Unrevealing components of risk assessment: escape behaviour and refuge use in a web-building spider.

Optimal escape models assume that decisions to escape after a predation attempt depend on the initial risk assessment. However, few studies have explored animals' decisions to return to normal activities after their first escape and how the ecological context influences these behaviours. To address these gaps, we investigated the influence of substrate type on escape behaviour and refuge use in the spider Aglaoctenus castaneus. Individuals forage on sheet webs but hide in funnels after detecting potential threats. Spiders from two populations build webs either on injurious substrates that can harm predators or on noninjurious substrates. The dominant injurious substrates include a large thorny palm species in the 'thorny palm' population and a razor grass in the 'razor grass' population. We hypothesized that spiders constructing webs on injurious substrates would (1) display reduced sensitivity to approaching predators by hiding in their funnels and (2) would return to their sheet web more quickly after detecting a predator. We also assessed whether these behaviours were correlated across and within populations and whether they were affected by the social context (couples or solitary individuals). Spiders inhabiting injurious substrates exhibited shorter flight initiation distances (FIDs) compared to those inhabiting noninjurious substrates. Interpopulation variation was observed, with spiders from the thorny palm population displaying shorter FID and lower hiding time than those from the razor grass population. However, there was no correlation between FID and hiding time. Spiders from the thorny palm population showed less sensitivity to predators, probably because thorny palm location was safer than the razor grass location. In conclusion, A. castaneus can adjust risk-taking behaviours depending on the damaging potential of the substrate for predators. We propose new hypotheses with the aim of improving risk assessment research. • We investigated the effects of substrate type on risk assessment of A. castaneus. • We observed spiders in two populations: thorny palm vs razor grass dominant. • Spiders on injurious substrates had shorter flight initiation distances (FIDs). • Hiding time was not correlated with FID. • Spiders in the thorny palm population had shorter FIDs and hiding times. [ABSTRACT FROM AUTHOR]