Sickness behaviour reduces network centrality in wild vampire bats

Simon P Ripperger; Sebastian Stockmaier; Gerald G Carter

To track dyadic associations among the bats, we used custom-built proximity sensors 85 (see [5, 8, 9] for details). The sensors weighed 1.8 g (including battery and housing) and were 86 glued to the dorsal fur using skin-bonding latex adhesive (Montreal Ostomy Skin-Bond). Tag 87 weights were 4.5 -6.9 % of each bat's mass, in accordance with recommendations for short-88 . CC-BY-NC-ND 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.30.015545 doi: bioRxiv preprint term tracking of bats [10] . We placed the antennas of a base station inside the roost for 89 encounter data download. Each encounter observation includes a duration and received signal 90 strength indicator (RSSI), which can be used as an estimate for a minimum distance between 91 two tagged bats during the encounter. We defined a 'proximity index' as the percent quantile of 92 all RSSI values. To define association, we used a proximity index of 85% (i.e. the top 15% of all 93 encounters ranked by signal strength, Fig. S1 ). We chose this value by using the same RSSI 94 value as a previous study linking wild associations to captive interactions (-27 dbm [8] ). Past 95 work [8] suggests these associations involve a proximity of about 0-50 cm. 96