Home About Me Dog Personality Survey Optimism in Animals Further Reading


Dogs have personality. We all know it, but how can it be measured or defined? Personality in humans isn't even that well defined, but broad definitions are favoured, such as consistent patterns in the way an individual thinks, acts and feels (John 2008). Unfortuantely, we can't really ask dogs how they think and feel and how they think they would act like we can people, so in dogs and other animals, personality is usually studied and defined through behavioural observations. The problem with this is that behavioural observations can be collected in any number of ways and are generally interpreted as telling us something about a broader personality trait. For example, a dog that readily approaches novel objects may be considered self-assured. Unfortuantely, such a dog may also be considered as expressing curiosity and fearlessness, self-sufficiency, or investigation. These are subjective names given to describe a general trait, yet we don't know if those behaviours are in fact indicative of broader personality traits and whether, say self-sufficiency is the same as fearlessness.

Dog personality is often studied by way of owner surveys, which comes with similar problems with validity. Data reduction techniques mean that we can at least tell if there are clusters of data. For example, it seems trainability and sociability often appear together in the same dog. But it's hard to tell if 'socialibility' and 'trainability' mean the same things to different people. If we look at how different people that know a dog rate them in the same survey, we can tell if people tend to agree with each other on terms, which seemingly solves that problem, except that we still don't know if people are agreeing on a personality trait that dogs have or a concept of a personality trait that dogs may not have. And that brings us full circle back to behavioural observations.

Study of dog personality does have an important role to play. The better we understand personality the better we understand an individual dog and how it is likely to behave. Behavioural observations may be a good way to make those connections, but collecting behavioural data is time consuming and can be open to interpretation. Survey data offers the means to collect a lot of data with little effort, and can reveal some broad patterns that may inform further study. I used such a survey to ask Australian dog owners about their dogs' personality. I had 1057 usable responses and confirmed the existence of a 'super-trait' that is currently being called boldness. In dogs, boldness is associated with interest in other dogs and people, and low fear, and also trainability in other studies (but not mine). I found clear patterns between boldness in dogs and breed group, age of dogs, sex, and reproductive status (neutered or entire). The two papers reporting on the survey results are linked below.

Aside from flagging influences on broad behavioural tendencies such as age and reproductive status, the survey does seem to be a useful tool for asking people about their dogs' personality. The science doesn't stop here. We can use the survey in conjunction with behavioural observations to try to work out what exactly it is that people identify in dogs when they think about their personality. And with the ability to now investigate optimism and pessimism in dogs, we can link those results with personality survey results and look for significant patterns. This may help us adopt a more mathematical approach to dog personality study, like whether we can tell something about a dog's personality by how quickly they give up on a task if reinforcement is stopped, or how impulsive they are in their choices. This would solve some of the problems with validity by taking subjectivity out of behavioural observations and descriptions.





Last updated: October 2013