# Are Users Good Assessors of Social Dominance in Domestic Horses?

## TL;DR

This study shows that experienced horse practitioners can accurately assess social dominance among horses, which is important for rider and horse safety.

## Contribution

The study confirms practitioners' ability to detect dominance-submissiveness relationships and identifies specific behaviors linked to these rankings.

## Key findings

- Practitioners showed moderate to perfect agreement in ranking horses by dominance.
- Submissive horses ate less during feed confrontations and showed more withdrawal behaviors.
- Rolling when denied feed was identified as a redirected frustration behavior.

## Abstract

It is vitally important that horse riding instructors know which horses are dominant and which are submissive to prevent injuries resulting from social conflict during rides. Eight equine practitioners were asked to rank 20 horses according to their behaviour known from previous work and non-work observations. In pairs, the horses were also confronted with a limited amount of hay to test their dominance and associated behaviours. The time spent eating hay and submissive behaviours in the test were strongly related to each other, showing that dominant horses were eating hay for a longer time than submissive ones. The raters were moderately concordant in their ranking of horses, but when familiar with a group of horses they were able, consciously or unconsciously, to detect the dominance–submissiveness relationships between animals. This is key to assuring the safety of both horses and riders.

Horse users and caretakers must be aware of the risks of mixing social groups. The current study investigated whether eight equine practitioners can assess the social dominance rank of 20 horses. The horses’ feeding time and agonistic/aggressive and submissive behaviours were observed during the feed confrontation test, and the dominance index (DI) was calculated. Kendal’s W, Spearman correlations and factor analysis were applied to test the raters’ agreement, the relationship between dominance ranks and the behavioural variables, and to determine the clustered behaviours. The agreement between all raters in the classification of dominance order ranged from moderate to perfect. The ranking by every rater was strongly and negatively correlated with the time of eating in feed confrontation tests and with the DI, evidencing shorter feeding times for more submissive horses. The withdrawal of the horse when threatened was the behavioural variable that was most often correlated with raters’ ranking. The current study confirmed the abilities of practitioners to categorise the horses under their care according to their social interactions. Additionally, rolling when denied access to feed was proposed as frustration-releasing (redirected) behaviour.