Computations and mechanisms underlying sky- compass dependent head direction coding in the desert locust

• Applicants: Professor Dominik M. Endres, Ph.D.; Professor Dr. Uwe Homberg
• Subject Area: Cognitive, Systems and Behavioural Neurobiology ; Experimental and Theoretical Network Neuroscience
• Term: since 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 452193090

As highly mobile animals, insects and vertebrates face similar challenges during oriented locomotion. While in vertebrates, neurons encoding navigational aspects, such as head-direction cells, place cells and grid cells have been characterized for some time, brain areas and neural signatures of space-encoding neurons in insects have only recently been explored. One insect brain area in particular, the central complex, appears to play a key role in spatial orientation. In several species, neurons of the central complex are sensitive to sky compass signals such as the directional position of the sun (solar azimuth) and the angle of polarization of skylight, which is generated by the sun. In contrast to local landmark information providing directional information in mammals, these globally available cues appear to be used for head-direction coding in many other insect species and may be particularly suited for long-range seasonal migrants like locusts. In desert locusts, the solar azimuth and the overall pattern of polarization across the sky are encoded in the central complex in a compass-like fashion. Recent data show that both signals support each other, whereas the analysis of polarization in just a small spot of sky in the zenith may provide false information on solar azimuth. We will investigate under which conditions (environment and behavioral goal) such a representation is optimal, akin to ‘ideal observer’ models in human psychophysics. Although visual pathways from the eyes to the central complex have been studied in great detail, the computational mechanisms leading to the sky compass in the central complex are still unknown. The current proposal aims at modelling sun and sky polarization input to the compound eye, convergence of pathways from both eyes to the central complex and interaction of sun and sky polarization signals. Special attention will be paid to make the compass robust against different clouding conditions and solar elevations. We will construct an ‘ideal observer’ of the sky/sun, and translate this computational model into a neuronal implementation consistent with known anatomical and physiological constraints. The predictions of this neuronal model will be compared to data from electrophysiological studies, characterizing the integration of sun and polarization signals in the locust brain and identifying neurons shifting the peak of activity in the central complex compass during rotations of the animal. We expect that this study will uncover mechanisms in the locust brain leading to compass representation of celestial cues that can then be compared with head-direction coding in insects such as Drosophila that may navigate without sky cues. The results of this study will stimulate research on neural mechanisms in seasonally migrating vertebrates relying on a sun- and sky compass.

• DFG Programme: Research Grants