Severity assessment from an animal's point of view

Severity assessment in animal experimentation is a complex biomedical and ethical issue. While the requirements for laboratory animal science regarding assessment of severity are getting more demanding, criteria for such an assessment are still biased by uncertainty.The interpretation of physiological and behavioral measures in relation to animal welfare is difficult and often reflecting an educated gut feeling rather than scientifically sound conclusions. Here, we argue for the importance of including the animals' perspective into severity assessment. Preference tests are a straightforward approach in asking the appraisal of different goods. However, preference for one good over another does not necessarily indicate suffering if the access to the preferred good is denied. Nor can the overall severity of an experimental measurement be derived solely from the fact that the animal would rather not participate in such a procedure. To gain a better understanding of the valence of choices made, there is demand for sophisticated tests which allow estimating the strength of the respective preferences. Animals usually cannot choose to avoid experimental procedures and such experiences leave traces in internal affective states. These emotion-like states can be revealed using tests of cognitive bias, which ask the animal if future expectations are "optimistic"or "pessimistic". Advancing these methods on testing cognitive bias in mice allows a comprehensive severity assessment taking internal affective states into account. The set of measures proposed here include the animal's point of view in severity assessment with regard to their preference and valence of future expectations.