Mind the trap: large-scale field experiment shows that trappability is not a proxy for personality

Behavioural tendencies vary consistently among individuals and this variation is known as personality. Previous studies have found that personality traits measured through standardized behavioural tests predict trappability (i.e. 'trap happy' versus 'trap shy'). However, the nature of this relationship is unclear since it has been explored only within single species and never across environments. This is problematic because trappability is a labile characteristic that can vary between seasons, environments and years. It is essential to understand this link because there is great potential for the use of trappability as a proxy for personality. For example, if trappability reflects personality, this would allow researchers to extract personality data from long-term capture-mark-recapture data sets. To clarify this relationship, we designed a large-scale field experiment to measure personality and trappability in five small mammal species and across four distinct forest types. With an open field test, we quantified behaviour in 189 deer mice, Peromyscus maniculatus, 170 southern red-backed voles, Myodes gapperi, 42 American red squirrels, Tamiasciurus hudsonicus, 58 woodland jumping mice, Napaeozapus insignis, and 87 northern short-tailed shrews, Blarina brevicauda. We identified personality in all five of our target species, and through mixed-effects modelling we found that personality traits did not predict different aspects of trappability. Furthermore, trappability was not a repeatable measure (i.e. animals that were trap happy in one session were not necessarily trap happy throughout the trapping season). Our results suggest that trappability cannot be used as a proxy for personality.