Periodic Reporting for period 1 - PI_Memory_Circuits (The neural basis of path integration memory in insects)
Período documentado: 2021-09-01 hasta 2023-08-31
Imagine yourself hiking in the trailless wilderness of northern Scandinavia. You leave your tent for an excursion to admire the beautiful scenery. After passing the first moss-covered hill you already lose sight of the tent. Most humans would be utterly lost within an hour of undirected travel, unable to find the way back. Yet, many tiny invertebrate animals, such as bees and ants, with even smaller brains are able to return back to their nests with ease, even after navigating unknown, featureless terrain. They use a navigational neural computation called path integration to do so, which other animals, including mammals, rely on as well. Path integration is an efficient strategy to find a previously visited location: An animal monitors the turns it makes and distances it travels along its trip to continuously update an estimate of its location relative to the start of its journey. This estimate can be used to return to home along a straight line, termed the home vector. Besides returning home, path integration can be used to locate any previously visited location when defined as the origin, a strategy generally called vector navigation. In addition to following a home vector, some arthropods are able to construct novel shortcuts to previously known locations and use landmarks as navigational aids. Observing these impressive behaviors raises the question of how they are neurally manifested in the relatively simple brains of arthropods, some smaller than a pinhead. Much progress has been gained in unravelling the behavioral strategies of vector navigation in animals, the neural circuits responsible for compass and speed encoding required for insect navigation, and in the development of models for how neural circuits might manifest into navigation behaviors; however, the neural basis of vector memories underlying path integration remains unknown in any animal. My project aimed to fill this gap in knowledge, with the ultimate goal of elucidating the neural basis of path integration memory, a question that has been asked for decades. Combining behavioral, electrophysiological and neuroanatomical methods, I aimed at uncovering the core circuits underlying vector navigation in arthropods as an overarching goal. Over the course of the proposed project, I pursued two specific goals, first to develop a behavioral assay to define the fundamental characteristics of path integration in bumblebees and second, to develop methods to determine the neural correlate of vector memories in bumblebees.
By constructing circular arenas with artificial celestial cues connected to a living hive of bumblebees, I showed that walking bumblebees perform vector navigation behaviors while homing over distances less than the meter in the laboratory, consolidating behaviors naturally occurring over kilometers in nature. We found that during path integration, both directions and distances of vector memories are maintained by walking bumblebees. These bees would enact stereotyped search patterns when their path integrator did not lead them home. We demonstrated that during path integration, bumblebees orient using the two artificial celestial cues present in the arenas, an overhead polarization pattern and an artificial sun. We also have found that vector memories are maintained in the brain for at least 12 hours, are robust to chill-coma anesthesia, and are maintained after surgeries required for electrophysiological investigations, where the brain is exposed. These results have been published as a full-length article in the high-impact peer-reviewed journal Current Biology (doi.org/10.1016/j.cub.2022.05.010). Following these experiments, we have also found that walking bumblebees store path integration vectors in long-term memory and can recall them at a familiar location, something that has yet to be unequivocally demonstrated in any arthropod. These data offer a potential mechanism for a vector-based analog of a cognitive map in insects and are currently being prepared for publication.
To investigate the neural basis of path integration physiologically, we developed a methodology for investigating vector navigation behavior in virtual reality. In this virtual reality arena, bumblebees walk on a spherical air-cushioned treadmill surrounded by curved LED panels and an overhead polarized light field that mimics the orientation cues present in the real-world arenas used in the behavioral work described above. In preliminary work, homing bees walked on a spherical treadmill the approximate distance from the feeder to the nest in the direction of the nest predicted from a home vector informed by path integration. I have also established extracellular electrophysiological methods using a tetrode system to record from the bumblebee central brain in the virtual reality arena. Successful recordings from the central brains of completely tethered bumblebees have been achieved, suggesting that extracellular recordings in moving bumblebees is a realistic future goal.
The aforementioned results have been presented at international conferences and research institutions throughout Europe (6 times) and the USA (3 times).
To investigate the neural basis of path integration physiologically, we developed a methodology for investigating vector navigation behavior in virtual reality. In this virtual reality arena, bumblebees walk on a spherical air-cushioned treadmill surrounded by curved LED panels and an overhead polarized light field that mimics the orientation cues present in the real-world arenas used in the behavioral work described above. In preliminary work, homing bees walked on a spherical treadmill the approximate distance from the feeder to the nest in the direction of the nest predicted from a home vector informed by path integration. I have also established extracellular electrophysiological methods using a tetrode system to record from the bumblebee central brain in the virtual reality arena. Successful recordings from the central brains of completely tethered bumblebees have been achieved, suggesting that extracellular recordings in moving bumblebees is a realistic future goal.
The aforementioned results have been presented at international conferences and research institutions throughout Europe (6 times) and the USA (3 times).
Though much progress has been gained in recent years investigating the neural basis of animal navigation, the concrete neural basis of navigational vector memory is still unknown in any animal. The project has resulted in a robust behavioral assay to investigate vector navigation in the laboratory and neurobiological methodologies to examine the neural basis of these behaviors. Further, direct behavioral evidence of the storage and recall of navigational memories from long-term memory has been demonstrated, offering evidence for a mechanistic hypothesis for how a vector-based cognitive map might exist in insects. From these findings, more complex navigational feats of insects, including generation of food vectors and novel shortcuts and optimal foraging routes at multiple places is behaviorally accessible in the laboratory and have the potential to become neurally accessible with my methods. All these findings are extremely valuable for the fields of animal navigation and neuroscience in general, with the goal of understanding how complex behaviors are manifested in small brains with limited numbers of neurons.
Results from fellowship currently help inform models aimed at translating how actual neural navigational circuits in insects can be manifested in robotics. My host lab (the Heinze Lab in the Vision Group at Lund University) has established a collaboration with NanoLund, a center for physics/engineering, aiming at implementing insect path integration circuits with nanophotonics, developing nanoscale agents with autonomous navigation capabilities. These technologies have applications in any autonomously navigating agent, such as search-and-rescue drones in disaster areas or autonomous cars.
Results from fellowship currently help inform models aimed at translating how actual neural navigational circuits in insects can be manifested in robotics. My host lab (the Heinze Lab in the Vision Group at Lund University) has established a collaboration with NanoLund, a center for physics/engineering, aiming at implementing insect path integration circuits with nanophotonics, developing nanoscale agents with autonomous navigation capabilities. These technologies have applications in any autonomously navigating agent, such as search-and-rescue drones in disaster areas or autonomous cars.