biggest challenges in raising and training support dogs is the financial outlay and time involved. For example, according to the Autism Dogs Charity who provide support dogs to those suffering from Autism, the cost of caring and training a service dog costs £23,000 and takes months of specialised training. As a charity, Autism Dogs, can cover £8,000 of these costs. However, this leaves a further £15,000, which families themselves must fund.

AI technology that can assess canine personality could help to evaluate the most suitable dogs for support dog roles. Therefore saving time and finances.

Conclusion.

The research aims to inform more effective training programmes not just for support dog organisations, but for pet dog owners too. Ultimately, the findings could inform better matching between dogs and their owners and foster stronger bonds between existing partnerships between dog and human.

The results also indicate that machine learning models can learn to assign dogs to specific roles. The findings have quite significant implications for dog breeding and training programmes world-wide.