Lund Vision Group, Department of Biology, Lund University, Lund, Sweden.
Neural Circuits and Evolution Lab, The Francis Crick Institute, London, UK.
J Comp Physiol A Neuroethol Sens Neural Behav Physiol. 2023 Jul;209(4):563-591. doi: 10.1007/s00359-023-01625-x. Epub 2023 Apr 5.
Spatial orientation is a prerequisite for most behaviors. In insects, the underlying neural computations take place in the central complex (CX), the brain's navigational center. In this region different streams of sensory information converge to enable context-dependent navigational decisions. Accordingly, a variety of CX input neurons deliver information about different navigation-relevant cues. In bees, direction encoding polarized light signals converge with translational optic flow signals that are suited to encode the flight speed of the animals. The continuous integration of speed and directions in the CX can be used to generate a vector memory of the bee's current position in space in relation to its nest, i.e., perform path integration. This process depends on specific, complex features of the optic flow encoding CX input neurons, but it is unknown how this information is derived from the visual periphery. Here, we thus aimed at gaining insight into how simple motion signals are reshaped upstream of the speed encoding CX input neurons to generate their complex features. Using electrophysiology and anatomical analyses of the halictic bees Megalopta genalis and Megalopta centralis, we identified a wide range of motion-sensitive neurons connecting the optic lobes with the central brain. While most neurons formed pathways with characteristics incompatible with CX speed neurons, we showed that one group of lobula projection neurons possess some physiological and anatomical features required to generate the visual responses of CX optic-flow encoding neurons. However, as these neurons cannot explain all features of CX speed cells, local interneurons of the central brain or alternative input cells from the optic lobe are additionally required to construct inputs with sufficient complexity to deliver speed signals suited for path integration in bees.
空间定位是大多数行为的前提。在昆虫中，潜在的神经计算发生在中央复合体（CX），即大脑的导航中心。在这个区域，不同的感觉信息流汇聚在一起，从而能够做出依赖情境的导航决策。因此，各种各样的 CX 输入神经元提供了关于不同导航相关线索的信息。在蜜蜂中，偏振光信号的方向编码与适合编码动物飞行速度的平移视流信号汇聚在一起。CX 中速度和方向的连续整合可用于生成有关蜜蜂当前在相对于其巢的空间中的位置的向量记忆，即执行路径整合。这个过程依赖于 CX 输入神经元对视流编码的特定的、复杂的特征，但尚不清楚这些信息是如何从视觉外围衍生出来的。在这里，我们旨在深入了解简单的运动信号如何在上游的速度编码 CX 输入神经元之前进行重塑，以产生它们的复杂特征。使用电生理学和 Halictid 蜜蜂 Megalopta genalis 和 Megalopta centralis 的解剖分析，我们确定了一系列与中央脑连接的光叶的运动敏感神经元。虽然大多数神经元形成了与 CX 速度神经元不兼容的特征的途径，但我们表明，一组 lobula 投射神经元具有生成 CX 视流编码神经元视觉反应所需的一些生理和解剖特征。然而，由于这些神经元不能解释 CX 速度细胞的所有特征，因此还需要中央脑的局部中间神经元或来自光叶的替代输入细胞，以构建具有足够复杂性的输入，从而提供适合蜜蜂路径整合的速度信号。
