Ekeberg Orjan, Blümel Marcus, Büschges Ansgar
Department of Numerical Analysis and Computer Science, Royal Institute of Technology, 10044 Stockholm, Sweden.
Arthropod Struct Dev. 2004 Jul;33(3):287-300. doi: 10.1016/j.asd.2004.05.002.
Insect walking relies on a complex interaction between the environment, body segments, muscles and the nervous system. For the stick insect in particular, previous investigations have highlighted the role of specific sensory signals in the timing of activity of central neural networks driving the individual leg joints. The objective of the current study was to relate specific sensory and neuronal mechanisms, known from experiments on reduced preparations, to the generation of the natural sequence of events forming the step cycle in a single leg. We have done this by simulating a dynamic 3D-biomechanical model of the stick insect coupled to a reduced model of the neural control system, incorporating only the mechanisms under study. The neural system sends muscle activation levels to the biomechanical system, which in turn provides correctly timed propriosensory signals back to the neural model. The first simulations were designed to test if the currently known mechanisms would be sufficient to explain the coordinated activation of the different leg muscles in the middle leg. Two experimental situations were mimicked: restricted stepping where only the coxa-trochanteral joint and the femur-tibia joint were free to move, and the unrestricted single leg movements on a friction-free surface. The first of these experimental situations is in fact similar to the preparation used in gathering much of the detailed knowledge on sensory and neuronal mechanisms. The simulations show that the mechanisms included can indeed account for the entire step cycle in both situations. The second aim was to test to what extent the same sensory and neuronal mechanisms would be adequate also for controlling the front and hind legs, despite the large differences in both leg morphology and kinematic patterns. The simulations show that front leg stepping can be generated by basically the same mechanisms while the hind leg control requires some reorganization. The simulations suggest that the influence from the femoral chordotonal organs on the network controlling levation-depression may have a reversed effect in the hind legs as compared to the middle and front legs. This, and other predictions from the model will have to be confirmed by additional experiments.
昆虫行走依赖于环境、身体节段、肌肉和神经系统之间的复杂相互作用。特别是对于竹节虫而言，先前的研究突出了特定感觉信号在驱动单个腿部关节的中枢神经网络活动时间安排中的作用。本研究的目的是将从简化标本实验中得知的特定感觉和神经元机制，与单个腿部形成步周期的自然事件序列的产生联系起来。我们通过模拟与神经控制系统简化模型耦合的竹节虫动态三维生物力学模型来实现这一目标，该模型仅纳入了所研究的机制。神经系统将肌肉激活水平发送到生物力学系统，生物力学系统进而将正确定时的本体感觉信号反馈回神经模型。最初的模拟旨在测试当前已知的机制是否足以解释中腿不同腿部肌肉的协同激活。模拟了两种实验情况：受限步态，即只有基节 - 转节关节和股节 - 胫节关节可自由移动，以及在无摩擦表面上的无限制单腿运动。实际上，这些实验情况中的第一种类似于用于获取许多关于感觉和神经元机制详细知识的标本制备。模拟结果表明，所包含的机制确实可以解释两种情况下的整个步周期。第二个目标是测试相同的感觉和神经元机制在多大程度上也足以控制前腿和后腿，尽管腿部形态和运动模式存在很大差异。模拟结果表明，前腿的步态基本上可以由相同的机制产生，而后腿控制则需要一些重新组织。模拟结果表明，与中腿和前腿相比，股弦音器对控制抬腿 - 压腿网络的影响在后腿中可能具有相反的效果。该模型的这一以及其他预测将必须通过额外的实验来证实。
