Effect of training method and dairy cow personality traits on adaptation to an automated milking system

Training dairy cows on the automated milking system (AMS) can improve their performance, however, the ideal training method and how their individual personality relates to training is unknown. The objectives of this study were to determine: 1) the effect of AMS training method on adaptation of primiparous cows to an AMS, 2) whether the personality of a cow affects their adaptation to the AMS, and 3) whether certain personality types adapt better based on the type of training they receive. Forty-eight primiparous Holstein dairy cows, who had not previously been milked by an AMS, were enrolled in the study at ∼85 DIM and were assessed for personality traits using a combined arena test comprised of 3 parts: a novel arena test, a novel object test, and a novel human test. Cows were allocated into 1 of 4 AMS training treatments (n = 12): control (CON; no training), environment exposure only (ENV), environment and AMS concentrate provision (PEL), and environment and mechanics exposure (MEC). Cows in the ENV, PEL, and MEC treatments were subjected to training 4 × /d for 4 d, and then cows from all 4 treatments were milked on the AMS in the evening of the 4th training d. Milking activity and production were recorded for the first 14 d on the AMS. PEL cows had fewer visits to the AMS, fewer milkings, and fewer voluntary milkings compared with CON cows across the 14 d. Cows on all experimental treatments had more fetching events than CON cows, while MEC and ENV treatment cows had a higher milk yield compared with those on the CON treatment. Principal component analysis of the behaviors recorded during the personality assessment revealed 2 factors interpreted as personality traits, “explore-active” and “bold.” Across all cows, more “explore-active” cows had greater milk yield. Only the performance of ENV cows was associated with personality; ENV cows who scored low for “explore-active” differed from CON cows, through increased visits and voluntary milkings. Overall, the results demonstrate that providing concentrate on its own during AMS training may not be motivating enough to improve adaptation and performance. However, exposing cows to the mechanics (e.g., the robotic arm and sounds of the AMS) before first milking may improve milk yield. Finally, cow personality may influence the performance of cows trained using less complicated training methods, such as only exposing them to the environment of the AMS.