Nursing behaviour, postpartum activity and reactivity in sows: Effects of farrowing environment, previous experience and temperament

Abstract

Postpartum behaviour the first 24 h after farrowing (day 0 pp) and on day 10 pp of 40 Landrace/Yorkshire sows was studied in their first and second parity in a crossover experiment regarding farrowing environment (crate vs. get-away-pen (GAP)). The behaviour was compared with the performance of the same individuals during nest building and farrowing, and with their performance in prepubertal tests of behavioural reactivity (open field test and human test) as well as reactivity in the HPA-axis during the open field test. In the first parity, the duration of activity during farrowing correlated positively to the frequency and duration of activity on day 0 pp ( r s =0.54; P<0.001; N=33 and r s =0.47; P<0.01; N=33 , respectively). Being behaviourally active in the prepubertal open field test correlated positively to the number of observations in lateral recumbency on day 10 pp (r s =0.60; P<0.001; N=32) and negatively to the percentage of sow terminated nursings (r s =−0.46; P<0.01; N=32) , both in the first parity. On day 0 pp, the sows in the GAPs, in both parities, had a longer duration of activity, as compared to crated sows ( F 1,28=8.26; P<0.01 and F 1,28=11.01; P<0.01 , respectively). Furthermore, the second parity sows were more active than the gilts, e.g. made more postural changes (F 1,23=5.62; P<0.05) . The duration of lateral recumbency on day 0 pp was longer in the crates in both parities ( F 1,30=7.51; P=0.01 and F 1,30=5.73; P<0.05 , respectively), and the piglets were making more unsynchronised behaviour in the crates ( F 1,30=15.83; P<0.001 and F 1,30=2.53; P<0.05 , respectively). On day 10 pp, exposure of the udder by means of the sow being in lateral recumbency was longer in the second parity (F 1,17=6.28; P<0.05) and in both parities longer in the crates ( F 1,17=7.48; P=0.01 and F 1,17=15.83; P<0.05 , respectively). The proportion of nursings terminated by the sow was significantly higher in the second parity (F 1,17=10.81; P<0.01) and in the first parity the gilts in GAPs ended more nursings (F 1,17=6.37; P<0.05) . The latency to the first reaction in the “screaming piglet test” was shorter for gilts in GAPs as compared to gilts in crates (median: 29 s vs. 0 s; χ 2=11.00; P<0.001 ), as was the latency to an effective reaction in the second parity sows housed in GAPs compared to crates (120 s vs. 44 s; χ 2=6.88; P<0.01 ). In conclusion, activity during farrowing correlated to activity on day 0 pp, whereas activity in prepubertal tests correlated to the sow having less control of the nursing behaviour on day 10 pp. However, maternal behaviour developed over at least two parities, and a stimulating farrowing environment apparently improved the maternal behaviour, probably due more relevant environmental feedback. This was manifested in a progressively earlier development of cyclic nursing on day 0 pp and a better control of the nursing behaviour on day 10 pp by parity and by housing in an enriched environment. These results are promising for the future investigation of consistent individual differences in pigs, including the maternal behaviour.