Meeting Abstract
S8-9 Monday, Jan. 6 13:30 - 14:00 Visual behavior in flying snakes: measurement and exploration with virtual reality ZAMORE, SA*; ARAUJO, N; SOCHA, JJ; University of Colorado Boulder; Virginia Tech; Virginia Tech sharri.zamore@colorado.edu http://www.curiousdrz.com
Flying snakes (genus Chrysopelea) are highly visual animals that climb, jump, and glide while navigating through tropical rainforests. These arboreal behaviors likely produce different visual problems than terrestrial locomotion. For example, gliding requires visual assessment to determine position, distance, and speed. To wit, flying snakes have large eyes, and perform visually-guided behaviors, such as tracking birds that are flying overhead. While some visual behaviors in arboreal snakes are well-described, their visual capabilities and its role in decision-making remains largely unknown. The visually sensitive flying snake is an ideal model to explore this relationship.
The use of closed-loop stimuli—stimuli that is modified based on behavioral output—has proven to be a useful tool for exploring visually-guided behaviors. To test the functionality of closed-loop systems on visually-guided limbless locomotion, we developed an Immersive Virtual Visual Arena (IVVA). The arena is a 3-foot cube with translucent walls onto which images were back-projected from LCD projectors. The projected visual stimuli were created with Unity 3D game development software. We tracked a marker on the head of the snakes using a Leap Motion Sensor, d its position using OpenCV, and updated the stimuli in Unity. The snakes were placed on an air table inside the cube, which inhibited translative locomotion, while allowing them to freely move.
Preliminary results show that behavioral responses in the virtual arena approximate those seen in an “analog” setup. This talk will 1) review previous visual research on the flying snake (Chrysopelea paradisi) that contributed to the specifications of the IVVA, and 2) discuss the design and development of an affordable IVVA, and its performance when exploring visually guided behaviors.