As experiences and emotions are private and subjective, they can’t be accessed directly. Instead, scientists must use indirect measures which appear to reflect the animal’s emotional state.
While physiological measures, such as heart rate, are objective and may reflect levels of arousal, it is hard to assess the valence of such measures (the attractiveness or aversiveness of a situation). For example, heart rate increases in the presence of food (positive situation) but also when an animal is frightened (negative situation).
Behavioural measures, such as how long it takes an animal to approach food, enable scientists to gain a non-intrusive representation of an animal’s state, but they are often hard to interpret and subject to observer bias. For example, animals may approach things that they fear as well as things that they desire.
It is clear that both behavioural and physiological measures have their limitations, and as a result we must develop more accurate ways of assessing animal welfare. One way we can do this is to consider how an animal responds to a change in reward, known as reward sensitivity.
Find out more about reward sensitivity here.