reported in literature; and to test the validity of these as predictors of horses’ suitability as police horses. Reliability of response to novel object, social isolation and handling tests was established by tests repeated at three week intervals on 33 horses. The majority of the tests were only moderately reliable (Weighted Kappa (Kw) .0.4) though the reactivity tests showed good or better reliability (Kw .0.6). These tests were included in the validity testing that com- pared the horses’ responses to the behavioral tests with their handlers’ ratings of key aspects of police horse performance, developed from interviews with instructors and yard man- agers. Bonferroni corrected correlations between the horses’ behavioral responses to the tests and subjective ratings of suitability for their use revealed only moderate associations between the reactivity tests and suitability. These findings suggest that standardized behavioral tests are not strongly reliable in horses and that either the tests found in the literature and used in this study do not serve as predictors of suitability for use, or that behavioral responses to standardized tests cannot be generalized to the wider challenges faced by police horses in their roles. Reactivity tests did reveal stronger reliability and may have some validity in measuring a horse’s suitability. Overall, the findings of this study cannot support claims made for these (or similar) tests to be valid predictors of a horse’s suitability for use. Key words: horse; behavior test; reactivity test HORSE PERSONALITY PROFILES AND PERFORMANCE Das ˇa Dasha Grajfoner 1, *, Elizabeth J Austin 2 , Franc ¸oise Wemelsfelder 3 1 18/2F3 Waverly Park, EH8 8ET Edinburgh, Scotland 2 The University of Edinburgh, Psychology Department, 7 George Square, Edinburgh, EH8 9JZ, Scotland 3 Scottish Agricultural College, Bush Estate, Penicuik EH26 0PH, United Kingdom *Corresponding author: Dasha@Grajfoner.com Horses play a significant role in animal-assisted therapy, they are trained and employed by the police and military, and they are used in sports. In all of these situations horses’ suitability and performance are fundamental. Behavioral or personality traits may determine their success. Horses used in therapy are required to be placid and calm, friendly and approachable, which may not be desirable for horses in sports. We report on the personality profiles of individual horses, rated as high and low performers by their trainers. Thirty-eight horses from three different stables were rated on a nine point single item performance scale (‘‘1 5 poor performer – 9 5 excellent performer’’) and horse personal- ity traits, generated by 24 familiar and experienced raters, using Repertory Grid Technique. Inter-rater agreement on horse performance was significant for all horses (Stable 1: mean W 5 0.242, p , 0.05, Stables 2 and 3: mean W R 0.375, p , 0.001). Personality profiles of the highest and lowest performers indicate that most high per- formers were rated as ‘‘nice’’, ‘‘gentle’’, ‘‘easy to work with’’, ‘‘easy to handle’’ and ‘‘patient’’, and most low performers as ‘‘difficult to handle’’, ‘‘inexperienced’’, ‘‘immature’’, ‘‘tense’’, ‘‘nervous’’, ‘‘scared’’, ‘‘hesitant’’, ‘‘impatient’’ and ‘‘unsociable’’. Individual personality profiles of the best and the poorest performers in each stable indicate that it is not only one personality trait but rather a combination of traits that is associated with high or low performance in horses. Key words: equine personality; performance; horse personality profile; Repertory Grid Technique SALIVARY AND FECAL CORTISOL AS MEASURES OF STRESS IN HORSES Tamsin Hughes 1, *, Emma Creighton 2 , Rob Coleman 2 1 Anthrozoology Unit, Chester Centre for Stress Research, University of Chester, UK 2 School of Agriculture, Food and Rural Development, Newcastle University, UK *Corresponding author: t.hughes@chester.ac.uk Valid measures of stress are needed in horses to determine over-all stress levels and to identify stress triggers, used to ensure management is kept within the animals’ ability to cope and welfare is not compromised. Levels of circulating cortisol reflect HPA-axis activity, and excretion into saliva and feces allow non-invasive sampling. We validated an enzyme-linked immunoassay (ELISA) for horse salivary and fecal cortisol and validated these as indicators of acute and over-all stress levels in riding horses. Saliva was swabbed every 30-minutes over three days in N 5 15 horses with eight in light exercise. Feces were collected from N 5 9 working horses on stabled workdays and at rest at grass for three consecutive weeks. Immunological validity of the ELISA was demonstrated by high specificity, accuracy, precision and sensitivity. Biological validity of salivary cortisol was demonstrated by diurnal decline and elevation post-exercise both mirroring known patterns in plasma cortisol; and by a trend towards elevation following 10 minute exposure to a known stressor. Fecal cortisol was biologically validated by decline between working and rest days. Large individual differences in assay values were found and not all individuals followed the group means. Salivary cortisol was labile, and although it has a close tem- poral relationship to circulating cortisol, measures may be confounded by environmental disturbance, pulsatile release patterns and diurnal rhythm. Careful sampling protocols are therefore needed. Fecal cortisol as an index of circulating cortisol has a 24-hour time lag to excretion, and collection protocols must evenly sample total fecal mass due to un- even hormone distribution and be frozen immediately post-excretion to avoid degradation. With careful sampling, salivary cortisol may be used to measure acute stress re- sponses to identify stress triggers, and fecal cortisol may Abstracts 59