Differences between horses (Equus caballus) displaying oral and locomotor stereotypies in their responses to a stable mirror.

Recent cross-species analyses suggest that the different forms of stereotypic behaviour have heterogeneous causal mechanisms. In ungulates, there are indications that some forms of locomotor stereotypy may be linked to social factors. Using the domestic horse as a model species within this clade, we investigated whether individuals displaying the locomotor stereotypy box-walking (n = 6) demonstrated greater responses to a stable mirror, intended to mimic presence of a conspecific, than horses displaying the oral stereotypy crib-biting (n = 5) and non-stereotypic horses (n = 5). After a 24 hr baseline phase, horses were provided with a 1 × 1.5 m stable mirror for 24hrs. This was followed by a 24 hr post-mirror phase. Responses were quantified using measures of behaviour, orientation, and ear and head/neck movements. Box-walking, but not crib-biting, was reduced when the mirror was present, and box-walking remained at this lower level following mirror removal. Box-walkers and non-stereotypic horses spent a greater proportion of time orientated towards the mirror than crib-biters. Box-walkers did not, however, demonstrate increased use of ear or head/neck movements associated with affiliative social behaviours, nor did they exhibit more behaviours indicative of frustration, once the mirror was removed, than other animals. There were, however, other unexpected differences in mirror responses between the stereotypy groups. This study provides an initial indication that stereotypic horses respond differently to mirrors, however a sufficiently powered follow-up, with additional controls, is required to establish if box-walking horses are more socially motivated.