Environmental drivers and consequences of lekking by white bearded manakins

This project advances our ability to address a growing need in the field of behavioral ecology: how high-dimensional, correlated social and ecological variables interact to shape animal behaviors in the natural world. To meet this need, our team of AI experts and seasoned behavioral ecologists will pioneer the development of two complementary computational tools – AI interpretation of videos and neural network causal inference modeling – of broad relevance to field-based studies of behavioral ecology. In doing so, we offer two novel approaches to make sense of the many social and ecological factors that could influence mating behavior in birds, including where and when to display?; how do males optimize display phenotype and mating success?; and, how do display behaviors of frugivorous birds impact plant community succession in tropical forests? In addition to addressing these long-standing research questions, this project will result in shared research infrastructure – hundreds of thousands of hours of enriched videos and accompanying open-source software – to enhance integration of AI into field-based behavioral ecology studies. To amplify the societal impact and reach of this work, we will produce a broadcast-quality short film to engage public audiences with the natural history and ecological significance of mating systems and display behaviors in tropical birds manakins. We will also offer training opportunities to undergraduate and graduate students, a post-doctoral fellow, as well as high school teachers. This work advances the NSF priority area of use of Artificial Intelligence in research and in workforce training and development. This project investigates the eco-evolutionary dynamics of lekking behavior in the white-bearded manakin (Manacus manacus), a tropical frugivore found in Ecuador’s Chocó rainforest. Lekking is a mating system where males gather to perform elaborate displays without providing parental care, resulting in strong reproductive skew and some of the most extreme examples of sexually selected traits observed in nature. The proposed study concurrently evaluates how ecological and social environments interact to influence lek formation and dynamics, as well as how lekking birds may shape the environments they inhabit. First, we document, in real time, ecological and social conditions before, during, and after lek establishment in a replicated and rigorous manner. Second, we gather a rich, multifaceted dataset to evaluate the relative impact and chains of causality among multiple, inter-correlated variables on mate choice dynamics at active leks. Third, we model seed dispersal and demographic trajectories of favored plant species to better understand the degree to which frugivorous birds may act as ‘ecosystem engineers’ by shaping plant communities. Our multidisciplinary team combines deep knowledge of the natural system with expertise in contemporary computational methods, including AI-based video interpretation and neural network modeling approaches. Building from these strengths, we expect to deliver a comprehensive understanding of how ecological and social drivers interact to shape lek establishment and mate choice dynamics at leks, as well as the ways that lekking behavior in turn shapes the surrounding environment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria. NSF Award ID: 2550002 | Program: 01002627DB NSF RESEARCH & RELATED ACTIVIT | Principal Investigator: Jordan Karubian | Institution: Tulane University, NEW ORLEANS, LA | Award Amount: $1,386,093 View on NSF Award Search: https://www.nsf.gov/awardsearch/show-award/?AWD_ID=2550002 View on Research.gov: https://www.research.gov/awardapi-service/v1/awards/2550002.html