A continuous recording approach to qualitative behaviour assessment in dairy buffaloes (Bubalus bubalis)

In order to develop a method suitable for a continuous qualitative evaluation of animal behaviour while it changes during the observation period, two recently described techniques, the Qualitative Behaviour Assessment (QBA) and the Temporal Dominant Behavioural Expression (TDBE), derived by the Temporal Dominance of Sensations, were used in combination and applied to 8 Mediterranean buffalo heifers. Animals were subjected to two isolation tests lasting 150s each. One consisted in isolating individual animals from the rest of the group in the indoor part of the home pen, the other in leading animals individually to an outdoor paddock, which was novel to the animals. The behaviour of the animals was video-recorded and the resulting 16 video clips were assessed by a twelve-member trained panel with previous experience in animal behaviour observation. The TDBE allowed the QBA to be performed continuously (C-QBA) during the observation period on a pre-determined list of six behavioural descriptors. Data were subjected to analysis of variance using observer (n = 12), replication (n = 4), animal (n = 16) and the first order interactions as factors. Changes of behavioural expressions during the test were assessed by dividing it into three intervals of 50 s and using the chi(2) one-sample test to check the dominance of each descriptor in each interval. The Wilcoxon test was used to compare quantitative and qualitative behavioural variables as assessed indoors and outdoors. The low level of variance explained for each descriptor by the interactions animal x replication and animal x assessor as compared with the animal (the F of the factor animal was roughly 10 times higher than that of the interactions), showed the intra- and inter-observer reliability of the panel. Indoors, the dominant descriptors of buffalo behaviour (calm, apathetic and curious) were all indicative of a low level of arousal and frequently switched from one another. Outdoors, most of the animals were described with terms indicating a high level of arousal (e.g. Active and Nervous) and only at the end of the test they were generally scored as Curious. Accordingly, the dominance of the descriptor Curious changed during the test when the animals were tested outdoors (chi(2) = 7.00, P < 0.05), whereas no significant changes were observed for all the other descriptors both indoors and outdoors (P > 0.05). The results indicate that C-QBA, if combined with the appropriate tools (i.e. TDBE), may be able to follow and describe the variations of animal behavioural expressions in time. (C) 2015 Elsevier B.V. All rights reserved.