Compliance with Twitch Restraint in Horses

Angela J. Davies
University of Massachusetts and University of Pennsylvania

The twitch is used as a method of restraint in horses. It applies pressure to the upper lip causing an endorphin release, which effectively increases the horse's tolerance of aversive procedures. The twitch is variably affective among horses. The varying behavioral response among horses to the twitch raises the question of whether previous experience with the twitch affects behavioral response. I

In Experiment 1 the behavior of twitch-naïve ponies was compared with that of twitch-experienced horses during use of the twitch. This experiment was based on our idea that horses with past twitch experience have learned to associate the twitch with either positive or negative experience. The twitch was applied in a gentle manner and kept in place for 3 minutes, during which a simulated ear clipping procedure was performed. The trials were videotaped, and behavior measures of compliance derived. Difference in compliance between twitch-naive and twitch-experienced animals were not significant (p<0.05, independent t-test).

A second experiment compared the effect of a series of three gentle twitch trials on the behavior of twitch-naïve and twitch-experienced horse and ponies. Control group subjects received a similar series of trials in which they were simply tied for 3 minutes with the experimenter present and no application of the twitch. Change in behavioral measures of compliance during twitching was not different between groups (p<0.05, independent t-test). Data were also analyzed based on improvement vs. no improvement for the twitch-naïve ponies and twitch-experienced horses using Fisher's Exact Test. There were mixed results. Some were improved, some were not. Differences were not significant.

For both experiments the findings were contrary to our expectations. This may indicate that our hypotheses were incorrect or that there were confounding factors. The type of subjects used could have affected results, as could the use of the humane twitch, which was not fitted for use with ponies. Also, the simulated clipping procedure could have been too great of a stimulus for the purposes of this study. In Experiment 2, the small number of subjects and the low number of trials performed may have been inadequate to detect a change in behavior.