FicTrac: a visual method for tracking spherical motion and generating fictive animal paths

Area of Science:

• Neuroscience
• Animal Behavior
• Computer Vision

Background:

• Understanding brain mechanisms of attention, learning, memory, sensory processing, and navigation requires studying animal behavior in controlled environments.
• Virtual reality (VR) offers a powerful tool for such investigations, but requires accurate methods to track animal movement within the virtual space.

Purpose of the Study:

• To develop and validate a novel, accurate, and robust vision-based tracking system for estimating animal paths in VR.
• To enable real-time sensory feedback for behaving animals in VR experiments.
• To demonstrate the system's flexibility and applicability across different species and experimental paradigms.

Main Methods:

• Development of FicTrac (Fictive path Tracking software), a vision-based system using a standard camera and computer vision techniques.
• FicTrac estimates an animal's path by analyzing its rotation of an air-supported sphere.
• Comparison of FicTrac's accuracy and robustness against a standard optical mouse-based approach.

Main Results:

• FicTrac demonstrated superior accuracy and robustness compared to optical mouse-based methods for generating fictive paths.
• The system successfully recorded tethered honeybee (Apis mellifera) behavior in response to visual stimuli.
• Closed-loop visual fixation was demonstrated in both honeybees and fruit flies (Drosophila melanogaster), highlighting the system's flexibility.

Conclusions:

• FicTrac is a simple, fast, and adaptable vision-based tracking system for VR experiments with animals.
• It enables real-time sensory feedback and can be combined with electrophysiology for neural mechanism studies.
• The system is suitable for a wide range of organisms, including invertebrates and vertebrates.