Indicators on the outside: Behaviour and equine quality of life

In order to evaluate the quality of life of an animal evidence of its subjective experience, ‘how it feels’, is required. For the horse, such information is needed to assess the impact of management and training and identify factors that require change. The aim of this systematic review was to assess the behavioural evidence of emotion in the horse and how this is indicative of equine quality of life. Electronic databases were searched using Science Direct, PubMed, Scopus, Web of Science, PubPsych and PsycINFO. Two separate sets of search terms were specified: 1. (Emotion* OR affective OR cognit*) AND (behav* OR welfare OR physiol*) AND (horse OR Equid* OR equine*); 2. (Horse OR equid* OR equine*) AND (stress AND respons* AND behav). All primary peer reviewed experimental and observational studies referring to equine emotional responses and emotional state were included in the initial pool (n=1446). The application of retention criteria (relevance to the objective of the review; involving > 4 equid subjects; clear description of methodology and justification for the attribution of emotion) resulted in 75 publications being included. Factors found to affect emotional responses included age, social grouping and previous training methods. There was inconsistent use of ethograms with limited justification for behavioural interpretation. Over 30 different journal articles and nine books were cited in relation to ethogram development. Differentiation between behaviours indicative of negative and positive emotional responses was context specific, with head and neck position / movement being most frequently referred to (positive: n=9; negative: n=28). More conclusive signs of response valence included vocalisation, facial expression and movement patterns (approach/avoid/escape /attack/freeze). The valence of underlying emotional state, potentially a better indicator of equine well-being, was implied by changes in the frequency of specific behaviours. Increased feeding behaviour, affiliative social interactions (horse and human) and interest in the environment were indicative of positive emotional state. Negative emotional state was associated with increased repetitive non-functional movement patterns, agonistic interactions, decreased feeding and reduced interest in the environment. Responses to cognitive bias testing and static and dynamic features of facial expression also reflected underlying emotional state. Further validation and the development of an evidence-based ethogram is now required. Equine quality of life is demonstrated during everyday behaviour and responses to training and once the key indicators have been validated, all horse owners and carers should be able to assess and improve the quality of life of horses under their care. Lay person message: This review was carried out to identify equine behaviour that is indicative of underlying mood state and general well-being of horses, i.e. how they feel. Everyday behaviour, including that associated with training, was found to be the clearest indicator of this and will underpin criteria to enable all horse owners and carers to assess and improve the quality of life of the animals under their care.