Graded variation in D notes of Carolina chickadees: Do signalers vary acoustic fine structure with context and do receivers respond to that variation?

Animal vocal signals vary in terms of repetition rate of signal units and graded acoustic properties such as frequency and duration. Variation in the signals is known to be associated with contexts directly related to fitness, such as predation threat and food detection, and with other external factors such as group membership, geographic location of the signaler, call structure, and so on. Among animal species that produce combinatorial signals such as some primates, meerkats, and birds in the Paridae, graded acoustic properties have gotten little attention. In this dissertation, I tested whether Carolina chickadees changed their vocal signal properties in predation and food contexts and whether receivers in turn changed their behavior according to that vocal variation (i.e. tested if communication occurs in food and predation contexts). Then I also tested whether the signal varied across flocks, geographic locations, and call structures. I found the very first evidence that chickadees varied their vocalizations across food and predation contexts by changing both repetition rate and graded acoustic properties including spectral and temporal measures. However, there was no evidence that receivers behaved differently according to the signal variation. In the last study, it turned out that flock membership and call structure were accurately classified with the graded acoustic properties of notes. This dissertation demonstrates the first evidence that acoustic properties of the same specific signal unit can be adjusted in two opposing fitness-related contexts, and also evidence of consistent vocal convergence regardless of food and predation contexts within flocks. Taken together, this study advances our knowledge of acoustic signal variation with various ecological factors and of receiver perception of signal variation.