Biosonar processing of rare objects in complex scenes

• Applicant: Privatdozent Dr. Uwe Firzlaff, since 2/2020
• Subject Area: Cognitive, Systems and Behavioural Neurobiology
• Term: from 2018 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 411419794

Vision provides animals with a high-resolution image of their environment, ideally suited for perceiving, sampling and evaluating a multitude of objects distributed in space. However, where light was amiss, echolocation evolved in animals, allowing the imaging of space through the auditory analysis of self-generated sounds.Vision and echolocation are both highly directional. In vision, we constantly move our eyes to detect new and relevant objects. As these objects are often rare, a process of selection is required to pick out relevant objects, based on a mixture of the visual salience of objects and our behavioral demands. In echolocation, the need to focus on specific objects of interest is even more imperative due to the non-spatial layout of the sensory epithelium, the cochlea, and the comparatively broad sonar beam which ensonifies multiple close-by objects simultaneously. These spatial properties of echolocation still allow a high acuity to pinpoint the direction and distance of a single target, but possess a very low angular resolution, preventing bats to resolve multiple targets that are present in complex natural scenes. Unlike in vision, where the processes that guide sequential sampling of space are well studied, very little is known about how echolocating bats scan complex natural scenes by echolocation and which reflective properties of objects guide the bat’s sonar beam to select an important but rare object among a background of unimportant information. Here we propose an integrative experimental approach with a common experimental paradigm comprising psychophysical experiments for bats in virtual echo-acoustic environments and single-cell neurophysiology to understand the behavioral strategy for the detection of rare objects and its neural basis. (1) We will work with trained bats in virtual echo-acoustic environments, allowing us to quantify echo-acoustic salience for isolated object properties and combinations thereof. (2) We will record neural activity, from both anaesthetized and awake bats to investigate the processing of rare objects in a well-established task which we refine for echolocation. Our common experimental paradigm will allow us to link animal behaviour, such as the dynamics of biosonar emission and ear and head movements, to object salience features and the underlying neuronal processing mechanisms. Our work will provide new insights into which strategies bats employ in echolocation to maneuver extremely fast in complete darkness through highly structured environments in 3D, avoiding obstacles while successfully detecting and capturing prey.

• DFG Programme: Research Grants
• Ehemaliger Antragsteller: Professor Dr. Lutz Wiegrebe, until 2/2020 (†)