17.1 Friday, Jan. 4 Hawkmoths of Endor: Navigational decision policies for obstacle navigation in Manduca sexta MUNK, Y*; WILKINSON, S; DANIEL, TL; Univ. of Washington; Morehouse College; Univ. of Washington firstname.lastname@example.org
Insects flying within complex environments avoid obstacles by making decisions primarily on the basis of visual information. To study how the availability of visual information about obstacles affects course planning, we constructed a virtual flight simulator in which tethered Manduca sexta hawkmoths were able to control their trajectories through a virtual 3-D forest environment by applying yaw torques measured by a torquemeter. We investigated how visual information regarding local obstacles influenced navigational decision policies by using the display software to reduce contrast between obstacles and the background on the basis of their virtual distance from the moth, effectively introducing a virtual fog into the environment. For each subject, we applied five levels of virtual fog at two flight speeds, with three replications for each combination, for a total of thirty trials per subject. For 8 subjects a variety of behaviors were recorded, including obstacle avoidance and attempted landings on virtual trees. Heavy virtual fog limited navigation for all subjects, leading to reactionary avoidance behavior without evidence of course planning. At light fog levels with longer range visibility, trajectories showed evidence of short range planning. Trajectories gathered for very long visibility ranges did not exhibit obvious differences compared to medium visibility trials. Our virtual reality system demonstrates a viable platform for investigating decision making on the basis of visual information, broadly applicable to a wide range of biological subjects and capable of easily generating extraordinarily rich datasets.