Qualitative behavioural assessment shows individual variation suggestive of affective state changes in early...

Qualitative behavioural assessment (QBA) is described as a 'whole animal' approach for evaluating welfare through posture and behaviour to judge their emotional state. Although QBA has been applied in dairy cows for a population-level assessment, its use in evaluating individual cattle remains limited. Given the current limitations in evaluating the affective state in cows, validated methods are needed, including the assessment of QBA, for the assessment at the individual level rather than at the herd level and to assess individual changes over time. In this study, 35 individual cows were assessed both within a group setting and through daily individual QBA immediately post parturition to assess behavioural patterns using principal component analysis (mean 12, range 4-19 days). A sub-sample of 20 cows was clustered according to the observed patterns using Gaussian mixture model-based clustering. A large variation in behaviour was observed among cows on an individual basis, with an increase in principal component 1 (PC1) being associated with a more agitated state, and principal component 2 (PC2) being associated with a more positive and playful/inquisitive state and high arousal states, suggestive of affective state changes. Linear mixed-effects models showed days in milk had a significant effect on both PC1 and PC2, with PC1 decreasing as days in milk (DIM) increased (-0.22, p < 0.001) and the opposite for PC2 (0.05, p = 0.01). Cows were clustered based on profile patterns of their PC1 and PC2 components, with some cows showing consistent patterns (n = 13) and inconsistent patterns of behaviour based upon PC scores (n = 7). As manual daily monitoring of cows is unlikely to be practical or feasible, sample frequency analysis was performed, and further analysis showed the potential for less frequent sampling to be able to define these patterns. Our results suggest that individual cow QBA could be useful to identify behavioural changes and pattern profiles after calving, suggestive of an improvement in affective state in housed early lactation dairy cows. QBA at an individual level, compared to group level, could aid welfare assessment, further helping understand dairy cows' affective states around the post-calving period.