The present thesis examined whether the corvid, Clark’s nutcracker, is able to discriminate knowledge states between human experimenters based upon gestural cues using an object-choice task. To do so, the knowledge state of two experimenters was manipulated – one experimenter was informed, and the other uninformed, as to the location of a hidden food reward. To find the reward, the birds had to use the gesture of the informed experimenter and refrain from using the unreliable gesture of the uninformed experimenter. The nutcrackers responded to the gesture of the informed experimenter at above chance levels when simultaneously presented with the uninformed experimenter’s gesture. When the uninformed experimenter’s gesture was presented alone, the birds continued to follow the gesture. These results suggest the birds learned the gesture was meaningful, perhaps by associative learning, yet when this mechanism was not reliable the nutcrackers based their choices on the knowledge states of the experimenters.

The present thesis examined whether the corvid, Clark’s nutcracker, is able to discriminate knowledge states between human experimenters based upon gestural cues using an object-choice task. To do so, the knowledge state of two experimenters was manipulated – one experimenter was informed, and the other uninformed, as to the location of a hidden food reward. To find the reward, the birds had to use the gesture of the informed experimenter and refrain from using the unreliable gesture of the uninformed experimenter. The nutcrackers responded to the gesture of the informed experimenter at above chance levels when simultaneously presented with the uninformed experimenter’s gesture. When the uninformed experimenter’s gesture was presented alone, the birds continued to follow the gesture. These results suggest the birds learned the gesture was meaningful, perhaps by associative learning, yet when this mechanism was not reliable the nutcrackers based their choices on the knowledge states of the experimenters.

Clark's Nutcracker (Nucifraga columbiana) is the primary seed disperser of whitebark pine (Pinus albicaulis), which is in decline throughout its range. There is concern that a decline in whitebark pine will lead to a subsequent decline in local populations of Clark's Nutcracker. Because natural regeneration depends on the presence of Clark's Nutcracker, the process of harvesting whitebark pine seeds needs to be fully understood. In addition, resource managers need a cost-effective method for monitoring nutcracker occurrence in whitebark pine stands during the seed harvest season. I visited eleven study sites in Glacier National Park, Montana, where I searched for Clark's Nutcracker and surveyed whitebark pine cones for seed harvesting scars, the presence of which indicated that nutcrackers harvested seeds. I documented cone use patterns of Clark's Nutcracker and the major cone predator, red squirrel (Tamiasciurus hudsonicus), at five sites. To identify factors that influence cone use, I ran a correlation analysis with nutcracker and red squirrel seed harvesting variables with physical, compositional, and whitebark pine-related factors. I found that nutcrackers harvested seed at every site that had cones available. Nutcrackers harvested seed from a greater proportion of whitebark pine cones in stands where they started intensively harvesting seeds earlier. Nutcrackers began intensively harvesting seeds earlier in stands with higher relative dominance of whitebark pine. Red squirrels depleted the cone source more rapidly in stands with greater whitebark pine mortality, and at one site depleted the cone source completely before nutcrackers began intensively harvesting seeds from that site. The results of this study suggest that Clark's Nutcracker will continue to harvest seeds even as whitebark pine declines, but the decline in whitebark pine may lead to decreased seed dispersal due to greater pre-dispersal cone predation by red squirrels. Finally, I evaluated direct and indirect monitoring methods to identify a cost-effective method to accurately monitor Clark's Nutcracker occurrence in whitebark pine stands during the seed harvest season. I found that surveying scars made by seed-harvesting nutcrackers on whitebark pine cones was the most accurate and economical method of monitoring Clark's Nutcracker occurrence in an area with a low population of Clark's Nutcracker.

This dissertation explored the social cognitive abilities of the Clark’s nutcracker (Nucifraga columbiana), a relatively non-social, food-caching corvid. Corvids are a family of large-brained birds, which are capable of remarkable cognitive feats (e.g., future planning, tool use). These cognitive abilities have been revealed predominantly by testing social species, supporting popular theories that living in social groups drove the evolution of complex cognition. However, few studies have investigated the social cognitive abilities of corvid species that do not live in large groups. Here, I developed novel procedures using the food-caching behaviour of Clark’s nutcrackers as a tool to explore two cognitive abilities predicted to be limited to social species: mirror self-recognition (Chapter 2) and cooperation (Chapter 4). In Chapter 2, birds cached food when alone, with a conspecific present, and with a regular or blurry mirror. The nutcrackers suppressed caching with a regular mirror (as done with a conspecific), but not with the blurry mirror. When integrated with the traditional ‘mark test’, the birds also showed evidence of self-recognition with the blurry mirror by attempting to remove a coloured mark placed on their body with the blurry mirror, but not with an opaque barrier. In Chapter 3, I discuss the importance of self-recognition as a precursor for complex and flexible social cognitive abilities such as cooperation. To investigate cooperation, in Chapter 4 the birds experienced having their caches exchanged with another bird over multiple trials. This procedure assessed whether the normal response of cache suppression with a conspecific could be over-ridden if the experimental contingencies made cache sharing beneficial. The nutcrackers continued to cache in this context, and male birds increased caching when cooperation from the conspecific was exaggerated artificially by the experimenter. Combined, the results indicate the non- social Clark’s nutcracker is capable of mirror self-recognition, and the ability to distinguish one’s ‘self’ from others may facilitate flexible caching decisions, contrary to the predictions of the social living hypotheses. The findings indicate social living alone does not strongly predict complex cognitive abilities and, instead, that multiple evolutionary paths exist for the development of complex cognition. / October 2016