ABS 2022
Does phenotypic variation affect receiver learning in a Batesian mimicry system?
Avery/L Russell1, Stephanie/R Sanders1, Liam/A Wilson1, Daniel/R Papaj2. 1Missouri State University, Springfield, MO, United States; 2University of Arizona, Tucson, AZ, United States

Flowering plants often offer pollen as a food reward to bees. Since plants benefit by maximizing pollen export to conspecific flowers, plants might cheat on pollen rewards. Often, rewardless female flowers mimic rewarding male flowers (models) on the same plant. Bees should therefore learn to avoid the unrewarding mimics. We tested whether plants make learning more difficult by producing phenotypically variable flowers that cause generalization among models and mimics during learning. We used partially artificial flowers (artificial petals, live reproductive parts) modeled after Begonia odorata to test whether the degree to which the size of models and mimic varied affected how quickly bees learned to correctly identify models and reject mimics. Mimics have 33% longer petals and 31% greater surface area than models, which bees should readily discriminate. Bees rapidly learned to forage less on mimics, but learning was unaffected by flower size variation. This was not a result of altering response speed to maintain decision accuracy. Contrary to our expectation, flower size variation does not appear to be an evolved strategy on the part of plants to exploit pollinator cognition.