Memory in wild mountain chickadees from different elevations: comparing first-year birds with older survivors

Understanding both inter- and intraspecific variation in animals' cognitive abilities is one of the central goals of cognitive ecology. We developed a field system for testing spatial learning in wild chickadees using radio frequency identification (RFID)-enabled feeders that allowed us to track individuals across multiple years. Mountain chickadees, Poecile gambeli, inhabit a continuous montane gradient, and individuals inhabiting higher elevations experience harsher winters than those at lower elevations. Previous studies found that chickadees at higher elevations cached more food and demonstrated better spatial memory, but they performed worse during reversal learning than chickadees at lower elevations. Here, we employed spatial learning, reversal learning and memory retention tasks to compare elevation-related performance of first-year juvenile birds with that of adults that had survived at least 1 year. Chickadees from high elevation performed better in the initial learning task but worse in the reversal task than birds from low elevation. There were no differences between first-year birds and adults in the initial learning task, but adults performed significantly better in the reversal test. First-year birds also made more errors associated with the initial target, which suggests higher levels of proactive interference. There were no significant differences between elevations or between juvenile and adults in memory performance after a 16-day retention. After retention, chickadees did not discriminate between the feeders that provided food during the initial task or during the reversal task prior to retention. These results are also consistent with the effects of proactive interference, as birds should have only visited the most recently rewarding feeder. Our findings suggest that the ability to quickly learn changing information is critical for chickadees at both elevations as surviving adults did better in the reversal task than first-year birds. Our results also suggest that selection may favour better reversal learning abilities associated with lower levels of proactive interference.