Dairy cows' motivation to nurse their calves dataset
Description
For seven weeks, dairy cows were housed with their calf on either full-time (full; 23 h/d, 28 pairs) or part-time (part, 10 h/d, 27 pairs). Additionally, 26 cow-calf pairs were used as a control treatment, and had no contact. Over two weeks, week 8 and 9, half of the contact pairs (full and part) experienced reduced contact (RC), to first approx. 50% of original contact (9.5 h for full, 4.5 h for part) and then approx. 25% of original contact (4.5 h for full, 2 h for part). The other half continued with unchanged contact (UC). The cows' motivation to nurse their calf during week 8 and 9 was tested in a maximum price paid (MPP) test, where cows had to pass through an increasingly heavier gate for the opportunity to nurse their calf. The weight on the gate increased by 1.0 bar at a time to a maximum test resistance at 8.0 bar. Cows were interpreted as having reached their MPP when they failed to pass the gate two days in a row. As an alternative, the cows could pass a gate to obtain partial contact (PC), or they could remain in the start box of the arena, obtaining no contact (NC). The gate leading to PC remained at a low resistance throughout the testing. We hypothesised that cows with reduced contact would express reduced motivation to nurse their calves, and thus have lower MPP than cows with unchanged contact. We also expected that motivation would affect the trainability of the cows, meaning that RC cows would be more likely to fail training than UC cows. Cows' latency to pass the gate was recorded, and it was recorded whether the calf suckled during the test or not. Additionally, we hypothesised that cows would utilise the option of partial contact, when the weight on the gate leading to nursing surpassed their maximum price. This was recorded for cows with two consecutive fails. Furthermore, it was recorded whether cows had attempted to reach their own calf before they failed. Contrary to our hypothesis, cows with RC had higher MPPs than cows with UC (significantly for full, numerically for part). Full UC cows had similar MPP as no-contact cows. RC calves were more likely to suckle. Fewer no-contact cows than any of the other treatments passed training. We found no indications of preference between PC and NC for the cows' first fail, but cows were more likely to chose NC in the second fail. No relationship was found between cows attempting to reach the nursing opportunity and what alternative they eventually chose. MPP was analysed using Cox' proportional hazards mixed effects models, trainability and the relationship between trying own calf and chosen alternative were analysed using pairwise Fisher's exact tests, likelihood of nursing was analysed using Chi^2, and choice of alternative was analysed using binomial tests. Latency was only descriptively analysed.