Effects of handling on fear reactions in young I celandic horses

Summary Reasons for performing the study Inclusion of objective temperament tests at practical horse breeding evaluations is of increased interest. It has been debated whether such tests may involve human handling, since there may be considerable differences in horses' handling experience. Objectives To investigate the effect of a short‐term standardised handling procedure on reactions of young horses in 2 types of fear tests (including and excluding human handling). Study design An experimental study with 3‐year‐old Icelandic horses (n = 24). Methods Handled horses (n = 12) were trained according to a standardised handling procedure whereas controls (n = 12) remained untrained. Behavioural and heart rate responses in a novel object test and 2 handling fear tests ( HFTs ) were measured. The HFTs were conducted with both an unknown ( HFT ‐unknown) and a known handler ( HFT ‐known). Results There was no effect of the handling procedure on the horses' behavioural and heart rate responses in the novel object test, nor in the HFT ‐unknown. In the HFT ‐known, however, handled horses showed a significantly shorter duration of reluctance behaviour compared with controls, whereas heart rates did not differ. Heart rates correlated significantly between tests. Conclusions Previous handling may affect the behavioural fear response of horses when handled by their usual handler, whereas this effect did not apply to an unknown handler. Heart rates appeared unaffected by handling and may be a more reliable indicator of fearfulness. Known handlers may ‘mask’ behavioural responses of horses in fear tests and thus handling by a known handler during testing may not be appropriate for objective evaluation of fearfulness in a practical situation.