Personality modifies the welfare effects of shelter screens in farmed blue foxes

Consistent differences in individual behaviour, commonly referred to as animal personality, can influence welfare through their interaction with management practices. This study investigated whether farmed blue fox females exhibit personality traits along a shyness-boldness axis and how these traits relate to adrenocortical activity under different housing conditions. Test-retest consistency and intra‑ and inter‑assessor repeatability were first confirmed for the behavioural tests (Stick test, Approach test, Titbit test), which were repeated monthly over six test sessions. Reliability analyses (Cohen's kappa, Prevalence-adjusted bias-adjusted kappa, and intraclass correlation coefficient) showed high assessor agreement and greater behavioural consistency in adults than in juveniles. Following these reliability analyses, individual‑level behavioural estimates were obtained as BLUPs from mixed‑effects models accounting for repeated measures and random effects. Because BLUP estimates from the three tests were correlated, principal component analysis was applied to the Z‑standardised BLUPs, and the first principal component was used as a composite shyness score. Juveniles (n = 38) and adults (n = 37) were analysed separately. Shyness scores were examined in relation to faecal cortisol metabolite (FCM) levels and housing conditions in a 2 × 2 factorial design (basic vs. enriched environment; with vs. without a shelter screen). Linear models revealed that individuals with higher shyness scores also had higher FCM levels in both age groups. In juveniles, shelter screens were associated with increased shyness scores but reduced FCM levels in consistently shy individuals, indicating that visual barriers may mitigate stress by enabling these animals to avoid human contact. This effect was weaker in adults, and enrichment showed no personality-dependent effects in either age group. Overall, farmed blue fox females exhibited personality traits along a shyness-boldness axis, with shyness associated with elevated adrenocortical activity. Accounting for personality improved the interpretation of stress indicators and highlighted that housing modifications, such as shelter screens, may be particularly beneficial for shy individuals. Incorporating personality-informed approaches into welfare studies can therefore guide management strategies that better accommodate individual differences.