An investigation of personality in the domestic ferret (Mustela putorius furo)

This thesis investigated personality structure in a small, social carnivore, the domestic ferret (Mustela putorius furo). Given that several species-specific behaviours are not well understood in ferrets, their underlying drivers were explored using personality as a predictor. Using a multi-modal approach of both owner-reported trait ratings and behavioural trait ratings, four distinct dimensions of personality appear to have emerged. This framework of ferret personality was then subjected to numerous validity measures, including inter-rater reliability and test-re-test measures, to assess its reliability for future research and potential applications. Finally, the adaptive significance of personality in a domestic species was explored by examining the intercorrelations with biological parameters, such as morphology and physiology.
For the first phase of this research, owner-reported trait ratings were collected using an online questionnaire consisting of 67 ferret traits (adjectives). Ferret owners (n = 743) then rated each trait on a Likert scale, according to how well they applied to their ferret(s). These were labelled as Subjective Ratings (SR).
The next step was to conduct a series of standardised behavioural tests on a sub-sample of the ferrets (n = 137). In addition to behavioural coding of responses in these tests, such as frequency of some responses and proportion of time spent performing key behaviours, the researcher and an assistant also completed the same trait ratings as the owners, based purely on direct observations from the behaviour tests. These were labelled as Behavioural Ratings (BR).
Principal components analysis (PCA) was then conducted independently on both the Subjective and Behavioural trait ratings. The two separate PCA analyses revealed four personality dimensions underlying both sets of ratings. Given that the traits groupings within each BR dimension were similar to the SR dimensions, they were afforded the same broad labels: SR Extraversion, BR Extraversion, SR Sociability, BR Sociability, SR Attentiveness, BR Attentiveness, SR Neuroticism, and BR Neuroticism.
Despite the similarities in the traits that they encompassed, Pearson’s Correlation coefficients between the SR and BR dimensions were weak, indicating that they did not converge strongly. Given this lack of convergence, both the SR and BR dimensions were explored individually in further analyses.
It was found that all BR personality dimensions were strong predictors of several behavioural variables. From a biological perspective, these correlations provide good indication of construct and discriminant validity in the personality dimensions. To further investigate the validation of the coding method used, inter-rater reliability coefficients were calculated for the researcher and the assistant. These coefficients were found to be high, indicating strong reliability. Conversely, after readministering the behavioural tests to 10 of the ferrets, test-retest values were relatively low suggesting a lack of repeatability in ferret behaviour.
The last stage of this research explored the potential heredity of personality by investigating correlations between the SR and BR ferret personality dimensions, and various morphological parameters including sex, coat colour, markings, total body length (mm), and head width-to-length ratio. It was confirmed that domestic ferrets are highly sexually dimorphic in all body measurements taken, with males being far larger than females. Only Neuroticism (both SR and BR) and Attentiveness were found to be significantly influenced by morphological parameters. These results suggest that personality and morphology are genetically linked to some extent which could suggest that the underlying mechanisms responsible for the adaptive significance of certain traits are still being maintained in domestic populations.
Overall, this research has answered several fundamental questions regarding domestic ferret behaviour which may have important implications for the welfare of this popular companion animal. On a broader scale, we can use these results to make cross-species comparisons and further inform the discussion regarding the evolutionary and adaptive relevance of animal personality.