Fly, Robot: The use of a controllable fly robot to explore object recognition and visual tracking during courtship in Drosophila

Thesis (Ph.D.)--University of Washington, 2015-11Though the application of genetic approaches has greatly increased our understanding of motion vision in the fruit fly, Drosophila melanogaster, little is known about higher-order functions of the visual system such as object recognition. Visual object recognition likely plays an important role during courtship: for successful courtship, males must be able to distinguish an object from the background and further categorize it as conspecific, female, and receptive. Because courtship between two flies happens at a close distance, typically within a few millimeters, the fly’s eye could possibly distinguish fine-scale patterns for use in object recognition. Even at a greater distance, the male would have access to large-scale cues like shape, size, and color, as well as motion cues like speed and patterns of motion, to aid in its visual identification of objects. In the following sets of experiments, I used a modifiable, programmable fly dummy coupled with automated behavioral tracking to explore how males use visual cues to dynamically modulate their courtship and the neurons that may be involved in these behaviors. I. The relative roles of vision and chemosensation in mate recognition: Animals rely on sensory cues to appropriately classify and respond to objects in their environment. However, the spatial structure of those sensory cues can greatly impact when and how they are perceived. In this study, I examined the relative roles of visual and chemosensory cues in mate recognition by pairing male flies with dummies of various shapes, sizes, and speeds, or coated with different pheromones. I determined that visual and chemical cues play specific roles at different points in the courtship sequence. Vision is essential for determining whether to approach a moving object and initiate courtship, and males were more likely to begin chasing objects that were the same approximate dimensions as another fly. However, whereas males were less likely to begin chasing larger dummies, once started, they would continue chasing for a similar length of time regardless of the dummy’s shape. The presence of female pheromones on the moving dummy did not affect whether males would initiate a chase, but it did influence how long they would continue chasing. Collectively, these results demonstrate that males can visually distinguish potential mates and that different sensory cues play a dominant role at different stages of courtship. II. Pigmentation and shape affect male’s positioning during courtship chases: Beyond helping a male identify a mate from afar, vision continues to be an important cue after courtship has begun. Vision allows a male to track and pursue a female, and it also influences how the male positions himself around the female during these chases. I found that as dummies increase in width or height, males demonstrate less wing extension and position themselves further from the dummy. I further found a role for dummy color in modulating mate attractiveness: males are not strongly attracted towards very dark or very light objects, but do show a robust courtship response towards gray-painted dummies. These experiments are the first to suggest that males use color to distinguish potential mates. I also found that male flies preferentially bias their chasing and courtship maneuvers towards the abdomen–and not the head–of female flies, irrespective of her direction of movement. This preference is maintained to some degree towards females that do not produce pheromones, suggesting that males can use vision to distinguish different parts of the female body. Color could be one of the visual cues males use to position themselves around the female, as males paired with dummies painted multiple colors bias their position towards the preferred color. III. Despite concurrent activation of P1 neurons, males still use visual cues to dynamically modulate courtship: The P1 neurons are a population of ~20 Drosophila male-specific, fruM-expressing neurons in the posterior brain that have recently been identified as the putative site where sensory cues are integrated and translated into persistent courtship behavior. Males with activated P1 neurons direct their courtship maneuvers towards nearby moving objects, and functional imaging studies demonstrate activation of these cells by both chemosensory and visual stimuli. I activated P1 neurons in male Drosophila using the heat-sensitive cation channel, TRPA1, and examined the effect on their courtship responses towards dummies of varying shapes or chemosensory content. While overall courtship levels were elevated, males still showed a preference for smaller, more fly-sized objects. Furthermore, males still used dummy shape or color to modulate their levels of wing extension and chasing position relative to the dummy. However, males with activated P1 neurons no longer modulated their chasing position in response to the dummy’s pheromone coating. Thus, visual information is able to modulate male courtship independent of P1 activation, but P1 neurons are important for modulating a males’ response to pheromones