Active and explorative individuals are often restless and excluded from studies measuring resting metabolic rate: Do alternative metabolic rate measures offer a solution?

It has often been proposed that bolder, more explorative or more active individuals also have a higher resting metabolic rate (RMR), indicating metabolic costs of these personality types. However, such individuals might often be restless and thus excluded from RMR datasets, leading to a significant sampling bias. We tested (1) whether such a bias occurs when animals are measured for a relatively common but short time period of 3h, and if so, (2) whether alternative measures of metabolic rate, that allow the incorporation of non-resting individuals, would reveal associations between metabolism and personality. For this, we studied free-living individuals of the African striped mouse (Rhabdomys pumilio) both during the moist season (N=25 individuals) with high food availability and the dry season (N=48 individuals) with low food availability. We assessed variation in the latency to explore a novel object, and the time spent active and time spent in the centre of a neutral arena. We examined links between personality and (i) RMR and (ii) four alternative metabolic rate (MR) metrics: average MR, highest MR, lowest MR and span of MR. Twenty-nine percent of the measured individuals had to be excluded from our RMR study because they remained restless during respirometry trials. Striped mice showed a behavioural syndrome where fast explorers also spent more time in centre and more time active than slow explorers. Individuals that did not rest during respirometry trials were faster explorers and in the moist season, they were also more active and spent more time in the centre than individuals that rested. We found no relationship between RMR and the behavioural syndrome, which might be due to the exclusion of individuals with a certain behavioural type, leaving a subset of compliant individuals. In the moist season, we found positive relationships between the behavioural syndrome and span of MR and lowest MR. In the dry season, low food availability may mask links between the behavioural syndrome and alternative MR measures due to reduced overall activity in striped mice. Our study demonstrated the importance to consider personality when measuring RMR and suggests that some alternative MRs may be useful to examine relationships between metabolism and personality when it is impossible to measure individuals over extended time periods.