Effect of early life and current environmental enrichment and personality on attention bias in pigs

Animals may show increased attention towards threatening stimuli when they are in a negativeaffective state, i.e. attention bias. A barren, stimulus-poor housing environment can inducestress and potentially a negative mood in pigs. Apart from current housing conditions, however,also the early life environment and personality characteristics might influence affective state.In this study, we aimed to investigate the effects of early life and current housing conditionsand personality (coping style) on attention bias in pigs. Pigs (n=128, 32 pens) housed inbarren or enriched housing in early life (B1 vs E1), experienced either a switch in housingconditions at 7 wks of age or not (creating B1B2, B1E2, E1E2 and E1B2 treatments). Theywere classified using a backtest as ‘high resister’ (HR, proactive coping style) or ‘low resister’(LR, reactive coping style) at 2 wks. Pigs were subjected to a 3-min attention bias test at 11wks of age. Half of the pigs were exposed to a 10-sec potential threat (T) and the other halfnot (C) in a test room with food in the centre. Attention towards the (location of the) threat,vigilance, eating and vocalisations were recorded. Firstly, behaviours of T and C pigs over thetest were compared. Secondly, for T pigs, effects of early life and current housing, coping styleand their interactions on behaviour during and for 150 sec after the threat were tested. Mixedmodels with random pen effects were used, except for squealing for which a Fisher’s exact testwas used. T pigs spent more time on vigilance behaviour (T: 13.6±1.4, C: 6.8±1.0%, P<0.001),less time on eating (T: 15.0±1.8, C: 27.8±2.4%, P<0.001), were more likely to squeal (T: 22%C: 6% of pigs, P<0.05) than C pigs, and paid more attention to the location of the threat (T:7.1±0.6, C: 0.5±0.1% of time, P<0.001) throughout the 3-min test, indicating that pigs didrespond to the threat. During presence of the threat, HR pigs showed more vigilance (P<0.05),particularly in E2 housing (E2-HR: 39.9±6.6, E2-LR: 6.7±2.9, B2-HR: 19.4±5.9, and B2-LR:12.1±4.4%, interaction P<0.05). E1-HR pigs (55.4±6.5%) tended to pay more attention to thethreat than E1-LR pigs (30.3±5.9%), with levels of B1-HR (46.4±6.8%) and B1-LR (48.3±7.6%)in between (interaction P<0.10). After presence of the threat, no effects of housing or copingstyle on vigilance, attention to location of the threat or eating were found. E2 pigs grunted moreoften than B2 pigs (9.6±1.7 vs 3.6±0.9 per min, P<0.01). E2 pigs were also more likely to squealthan B2 pigs (P<0.05), particularly the HR pigs (E2-HR: 50%, B2-HR: 0%, E2-LR: 21%, B2-LR:17%, interaction P<0.10). In conclusion, housing affected vigilance in a personality-dependentmanner during a short period of exposure to a potential threat. We found no strong effect ofearly life or current housing on attention bias after the threat, but current housing conditionsand personality did affect vocalisations.