Judgement Bias During Pregnancy in Domestic Pigs

In humans and rats, changes in mood and affect are known to occur during pregnancy, however it is unknown how gestation may influence mood in other non-human mammals. This study assessed changes in pigs’ judgment bias as a measure of affective state throughout gestation. Pigs were trained to complete a spatial judgement bias task with reference to positive and negative locations. We tested gilts before mating, and during early and late pregnancy, by assessing their responses to ambiguous probe locations. Pigs responded increasingly negatively to ambiguous probes as pregnancy progressed and there were consistent inter-individual differences in baseline optimism. This suggests that the pigs’ affective state may be altered during gestation, although as a non-pregnant control group was not tested, an effect of learning cannot be ruled out. These results suggest that judgement bias is altered during pregnancy in domestic pigs, consequently raising novel welfare considerations for captive multiparous species.


42
affective state is altered throughout gestation (for review see [2]) and it is clear that 43 pregnancy impacts maternal affective state.

45
Understanding an animals' affective state, or emotion and mood, is a key component of 46 animal welfare [10]. Affective state can influence and alter cognitive processes, such as judgement, [11][12] which may then be used to infer and understand an animals' affective 48 state. Cognitive bias or judgement bias is the influence of affect on information processing, 49 with more content individuals likely to make positive assumptions about ambiguous stimuli 50 [13]. Judgement bias tests have been used to assess changes in affective state in a range of 51 species, including pigs, dogs, honeybees and European starlings [14][15][16][17]. Research typically 52 focuses on the impact of external stimuli on judgement bias; this is likely to act via alteration 53 to the internal, physiological environment ultimately resulting in changes in behaviour and 54 judgement bias [11; [18][19]. As such, we would expect internal stimuli, such as physiological 55 changes, would also impact judgement bias directly even in the absence of external 56 influences. Pregnancy is one of the biggest physiological changes a mammal may experience, 57 involving major hormonal and cognitive adjustments [20][21], yet little is known of how 58 information processing and affective state may change in relation to pregnancy in animals.

60
The domestic pig (Sus scrofa domesticus) has been used as a human model in a wide range of 61 medical research such as infectious disease [22], nutritional [23] and neurological studies [24].

62
Pigs allow for longer lifespan studies and are more similar to humans than other laboratory 63 species, such as rodents [25][26] Figure 1). Pigs consistently approached the positive probe quickly and 170 the negative probe slowly (or not at all), getting generally slower during gestation (Figure 1).

171
However, whilst the mean speed of approach was fairly linear between positive and negative 172 pre-and early gestation (Figure 1a,b), by late gestation, pigs showed a shift towards pessimism, 173 such that the positive probe continued to be approached quickly but ambiguous probes were 174 approached more slowly (Figure 1c).

176
All models retained all interactions and gave qualitatively similar results. The best model was 177 model 1, where the intercept was allowed to vary for each pig at each gestation time (Table 1). However, the result for model 2, where the intercept was allowed to vary for each pig at each 179 gestation time, within each replicate, was equally well supported (delta AIC <2;   as the large physiological changes as associated with pregnancy, also have the potential to 216 influence affective state and therefore judgement bias. The aim of this study was to assess 217 judgement bias in domestic pigs throughout gestation. It was hypothesised that the gilts would 218 be more pessimistic during pregnancy than prior to mating, as indicated by an increase in 219 latency to approach the ambiguous cues. Our results showed this to be the case, with the gilts 220 taking longer to approach the ambiguous locations in the later stage of gestation than before 221 mating which indicates that judgement bias changed as gestation progressed. This was most 222 apparent at the middle and most ambiguous location ( Figure 1) and suggests the pigs were more 223 pessimistic during the late gestational stage. Crucially, the latency to reach the positive location 224 did not vary markedly throughout gestation, showing that other changes, for example, increase 225 in weight, did not affect response latencies (Figure 1)