Qualitative Behavioural Assessment of Pigs using Free Choice Profiling

Participating observers were 12 MSc students in Applied Animal Behaviour and Animal Welfare, who all had a general understanding of animal behaviour, but no specific expertise in pig behaviour. These observers scored pig behavioural expression in the OF and EPM following protocols as developed for Free Choice Profiling (FCP) methodology (Wemelsfelder et al., 2000, 2001 ). The first stage of the FCP method requires that observers generate their own descriptor lists for use in subsequent assessments. For this 16 demonstration clips (1 min long), deemed to be representative of the range of different behavioural styles and demeanours observed in the tests, were selected (eight for each apparatus) and shown to the observers. All demonstration clips and the subsequent study clips included sound. The 16 clips were shown to observers in two batches with a 10 min break between each batch. Each batch consisted of both OF and EPM footage but no more than two clips of the same apparatus were played in succession. The reason for showing recordings of both types of test in a single term generation session was that a single vocabulary would be produced that was applicable across both tests. After each individual clip, observers were given 2 min to write down as many descriptive terms for the observed pig as they thought were needed to adequately characterise that pig's behavioural expressions. All the terms generated in this way were then collated for each observer and used to create each observer's individual list of terms. When individual observers used both positive and negative antonyms (e.g. “confident” and “unconfident”, “comfortable” and “uncomfortable”), only the positive term was kept for use in subsequent scoring. The number of terms generated by each observer ranged from 11 to 43. For each observer's score sheet, the terms were arranged so that successive terms had contrasting meaning with (as far as possible) similar terms being listed further apart.
In the second stage of the FCP process, the observers scored the full set of clips, each using their individual list of previously generated terms. This scoring took place over four sessions on different days. On each day, observers were shown a batch (11 or 12 clips) of OF clips and a separate batch (13 or 14 clips) of EPM clips, with a break between batches. The order of OF and EPM batches was rotated each day. Clips were allocated to a session so that sessions were as far as possible balanced for whether pigs were drug-treated or not, by litter group, and recording time within the observation. The open field clips were also balanced for order of testing since the pigs were tested in the apparatus twice. Following viewing of each individual clip, observers were asked to quantify, for each term, the degree of expression shown by the pig, by marking a vertical line on a 125 mm visual analogue scale, ranging from minimum to maximum possible expression. Observers were unaware of any prior drug or other treatment applied to the pigs. They were instructed that the purpose of the experiment was to investigate and compare behavioural expression in the 2 different types of test.
Prior quantitative assessment of pig behaviour in these tests (Donald et al., 2010, 2011 ) identified three main factors that change with Azaperone treatment: activity, vocalisations and exploration. To provide a comparison with the qualitative measures of behaviour, these same quantitative measures of behaviour (see Table 1 for a summary Ethogram) were recorded (using event logging software: Observer 5.0, Noldus Information Technology) from the same 1 min long clips used for QBA.