Is the judgment bias test a good tool to assess the quality of horse management?

Animal welfare should include the possibility of animals experiencing positive emotions. Emotions influence the cognitive process, and judgment bias tests (JBTs) are employed in different species, to assess the optimistic or pessimistic expectation of an individual and its affective state. Only recently the JBTs have been applied to horses. This research aims to investigate the relationship between a spatial JBT and hypothalamic-pituitary-adrenal axis (HPA) chronic and acute activation in forty-one animals hosted in different kinds of environments: traditional stables (TS), natural boarding (NB), and ethological stable (ES). Fecal (FC) and horsehair (HC) cortisol concentrations were quantified for each subject through RadioImmuno-Assay (RIA). Body condition score (BCS), as an indirect index of animal motivation towards food, and personality traits were measured to explore their possible influence on JBT results. Horses had to distinguish a positive position (P), where a bucket full of food was positioned, from a negative one (N), with an empty bucket. Then, 3 intermediate positions (Near Negative-NN, Near positive-NP, and Medium-M) with an empty bucket were presented to the subject one at a time. Only 20 subjects out of 41 completed the JBT and were included in the statistical analysis, and both BCS and P position, whether at the right or the left of the subject, seems to have influenced the inclusion rate. Only the ES group registered a significantly lower score in NN, suggesting a more optimistic affective state, whereas NB and TS did not significantly differ in their responses. Despite this, horses from NB recorded higher FC concentration than TS subjects during all the phases of the test, but lower HC levels, which could suggest a generally lower level of chronic stress but its interpretation presents several confounding factors. These results put into question whether JBT is indeed a good test to monitor the quality of the management, as it does not seem to reflect the chronic physiological state of the animals and could be influenced by a state of acute stress, caused by the test procedure. Due to these confounding factors, this procedure should be accompanied by other indicators. Finally, to include more animals and exclude possible biases, the structure of the JBT and the employment of food as a reward should be evaluated considering the peculiarities of the species and individual motivations.