Observational learning of foraging strategies in Blue Jays

Background/Question/Methods

Observational learning has been shown to have broad ecological and evolutionary implications in areas such as mate choice, predator avoidance, and foraging.  Within the context of foraging, observational learning can allow exploitation of novel resources, increase foraging efficiency, or otherwise increase individual success. In this experiment, we examine intraspecific observational learning as it relates to foraging strategies, using captive Blue Jays (Cyanocitta cristata) as a study system. We compare the time it takes a non-observer to learn a foraging task with the time it takes a naive "observer bird" to perform the same task after watching the model for one trial.  Further trials monitored learning and behavior of the model and observer.  We used a well-board as a simple proxy for a natural foraging substrate a Blue Jay might encounter.  Model birds with extensive well-board experience attempted to access food hidden under paper covers in the wells while observers watched. During a 10-minute interval between model and observer trials the well-board was reset to the pre-model bird state before introduction of the observer bird.  We measured number of wells exploited, latency to uncover the first well, and before/after weight difference as a conservative estimate of food eaten.

Results/Conclusions

Preliminary results show that observers successfully learned to forage in covered wells faster than non-observers. Specifically, observers uncovered wells within 5 minutes of first exposure, while non-observers needed more than 2 cumulative hours of exposure to uncover their first well. Multiple strategies were observed for obtaining food from the wells (e.g., ripping the paper covering off, folding it inward). Consistent with the idea of imitation, observers initially copied the specific strategy used by their model, though they often shifted from the model’s strategy to a different one as they gained additional experience with the foraging task. Interestingly, it was noted that observer birds often avoided wells that they had seen model birds open, especially as trials progressed.  The presence of food cannot explain this trend; all wells were refilled with food and re-covered before testing. This suggests that observers keep track of which specific locations have been exploited. Future studies will address these latter trends in more detail, or build on this approach to investigate the role of observational learning in aposematism and predator memory.