Hucul horses’ learning abilities in different learning tests and ue the association with behaviour, food motivation and fearfulness

Abstract

Research on learning in domestic animals is often based on food-rewarded tasks. It is still unclear however, how much intrinsic factors such as food motivation and fearfulness influence performance in food-rewarded operant tasks. This study aimed to investigate the association between learning ability in different tests, food motivation and fearfulness. Twenty-three young, naive Hucul horses were tested. The learning tests included a negative reinforcement test (NR), where the horses were trained to yield to rope and hand pressure; a clicker test (CT), where the horses were taught to follow a target for a food reward; and a visual discrimination test (VD), where the horses had to learn to recognise a correct bucket (containing food). Food motivation was estimated through a free feeding test, where the latency to finish eating a standard meal was measured. In the fear test, a plastic bag was waved in front of the horse in a standardised manner to induce a startle response. The escape distance and latency to resume eating was measured. Further, heart rate (bmp) and salivary cortisol (ng/ml) were measured before and after each learning test and the fear test. All horses reached the learning criterion in the learning tests (number of sessions to reach criterion (mean ± SD): NR 1.87 ± 0.63; CT 1.74 ± 0.96; VD 5.00 ± 1.71). No correlations were found in learning performance between the different learning tests, and food motivation did not appear to influence the speed of learning. Behavioural reactions in the fear test correlated moderately to performance in the CT test (e.g. sessions to reach criterion and escape distance, rs = 0.43, P = 0.04), suggesting that fearfulness may affect performance in some types of learning tests. In line with previous studies, we conclude that learning performance differed between the various types of tests. We further conclude that food motivation – measured as latency to consume a meal – did not influence performance in the food-rewarded tests. This may suggest either that the small food rewards delivered in the tests triggered food motivation in all horses regardless of general food motivation; that our test was inappropriate for assessment of food motivation; or that other intrinsic and extrinsic factors govern performance in learning tests.