Masters Project

Navigation memory in bumblebees


  • Lina O'Reilly

One of the defining features of animals is the ability to move through their environment. These movements are controlled by the animals’ brains, which use sensory information, combine it with previous experience and generate behavioral steering commands. These steering commands are the result of a comparison of the animal’s current heading with its desired heading. Whenever a mismatch between these two directions is detected, corrective steering pushes the animal back on track. These elementary navigational decisions have to be carried out in each moment in time and occur independent of the overall behavioral strategy. In our group we aim at understanding these decisions on the level of neural circuits consisting of identified neurons. We have therefore developed the bee central complex, a higher order brain center shared across all insect brains, as a model to investigate these neural processes.

Bees use a computation called path integration to return in a straight-line path (home vector) back to their nest after a convoluted foraging flight. During this homing behavior, the bee brain obtains the current heading estimate from sky compass cues, and the desired heading is equivalent to the angular component of the home vector. To obtain the home vector, bees have to monitor the directions and distances flown during the outbound trip. While it is known that honeybees use optic flow for measuring distances and polarized skylight for measuring directions, the neural mechanisms are unclear in nay species. We have recently established the bumblebee as an accessible model system for investigating these questions with electrophysiology, behavior, molecular methods, and connectomics.  Additionally, we have discovered neurons in the central complex of bees that encode optic flow based movement speed as well as another set of neurons encoding compass directions. Using a computational model, we have identified likely candidate neurons that are suited to integrate both pieces of information into a representation of the home vector.

The current project builds on these recent developments and is aimed at facilitating a deeper understanding of the path integration memory in bumblebees and to pave the way towards locating this memory to specific neurons of the brain. We will pursue a dual strategy of two parallel sub-projects. Project one will comprise behavioral experiments in flight tunnels aimed at verifying that bumblebees indeed use optic flow to measure distances, to illuminate how optic flow based path integration memory is used in conjunction with landmark based memory (hierarchy of navigation strategies), and finally to start characterizing the physiological properties of each of these memories by attempting to disrupt them experimentally. The second project will establish a novel paradigm to locate brain activity resulting from path integration behavior. We will use fluorescence labeled sugar that can be taken up by the brain, but cannot be metabolized to localize neurons with high sugar demand, i.e. high metabolic activity. As metabolic activity is a proxy for neuronal firing rates, active neurons should be highlighted by this method. As this method has not been used in bees before, we will carry out proof of concept studies aimed at establishing the activity patterns resulting from visual stimulation and motor activity of bumblebees. The final goal is to use this method in conjunction with the behavioral experiments of project one and locate neurons that correlate with the acquired path integration memory.