Day Three of the ISES 2026 Conference brought the meeting to a close with research spanning riding school welfare, equipment and communication, learning and affective state, working equids, training, movement and positive human–horse interactions.
Lisa Ashton began by introducing the ABRS+ Riding School Futures Project, which brings evidence-based equine welfare directly into operational riding schools. Built around the question “What is the horse experiencing?”, the project uses frameworks including the Five Domains Model and ISES First Principles to support changes in staff understanding, training and everyday management, while recognising that horse welfare and sustainable riding schools can work together.
Prof. Paul McGreevy introduced the COMPASS Guidelines, a new research standard specifically addressing studies involving behaviour modification and equipment. The guidelines aim to strengthen study design, biological grounding and ethical rigour while helping research translate more effectively into practice.
Prof. Dr. Uta König von Borstel presented data from more than 6,000 Icelandic horse competition starts. Cavessons incorporating a lower noseband were associated with a greater likelihood of oral lesions, while the large dataset also allowed relationships with bit design and lower-limb lesions to be explored.
Dr. Kathrin Kienapfel-Henseleit explored equipment use and conflict-related behaviour in Standardbred trotters. Equipment changed substantially between warm-up and pre-race phases, alongside increased head movements, while less restrictive equipment tended to be associated with better race placing.
Susanne Lundesjö Kvart compared riders’ perceptions when using snaffle and sidepull bitless bridles. Bitted bridles were generally perceived as providing clearer, more precise communication, whereas bitless riding increased awareness of seat and leg aids and was associated with greater perceived relaxation in some horses.
The programme then moved into learning and affective state.
Dr. Cathrynne Henshall investigated how reward learning and changes in expected rewards may influence mood in Thoroughbreds. The findings provided preliminary evidence that learning experiences can influence horses’ subsequent affective states, suggesting that learning is not emotionally neutral and may leave an emotional footprint beyond the training session.
Dr. Camie Heleski explored human judgement of equine emotional states as a foundation for developing AI welfare assessment tools. Agreement was strongest when identifying negative affective states, while neutral states were substantially harder for people to classify consistently, highlighting an important challenge when creating reliable datasets for AI.
Dr. Johanna Henke-von Der Malsburg asked whether horses can “learn to learn.” Horses taught several negative reinforcement tasks in parallel did not learn their original task less effectively and appeared to acquire additional tasks more efficiently, suggesting structured task variation may facilitate learning rather than cause confusion.
Lara Klitzing investigated young Thoroughbreds before and after a 12-week pre-training programme. Horses became more active and exploratory after training, while individual differences became more pronounced, demonstrating why novel object responses should be interpreted using multiple behavioural measures rather than a single indicator.
Dr. Franziska Pilger compared physiological stress responses to pre-training events in yearling and two-year-old Thoroughbred fillies. Yearlings did not experience the carefully conducted programme as more physiologically stressful than two-year-olds, providing no evidence of a stress-related benefit from delaying pre-training by one year.
Debora Cinganelli investigated whether social companionship could buffer horses’ responses to a novel stressor. Horses accompanied by either a familiar human or a familiar horse had lower heart rate and cortisol responses than horses facing the stressor alone, providing evidence for both inter- and intra-specific social facilitation.
Prof. Dr. Kathalijne Visser compared horses following therapeutic and conventional riding lessons using cognitive and physiological measures. Therapeutic riding produced no evidence of a more negative affective state and was associated with lower physiological arousal, although differences in physical workload may contribute to the physiological findings.
The afternoon moved into working equids and free papers.
Pantaleo Gemma assessed 100 working horses in Bishoftu, Ethiopia, alongside their owners’ knowledge and attitudes. Despite generally positive owner perceptions, lack of water and shade, skin problems and rough handling revealed important differences between perceived and observed welfare.
Dr. Mactar Seck presented Brooke West Africa’s Compassionate Handling Project in Senegal. Through management changes, positive reinforcement, systematic desensitisation and other compassionate approaches, challenging behaviour disappeared in 70% of cases, while 79% of owners showed positive changes in their own behaviour.
Lina Nyberg explored how experienced harness racing trainers and drivers understand driveability. Participants described a driveable horse as calm but alert and responsive, with calm, progressive training viewed as important for welfare and performance. Driveability emerged as something shaped by the horse, driver, training and context, rather than simply a fixed characteristic of the horse.
Dr. Renata Farinelli Siqueira challenged the assumption that locomotor asymmetry necessarily represents dysfunction. Different Quarter Horse disciplines showed repeatable discipline-specific patterns of asymmetry, suggesting that some lateralised movement may represent functional adaptation to athletic demands rather than pathology.
Dr. Sophie Côté presented preliminary work exploring memory reconsolidation as an intervention for established stereotypic behaviours. Stereotypic behaviours reportedly ceased across 25 behavioural targets following an average of approximately four sessions, although controlled research is needed to establish the effectiveness of the approach.
Finally, Taru Niittynen explored whether horses experience scratching differently depending on who provides it. Horses showed more behavioural indicators associated with enjoyment when scratched by familiar people, while privately owned horses initiated more contact with humans than riding school horses. Biomarkers including oxytocin and cortisol were also being investigated to explore whether behavioural responses are reflected physiologically.
Across the final day, a common theme emerged: understanding horse welfare requires asking not only what humans are doing, but how the horse experiences it. From riding schools and competition equipment to learning, social support, working equids and even scratching, Day Three demonstrated how equitation science can translate that question into practical changes for horses and the people who work with them.
And with that, ISES 2026 came to a close, but not before one final announcement: the ISES International Conference will be heading to Portugal in 2027! 🇵🇹
