Compact Internal Representation - A Mathematical Model of Drosophila‘s Prediction Capabilities in Flight Control

Abstract

Easily one can imagine, how difficult it must be for a flying fly to navigate in an environment rich of moving objects. How to avoid collisions and make a soft landing, when for instance seesawing plants provide ample distraction? Flies have to rely on their sensory systems to survive, but the information they perceive is rich and complex. Remarkably, flies show prediction capabilities, when their environment changes in a regular manner with time. The aim of this project is to investigate, how flies perform the processing of systematically time-dependent sensory information, and to find brain areas involved in the high-level processing of such information. We try to understand this information processing by starting from the mathematical concept of Compact Internal Representation (CIR).