Interactions between housed dairy cows during feeding, lying, and standing

Three experiments examined the extent of behavioral interactions between housed dairy cows. In Experiment 1, 6 different cows were selected every other day from a dairy herd and allocated to 3 pairs for recording of feeding, standing, lying, and leg-raising behaviors. There was no difference in the between-pair and within-pair coefficients of variation for the behaviors, suggesting that the performance of these behaviors was independent of influences of other cows. Experiment 2 used 9 pairs of cows in a 3-period changeover design. The same behaviors were recorded as in Experiment 1, and data were used only from the last 2 periods, which ensured that cows had been together previously for at least one 48-hour period. There were again no differences in the between- and within-pair coefficients of variation when using behaviors as replicates, either for the first 2 hours, during which time no treatments had been applied, or the entire 4 hours, when some cows had been prevented from lying down and feeding. Experiment 3 investigated the consistency of between-cow associations for repeated observations of feeding, lying, and entering the milking parlor. Although there was evidence of a similar order of entry into the milking parlor between days, the number of feeding associations was similar to that predicted by a model of random behavior, and the pattern of lying associations also suggested a random pattern of associations. Across behaviors, cows were observed with specific partners more often than expected by chance. Thus although there was no evidence of interdependence of feeding, lying, or standing behavior, consistent associations were observed, particularly in parlor entry.