Abstract:

The ability of a primate to efficiently find and relocate potential food items within a complex tropical forest likely influences its survivorship and reproductive success. Yet, we still know relatively little about the methods wild primates use to find desired resources across heterogeneous landscapes. Several factors have been proposed to impact spatial movement decisions in primates (including resource availability, forest structure, predation, and competition), but the relative importance of the spatial distributions of these factors in determining primate space-use patterns is unknown. In this dissertation, I address this topic by using multivariate methods to explore the relative influence of various ecological and social factors impacting the spatial foraging and movement decisions of mantled howler monkey ( Alouatta palliata ) groups on Barro Colorado Island, Panama. Chapter One evaluates the relative importance of canopy connectivity and resource distributions in the placement of arboreal pathway networks. Results indicate that canopy connectivity is a highly significant predictor of path selection, and that the relative importance of canopy connectivity and resource distributions likely varies depending upon context. Chapter Two investigates the types of spatial information howler monkey groups use when selecting foraging sites, and evaluates evidence for rule-based foraging decisions (i.e. optimization, 'win-stay' foraging, and the prioritization of a few key species). Support was found for the theory that howlers maximize exposure to fruits and flowers, while minimizing distance traveled. However, analyses indicated that foraging site selection was influenced not only by the characteristics of the resources at a given site (e.g. a particular target tree), but also by the resources in the immediately surrounding areas. Chapter Three examines the role of vocalizations in mediating inter-group spatial interactions. Results indicate that interaction history is a greater predictor than numeric odds in determining a group's spatial response to howls by known neighbors. In addition, groups approached howls more often when resources were low, when the howling group was closer to the responding group's home range center, and when the rate of howls (howls per minute) was low.