Can we predict behaviour in pigs? Searching for consistency in behaviour over time and across situations

Individual differences in animal behaviour could elucidate the differences in stress coping style, which have consequences for production, health and welfare. Therefore, individual behavioural differences in pigs and consistency of responses in different test situations were studied. If differences in behaviour reflect coping characteristics, then behaviour in one situation should predict behavioural reactions in other situations and at other times. In this study, a backtest was performed on 315 Great Yorkshire* Dutch Landrace piglets at 3, 10 and 17 days of age. On day 3, groups of approximately 10 piglets per sow were formed, based on escape behaviour in the first backtest (backtest score): high resisting (HR, all scores >3), low resisting (LR, all scores <3), miscellaneous (MISC, various scores between 0 and 10) or original (OR) litters to determine if group composition would influence coping behaviour. In weeks 5–7 and/or 10–12, a human approach test (HAT), a novel object test (NOT), and an open door test (ODT) were performed with all pigs simultaneously, in the home pen. Pearson correlation coefficients were calculated between the test results and a factor analysis was performed. Furthermore, data were analysed on pen level, and within MISC- and OR-pens on animal level, using multivariate linear models. Significant correlations were found between the backtests and between HAT, NOT and ODT. Backtest results on three ages loaded on the same factor, and HAT, NOT and ODT at one age also loaded on one factor. No differences were found in HAT, NOT and ODT for the different pens (HR, LR, MISC and OR). On animal level, animals with higher backtest scores also had higher HAT scores at 5–7 weeks (P