23 26 2
Could computers tell us what our horses are thinking and feeling? British researchers are working on software which could help us do just that.
The work is being led by Dr Steve North, a computer scientist with a passion for horses who has embarked on an interdisciplinary animal-computer interaction research project.
North, a research fellow in the University of Nottingham’s Mixed Reality Laboratory, is developing a Horse Automated Behaviour Identification Tool (HABIT), which is new animal-computer interaction software.
It aims to automatically identify horse behaviour from amateur video so humans can interpret each animal’s reactions and understand why they are happening. It has the potential to indicate whether a horse is stressed, sick or suffering. It could be used on farms, at home, in zoos and in veterinary practices.
The research is being carried out in collaboration with Dr Mandy Roshier, an expert in anatomy and behaviour at the university’s School of Veterinary Medicine and Science; Dr Carol Hall, an expert in equitation science at Nottingham Trent University; and Dr Clara Mancini, a senior lecturer in computer interaction design at the Open University.
Human-computer interaction is already well established. The emerging scientific discipline of animal-computer interaction is focused on the relationship between animals and technology.
The HABIT project is pulling together experts in animal computer interaction, equitation science, ethology and animal behaviour.
North received funding from the laboratory, which is based in the School of Computer Science, for an Engineering and Physical Sciences Research Council exploratory project to develop his ideas and bring together the interdisciplinary team of experts.
“Horses and all nonhuman animals are entitled to interaction technologies that enrich rather than exploit,” he says.
“Anthropocentrism limits our understanding of human interaction in a multispecies world and currently there isn’t any software that can reliably analyse video footage and log what behaviours it sees and when. We hope HABIT will also be able to assess how animals react to new surroundings.”
Mancini, who heads the Animal Computer Interaction Laboratory at the Open University, said: “We are coming to a point where technology is so widespread in society that animals are becoming exposed to it and interacting with it.
“However, we are still in the very early stages of developing technology that can interact with them in a user-centered way.”
North’s recent paper, “Do androids dream of electric steeds? The allure of horse-computer interaction”, was published in the academic journal Interactions in March. A further article is due to be published in Interactions this month.
Hall and Roshier have just published an article which considers the challenges involved in measuring and interpreting animal behaviour. “Getting the Measure of Behaviour – is seeing believing?” was written for a special section on ACI in the July-August edition of Interactions.
Hall is based at Nottingham Trent University’s School of Animal, Rural and Environmental Sciences at the Brackenhurst Campus. Her research focuses on the behaviour and welfare of horses, in particular ridden horses. She is a member of the International Society for Equitation Science, a director on the board of the National Equine Welfare Council, and is a British Horse Society Intermediate Instructor.
“This project has the potential to speed up the process of behavioural analysis and increase the objectivity and accuracy of recording behaviour,” she says. “Currently, behavioural assessment is largely based on human judgment leading to issues of consistency and experimenter bias.”
Roshier’s research interests include biomedical engineering, biomechanics, animal communication, human-animal interactions, and animal behavior in veterinary medicine. She is a committee member of the British Veterinary Behaviour Association.
“It is really important that our vet students can interpret what an animal is telling you through its body language,” Roshier says.
“This can go some way to understanding its emotional state. Using equipment that can help us measure and understand behaviour would provide important insights into how we can communicate with animals more.”