Looking on the bright side of bias—Validation of an affective bias test for laboratory mice
Publication Type: |
Journal Article |
Year of Publication: |
2016 |
Authors: |
Dana Marie Graulich, Sylvia Kaiser, Norbert Sachser, S. Helene Richter |
Publication/Journal: |
Applied Animal Behaviour Science |
Keywords: |
*cognitive bias, affective bias test, animal welfare, emotion assessment, environmental enrichment, laboratory mice |
ISBN: |
0168-1591 |
Abstract:
Based on the concept that experience-dependent memory may be biased by affective state at the time of learning, the affective bias test represents a novel approach for the assessment of emotions in animals. The task uses a within-subjects study design, where animals encounter two independent discrimination tasks on separate test days and learn that digging in one out of two substrates is rewarded with food. While one pair of substrates is associated with an affective manipulation, the other functions as a neutral control. Potential affective biases are then quantified by a preference test in which both initially rewarded substrates are tested against each other. Since the test was originally developed for the work with rats, the aims of the present study were (1) to translate the paradigm from rats to mice, and (2) to investigate the impact of an environmental enrichment on affective state in male C57BL/6J mice. For this reason, two control groups were included in the study that served to validate the overall testing procedure. One control group experienced equal absolute reward values for both rewarded substrates, assuming that this would prevent a substrate preference. In the second control group, we tripled the reward for one substrate, hypothesizing that mice would bias their choice towards this higher rewarded substrate in the preference test. Indeed, animals of the first control group did not show any substrate preference, while subjects of the second control group preferred the substrate associated with the high reward. Thus, mice were able to learn the association between absolute reward value and respective substrate, indicating that the affective bias test has the potential to work in mice. Besides these control groups, a third group was implemented in the study: As for the first control group, the absolute reward value was the same for both rewarded substrates, but in addition to that, affective state at the time of learning was modified for one substrate by subjecting mice to a highly enriched cage directly after testing. Since enrichment is known to influence the behaviour and physiology of rodents mostly in a positive way, we expected to observe a preference for the substrate associated with the enriched cage. However, in contrast to our expectations, mice did not show any substrate preference, suggesting that the additional enrichment provided in the enriched cage was probably not sufficient to induce measurable changes in an affective state.