Laterality and Cognitive Bias: Non-Invasive Indicators for Animal Welfare in the Development of Juvenile Horses (Equus caballus)

• Applicants: Professorin Dr. Konstanze Krüger; Professor Dr. Volker Stefanski
• Subject Area: Animal Breeding, Animal Nutrition, Animal Husbandry ; Biological Psychology and Cognitive Neuroscience ; Cognitive, Systems and Behavioural Neurobiology
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 564183104

There is a growing demand for non-invasive stress indicators to assess animal welfare, both from scientists (prioritizing minimized disturbance to the animals) and from the general public. Behavioral methods, such as observing shifts in sensory and motor laterality (i.e. an expression of stress-induced changes in hemispheric laterality), and a cognitive bias (i.e. shifts from optimistic to pessimistic perception), combined with non-invasive physiological stress indicators, are considered particularly useful for this purpose in horses. However, to date the animals’ status of laterality and cognitive bias after birth and its development at an early age are not satisfactorily elucidated and will, therefore, be the focus of the proposed study. The study analyses three areas of ongoing debate: a) whether motor and sensory laterality are developed at birth in a precocial species such as domestic horses (Equus caballus), and, if so, b) whether stress at an early age may induce changes in sensory laterality, motor laterality and a cognitive biases in horses, and c) whether changes in laterality and cognitive bias from birth up until the first 18 months of a horse’s life will change in the form of a step-like (continuous) increase at distinct stress events, or in wave-like (temporal) increases with recovery periods in between stress events. These changes should also be reflected in changes in physiological stress parameters such as fecal glucocorticoid metabolites and fecal calprotectin. The proposed study is based on a preliminary scientific study at the State Stud Marbach in Germany, supervised by the applicants and with a comparable setting, but with horses at three years of age. For the present experimental approach, 48 test horses will be examined from birth until the 18th month of life. The approach comprises a low stress condition group, exposed only to “unavoidable” stressors at the stud farm (weaning and a change of location), and an elevated stress condition group, exposed to four stressors (a stud show, weaning, and a change in location involving transport). Every three months and during the stress events, 20-minute video recordings will be conducted to allow "all occurrences sampling" of sensory laterality, motor laterality and social behavior. In addition, glucocorticoid metabolites and calprotectin will be analyzed from fecal samples. Cognitive bias tests will be conducted at the 6th, 12th and 18th month of life. Anonymized data will be uploaded to the university's data server, managed with a RDMO data management plan, and finally published on a data repository. For statistical analysis, the R statistical environment will be used when running a principal component analysis for identifying overlapping data at particular eigenvectors, which will then be analyzed with multivariate approaches, MANOVA and GLM for parametric, or GzLM and GLMM for non-parametric data. The research data will be made available by either publication or data access.

• DFG Programme: Research Grants