Date:
Location:
Title: Drosophila behavior under natural and semi-natural conditions
Abstract: Behavior is a highly context dependent trait that evolves in an organism in its natural habitat. When we bring animals into the lab and control environmental conditions to perform experiments, we are placing them into artificial contexts, which may affect the behavioral outputs that we observe. At the least, exploring the behavior of animals under natural or semi-natural conditions complements our understanding of laboratory findings and considers important evolutionary concerns that may not be at the forefront of traditional behavioral experiments, especially in Drosophila, a popular animal model for genetics and neuroscience.
In Chapter 1, I bring the wild substrate of Drosophila melanogaster, rotting fruit, into the lab by creating plastic surfaces that capture the geometry of the natural surfaces. I find that flies walking on naturalistic surfaces are attracted to certain geometric features, such as high and steep places, i.e., peaks and edges, and this preference is modulated by internal state, species, and essential sensory organs. In Chapter 2, I bring the lab into the wild by developing a field deployable device, ACORN, or Automated Capture and Observe Robot in Nature, that can measure turning behavior in a y-maze, a common behavior assayed in the lab. We find that wild flies assayed in the wild display behavioral individuality, agreeing with results from laboratory flies, and we also detect some differences in higher order behavioral measures over a field season, possibly reflecting changes in evolutionary strategy through time. Finally, in Chapter 3, I bring wild flies into the lab to investigate the evolutionary strategies of wild populations of Drosophila throughout the U.S. We find evidence of bet-hedging in the thermal preferences of flies from localities of fluctuating environmental conditions; for example, Boston undergoes major seasonal changes from winter to summer, and we see evidence of bet-hedging in local wild flies. This work illustrates how considering the wild context of an animal can enhance our understanding of its behavior and improve our interpretation of laboratory experiments.
Committee: Benjamin de Bivort (Advisor), Hopi Hoekstra, Venkatesh Murthy (MCB), Yun Zhang