Abstract: Most horses are trained by negative reinforcement. Currently, however, no standardised test for evaluating horses' negative reinforcement learning ability is available. The aim of this study was to develop an objective test to investigate negative reinforcement learning in horses. Twenty-four Icelandic horses (3 years old) were included in this study. The horses were tested in a pressure-release task on three separate days with 10, 7 and 5 trials on each side, respectively. Each trial consisted of pressure being applied on the hindquarter with an algometer. The force of the pressure was increased until the horse moved laterally away from the point of pressure. There was a significant decrease in required force over trials on the first test day (P<0.001), but not the second and third day. The intercepts on days 2 and 3 differed significantly from day 1 (P<0.001), but not each other. Significantly stronger force was required on the right side compared to the left (P<0.001), but there was no difference between first and second side tested (P=0.56). Individual performance was evaluated by median-force and the change in force over trials on the first test day. These two measures may explain different characteristics of negative reinforcement learning. In conclusion, this study presents a novel, standardised test for evaluating negative reinforcement learning ability in horses.

Abstract: Horses have been considered to rely on human gesticular cues (McKinley and Sambrook 2000, Anim Cogn 3:13-22; and recently Maros et al. 2008, Anim Cogn 11:457-466), however large variability among individuals tested in two-ways object choice tasks was found. Part of the horses in those studies (40 and 26 %, respectively) even failed to pass adequately through the training session which preceded the testing phase and served to learn a horse to carry out a task. Therefore, we alternated the experimental design designed by McKinley and Sambrook (reduced number of testing trials to 10 per horse to keep its attention, applied just one, a dynamic-sustained pointing cue with touching the bucket, etc.), and tested an effect of training method, sex, age, and learning on proportion of correct choices. We hypothesised, that horses trained by “traditional” method (TTM) will get lower score than those experienced with “horsemanship-based” methods (HTM), being characterized by closer and more frequent human-horse contact and also extended exercising “from the ground” with frequent using of arms cues. Despite simplification of the methods, only about 60 % of tested horses passed through the training phase (i.e., learned to come to and upturn the bucket with hidden treat). Successful completion of training phase was reached regardless of age or sex of a horse, but by the training method; HTM horses ran better compared to TTM ones. No significant effect of age, sex, or learning (i.e., trial order within all 10), and training method as well was found on proportion of correct trials in the testing phase. Horses made a correct choice in more than 70% of trials. Individual scores ranged from 50 to 100 %. In conclusion, horses showed high level of comprehension of human pointing gesture, regardless of their sex or age. No effects of training method or learning process within a test suggest low impact of handling and learning on the level of comprehension at least of the most vivid human pointing gesture. Horses trained by methods based on “natural human-horse communication” did enhance cooperation with people.

Abstract: Bringing the head and neck of ridden horses into a position of hyperflexion is widely used in equestrian sports. In our study, the hypothesis was tested that hyperflexion is an acute stressor for horses. Salivary cortisol concentrations, heart rate, heart rate variability (HRV) and superficial body temperature were determined in horses (n = 16) lunged on two subsequent days. The head and neck of the horse was fixed with side reins in a position allowing forward extension on day A and fixed in hyperflexion on day B. The order of treatments alternated between horses. In response to lunging, cortisol concentration increased (day A from 0.73 ± 0.06 to 1.41 ± 0.13 ng/ml, p < 0.001; day B from 0.68 ± 0.07 to 1.38 ± 0.13 ng/ml, p < 0.001) but did not differ between days A and B. Beat-to-beat (RR) interval decreased in response to lunging on both days. HRV variables standard deviation of RR interval (SDRR) and RMSSD (root mean square of successive RR differences) decreased (p < 0.001) but did not differ between days. In the cranial region of the neck, the difference between maximum and minimum temperature was increased in hyperflexion (p < 0.01). In conclusion, physiological parameters do not indicate an acute stress response to hyperflexion of the head alone in horses lunged at moderate speed and not touched with the whip. However, if hyperflexion is combined with active intervention of a rider, a stressful experience for the horse cannot be excluded.

Abstract: Since the time of domestication, humans have trained horses for the purpose of serving man. Different training methods have been developed throughout the centuries; some were developed with consideration for the horse's welfare, while others disregarded welfare to a great extent. Most present day training is based upon making the horse perform a desired behaviour through dominance and subordination. Although cooperative training techniques have gained popularity, everyday training lacks the application of learning theory or neglects the horse's learning capacities and their species' specific behaviour. Thus, the horse's welfare may be jeopardised.
The aim with this review is to consider methods that allow an objective assessment of the welfare of horses undergoing training. The review gives a brief insight into the history of horse training and handling. It proceeds with an overview of the horse"s learning abilities which is argued to be of paramount importance for effective training. The review then describes a few selected training techniques that are used today, based on negative and positive reinforcement, and discusses parameters from which it could be possible to assess the welfare of the ridden horse. The work concludes with suggestion for future

Abstract: In the last decade an increasing number of studies has been oriented towards equine social learning and their social behavior within the herd (Kruger‚ 2006-2008). In social species, social learning is important to learn and gain useful skills to move and live in their own social and environmental context. Group housing has been recognized as an important element to fulfill the physical and behavioral needs of horses, especially their need for social contact (Søndergaard‚ 2011). Still‚ when it comes to horse training, the social aspect and‚ in general‚ cognitive abilities of the horse are rarely taken into account. Although it is widely accepted that social isolation is stressful for horse (Mal et al, 1991a and 1991b) still isolating a young horse is the first step when it comes to training methods. Due to tradition and culture and our performance-oriented society it is both difficult to accept and apply a different social/cognitive training approach. Training sessions are focused on immediate results whereas in cognitive learning part of the process is latent and will not be visible immediately‚ but taking the cognitive skills into account plays an important role in avoiding tension both in the horse as in the human-horse interaction (Baragli and De Giorgio, 2011). In this study we tested the possibility to apply social learning by creating a social environment‚ favoring a cognitive learning approach‚ for the training of six young horses. The group existed in three males and three females, between two and three years old. All six showed initial difficulties and defense to human interaction. They were housed in two groups in adjacent spacious paddocks where they had ample opportunity to move and express their individual and social behavioral repertoire. Each horse had one training session per week without isolating it from the others. The training sessions were held following a cognitive-relational model defined as the equine-zooanthropologic approach (De Giorgio, 2010 – Marchesini, 2011). The learning objectives were to be able to handle each horse‚ conduct it‚ saddle and ride it within a maximum time-frame of two years. Every time a defensive or alert behavior would occur the training activity was re-arranged to not over-pressure the horse. Therefore the persons working with the horses carried out the activities without tight expectations focusing on the horses’ positive attention. After eighteen months all six horses were used to the saddle and to riding. None of the horses ever fled or showed defense behavior and in the case of unexpected events they showed no emotional reactivity/reactive behavior. Today the horses show the same calm behavior whenever worked individually. This preliminary study highlights how social learning applied to equestrian activity can be fundamental for safety and welfare and the establishment of a more problem-free relationship between horse and human. Safety as the defensive behavior seems to have been reduced and welfare as the horses have been trained in a social context without being isolated and thus without being stressed during the training experience.