Ali, A. B. A., Gutwein, K. L., Heleski, C. R. 2017. Assessing the influence of upper lip twitching in naive horses during an aversive husbandry procedure (ear clipping). Journal of Veterinary Behavior 21, 20-25.

When engaging in procedures that horses may find aversive, restraint methods are often used to help ensure safety of both horse and handler. Twitching is a common restraint method, but its use is sometimes considered controversial. Based on previous work, there is evidence supporting the concept of the twitch acting more nearly like acupuncture. Our aim was to compare behaviors, heart rate (HR), heart rate variability (HRV), and ease/time of completing procedure when naive horses were ear clipped with or without a twitch. Clipping out the inside of a horse's ear is considered a routine competition-preparation procedure for many horses, particularly in the United States. Many horses, however, find the procedure aversive and make attempts to avoid it. Eight Arabian horses (4 mares, 4 geldings; mean age, 2.8 ± 1.0 years) that were naive to ear clipping and twitching were tested during fall 2014. Baseline HR and HRV data were collected. Horses were randomly assigned to either being clipped with a twitch (ECT1) or clipped without a twitch (EC) as their first treatment, then vice versa for their second treatment. Later, each horse was twitched a second time and clipped (ECT2). Time elapsed to perform ear clipping, behaviors, summed ear clipping reactivity score, HR (beats per minutes), time domain indices of HRV (standard deviation of the beat-to-beat interval and root mean square of successive beat-to-beat interval), and power spectral analysis indices (high frequency [HF], low frequency [LF], and LF/HF) were recorded. Analysis of variance was used to compare different groups using SPSS, version 17.1 (SPSS Inc, Chicago, IL). Mean HR varied from 42.7 ± 1.6 beats per minutes (baseline) to 107.7 ± 6.2 (EC), 71.2 ± 3.5 (ECT1), and 59.4 ± 1.3 (ECT2) (baseline differed from EC, and both differed from ECT1, P < 0.05); mean LF/HF ratio varied from a mean of 1.5 ± 0.9 for baseline to 7.2 ± 0.2 (EC), 3.0 ± 0.3 (ECT1), and 2.1 ± 0.1 (ECT2) (baseline differed from EC, and both differed from ECT1, P < 0.05). Time to complete the task and the ear clipping reactivity score varied from 120.5 ± 13.1 seconds, 12.1 ± 2.5 for EC to 69.0 ± 7.4 seconds, 3.6 ± 1.3 for ECT1 and 48.4 ± 6.8 seconds, 1.1 ± 0.7 for ECT2 (EC differed from ECT1 and ECT2, P < 0.05). EC resulted in the strongest HR and HRV indicators of stress, the most behavioral indicators of aversion, and took the longest time to complete. Furthermore, the horses' second exposure to the twitch did not show evidence of increased aversion to the twitch. Based on the evidence in this study, we believe that lip twitches, when properly applied, should be considered a viable humane restraint for short usage situations. Our results should not be seen as an endorsement to use a twitch in place of careful methodical training.