Information-use strategy affects costs and benefits of foraging in ants with tandem recruitment (Temnothorax nylanderi)
Simone Monika Glaser, Simone Monika Glaser , Christoph Grüter
Institute of Organismic and Molecular Evolution, Johannes-Gutenberg University, Mainz, Germany; Institute of Organismic and Molecular Evolution, Johannes-Gutenberg University, Mainz, Germany ; Institute of Organismic and Molecular Evolution, Johannes-Gutenberg University, Mainz, Germany
Animals have to continually acquire information about their environment, including food sources, nesting sites or dangers. They can collect new information through costly individual trial-and-error sampling, i.e. scouting, or, alternatively, obtain information from observation of other individuals, i.e. social learning. Afterwards, they can rely on private information. Social learning is often used in social insects like bumblebees, honeybees and ants during foraging. For example, some ant species teach food source locations to their nestmates via tandem running. We hypothesized that the information-use strategy affected quality and the time of discovery of food sources. To test our hypotheses we experimentally manipulated both the number (2 vs. 10) and the variability (0.1 vs. 1.0 M sucrose solution) of food sources. We expected that (1) social learning is more often used and more time efficient when food sources are hard to find and (2) that social learners find better food sources than individual learners. We found that if there were few food sources, scouting was relatively rare and tandem running more likely. However, actual use of social information was quite rare as foragers mostly relied on private information to return to known food sources. When food sources were scarce, follower were equally likely to learn about good and bad food sources. When food sources were common, social learners discovered above-average food sources, most likely due to information-filtering by tandem leaders. Scouts were more likely to switch food source than followers, especially after finding low-quality food sources. Our results confirm that the use of social learning increases reward quality, but these effects depend on how common and variable food sources are.