New ways of assessing emotions in farm animals
Caroline Lee (view profile) and Jessica Monk
The objective assessment of animal welfare is an important first step towards improving the life experience of animals. Current methods for assessing animal welfare largely focus on quantifying the biological cost to the animal, examining their physiological and immunological responses to potentially negative events and environments. Studies of animal behaviour have also been used to indicate more obvious welfare states such as pain or discomfort, or to determine preferences for different resources. While these methods are all useful for assessing animal welfare, they do not necessarily tell the whole story. A major knowledge gap is our capacity to assess the affective state of animals (i.e. how they feel) in the various situations in which we farm them. Developing methods which can measure affective states (emotions) is internationally recognised as being important for the assessment and improvement of animal welfare.
As animals are unable to speak, assessment of their affective states is particularly challenging and requires innovative methodologies to gain insight into their minds. The challenge has been to achieve this insight in a scientifically rigorous way. Recently, there has been a strong research focus on tests that measure cognitive biases, which have been shown to indicate emotional states in humans and animals. In particular, judgement bias tests have been widely used in animal welfare research. Judgement bias is where the emotional states of humans and animals alter the way they interpret ambiguous cues. By measuring the responses of animals to ambiguous cues in a defined test, inferences can then be made about their underlying emotional states. Such tests have been developed in a range of farm animal species, including pigs, sheep, cattle and poultry, however a key limitation of the methodology is that it requires extensive training periods and is therefore impractical in applied contexts.
The CSIRO Welfare Team have recently been investigating attention bias tests as a more practical cognitive method to assess affective states in farm animals. It has been shown that humans and animals in increased states of anxiety show increased attention towards threats, known as an attention bias. Based on this concept, we have developed an attention bias test for farm animals that can be conducted without prior training. The test involves moving an animal into an arena containing a food source and a potential threat (previous location of a dog), then assessing the degree to which the animals attention is captured by that potential threat. Using drugs to induce states of increased and decreased anxiety, we confirmed that the test reflects states of anxiety as predicted. Both sheep and cattle showed increased attention towards the previous location of a dog after receiving an anxiogenic drug compared to a control or anxiolytic drug. Further studies have confirmed that attention biases can be detected in laying hens and that longer term states such as depression can also be identified using the test. Overall, we have shown that measurement of attention bias has the potential to provide a practical indicator of various short and long term affective states in animals. The ability to measure affective states provides a window into the mind of the animal and will provide a more complete assessment of farm animal welfare.