Open Access

Jesús Rafael Hernández-Montero

• Bats spend half of their life at roosting sites. Hence, exploring for potential roosts is an essential task for their survival, especially for those species which switch roosts regularly, such as several temperate bat species. However, localizing new roosts is a difficult task due to bats’ sensory limitations (e.g., vision, echolocation range). To compensate such constrains, it has been hypothesized that bats rely on cognitive processes like associative learning, spatial memory, social information use and memory retention for an efficient roost localization. However, no previous study has assessed these cognitive skills under natural conditions. The aim of my thesis was to assess how individually RFID-marked, free-ranging bats use different cognitive processes when localizing suitable day roosts. For this purpose, I used a pairwise roost-quality (suitable vs. unsuitable) choice experiment with automatic monitoring and assessed bats’ cognitive processes according to different cues available. Cues were echo-reflective (spectral signature of boxes), spatial (position of the box within the experimental pair) and social (presence of conspecific at roosts), each one linked to a different cognitive process. I found that Bechstein’s bats (Myotis bechsteinii) used associative learning to discriminate between suitable and unsuitable newly placed boxes according to their echo-reflective cues. However, when individuals returned to known suitable roosts, they relied more on spatial memory to localize them. This was evidenced by the higher proportion of visits to the unsuitable boxes after swapping box positions within the same experimental pairs. When social cues were available, bats discovered a higher number of suitable roosts and re-localized previously occupied roosts more accurately. Taken together, Bechstein’s bats used multiple cognitive processes and prioritized one process over another depending on the relevance of the cues and search context. Memory retention of the learned association was analyzed one year later, after the bats had returned to their breeding sites from their hibernacula. I found no evidence that individuals remembered the association between roosts’ suitability and their respective echo-reflective cue. The lack of memory retention could be attributed to hibernation or the duration of the period that the bats spent away from their summer habitat without the opportunity to reinforce the association contingencies. Nevertheless, bats quickly relearned the same association in a short period of time. This emphasizes the high behavioral flexibility of the bats. Given the ability of Bechstein’s bats to quickly learn to discriminate roosts based on their external echo-reflective cue via associative learning, I investigated whether the use of echo-reflective cues improves box detectability and further occupancy. This was also assessed in free-ranging Natterer’s (Myotis nattereri) bats and the brown long-eared bats (Plecotus auritus). I found that the use of echo-reflective cues did not improve the detectability and occupancy of newly placed boxes despite the previous experience of the colonies with such cues. There were differences among species in the number of discovered boxes, visits and roosting days. These differences could be related to the species-specific explorative behavior and roost-switching behavior. Box supplementations programs aimed to conserve or relocate bat colonies should consider these behaviors to increase their likelihood of success even when bat colonies are used to roosting in artificial shelters. My research underlined the importance of evaluating multiple cues under natural conditions to understand how natural selection has shaped the cognitive process used for localizing resources. Cognitive field studies are logistically challenging given the number of factors to control. However, automatic monitoring techniques like the one used in this study give the possibility to deepen the understanding of the cognitive ecology of animals. I finally discuss two venues of further research to understand the spread of information within colony members about novel roosts and the recruitment dynamic to novel roosts.