Individual variation in temporal activity patterns in open-field tests

The open-field test, which measures animal activity or distance covered by individuals subjected to a novel arena, is used frequently in behaviour studies. Studies generally report consistent individual differences in activity level in the open-field, but no study has tested whether individuals change their activity level differently within a single open-field test. If such differences exist, then the reliability of activity levels measured over the whole test may be affected by the duration of the test. Here, we present analyses showing how a reaction norm approach can be used to account for individual differences in temporal activity patterns. Specifically, we measured individual temporal patterns of activity in wild eastern chipmunks, Tamias striatus, and house mouse, Mus musculus, strains by decomposing the total duration of open-field tests into successive shorter time intervals. Open-field tests on chipmunks and mice lasted 3 min and 5 min, respectively. We detected repeatable individual and strain differences in temporal activity patterns. Activity during the first time interval (at the beginning of the test) was positively correlated with activity measured over the whole test, meaning that short tests yield a valid measure of activity in chipmunks and mice. We recommend analysing open-field data using random regressions to obtain activity measurements that are robust against differences in temporal activity patterns.