In recent years, consistent individual differences in behavior, or personalities, have been a topic of increasing interest as researchers strive to understand and predict the responses of individuals and populations to anthropogenic changes. Behavioral studies in wild populations often require that animals are live trapped before behavioral observation can occur, and this is especially true in studies investigating animal personalities. However, it is unknown whether the amount of time confined to a live trap may regulate the behavior of trapped individuals. Specifically, if the duration of trap confinement directly influences behavior, then by obtaining wild animals through live trapping we may be confounding the very measurements of greatest interest. To investigate whether the duration of trap confinement influences the behavior of trapped individuals, we performed a study on two small mammal species, focusing specifically on personality traits. We positioned high-definition trail cameras facing Longworth small mammal traps in the field to observe capture events and record the time of capture. We then measured personality in captured deer mice (Peromyscus maniculatus) and southern red-backed voles (Myodes gapperi) using three standardized tests, and through linear and generalized linear models we found that the time an animal had spent confined to a trap before testing did not affect 86% of behaviors exhibited. Our results showed two weak behavioral effects of confinement duration on boldness and docility resulting from an interaction between the duration of confinement and whether or not an individual was naïve to trapping. Our results suggest that behavioral measurements of wild, trapped small mammals are not determined by the time spent confined to a trap. However, researchers should use caution and consider whether an animal is naïve to trapping during analysis since habituation to the live trap may play a role in the effects of confinement duration on behavior. OPEN ACCESSCitation: Brehm AM, Tironi S, Mortelliti A (2020) Effects of trap confinement on personality measurements in two terrestrial rodents. PLoS ONE 15(1): e0221136. https://doi.org/10. Observations 43� Only results from the top model (based on AICc scores) are shown. We have omitted occasions where the null model was the top model. See materials and methods for more information.

8In recent years individual differences in the behavior of animals, or personalities, have been 9 shown to influence the response of individuals to changing environments and have important 10 ecological implications. As researchers strive to understand and predict the responses of 11 individuals and populations to anthropogenic changes, personality studies in wild populations 12 will likely continue to increase. Studies of personality in wild populations often require that 13 animals are live-trapped before behavioral observation can occur; however, it is unknown what 14 impact live trapping may have on the behavior of trapped individuals. Specifically, if the 15 duration of trap confinement directly influences behavior, then by obtaining wild animals 16 through live-trapping are we confounding the very measurements we are most interested in? To 17 investigate this question, we performed a study using two small mammal species. We positioned 18 high-definition trail cameras on Longworth small mammal traps in the field to observe capture 19 events and record the time of capture. We then measured personality in captured deer mice 20 (Peromyscus maniculatus) and southern red-backed voles (Myodes gapperi) using three 21 standardized tests. With a repeatability analysis, we confirmed which behaviors could be 22 considered personality traits, and through linear and generalized linear models, we found that the 23 time an animal had spent confined to a trap before testing did not affect the majority of behaviors 24 exhibited. Our results showed two weak behavioral effects of confinement duration on boldness 25 and docility depending on whether an individual had been trapped previously. Our results 26 suggest that personality measurements of wild, trapped small mammals are not determined by 27 trapping procedures, but that researchers should control for whether an animal is naïve to 28 trapping during analysis.Over the past few decades, the acknowledgement that many species of animals display 32 consistent individual differences in behavior, or personalities, has become widespread (1-4). 33Personalities are heritable (5), have consequences for fitness (6-9), and can limit the ability of 34 individuals to exhibit behavioral plasticity (10) resulting in trade-offs where certain personality 35 types perform well in some ecological contexts but not in others (11). Because individual 36 personalities can determine the response of individuals to changing environments (12,13) and 37 have important ecological implications (14-16), personality studies in wild populations will 38 likely continue to increase as researchers strive to understand and predict the responses of 39 individuals and populations to anthropogenic changes (17)(18)(19)(20). 40Studies of personality in wild populations usually require that wild animals are live-41 trapped so that one or more standardized behavioral tests can be undertaken (21-24) but see (25) 42 for a method of personality observation in non-captured animals. Because being trapped may 43 induce st...

No abstract