SANS annual report 1994

Previous Up Next


Generation and Control of Swimming

Generation and Control of Swimming

To be able to do realistic simulations of a particular neural network, detailed knowledge is needed both to set the connectivity and the cellular properties correctly. This is generally done either by limiting the study to simple invertebrate animals or to do extrapolations from simpler systems. The lamprey provides an intermediate alternative: being a lower vertebrate (a simple fish) it shares much of the structure with higher animals; yet, its nervous system is fairly accessible for experiments and has been studied intensely over the years.

At the Nobel Institute for Neurophysiology at Karolinska Institutet , the neural circuitry responsible for the generation of the rhythm involved in lamprey swimming has been studied for a long time. The synaptic connectivity as well as the cellular properties have been well characterized. The system, as it is known today, is too complex to be grasped and understood from the knowledge available for the individual components. However, being comparatively well characterized, it is a good candidate for computer simulation.

Together with the group at Karolinska Institutet, we have made a computer model of the segmental rhythm generator capable of mimicking both fast and slow swimming as well as many other experimental conditions (Wallén et al 1992), (Hellgren Kotaleski et al 1992).

Current activities include an extension towards simulating the spinal cord in its full extent using a more realistic number of neurons; a hypothesis for the intersegmental coordination and generation of steering commands; and modeling the entire swimming behavior also including body mechanics and body-water interactions.

Subtopics

  • The lamprey continuous spinal network model
  • The crossed-oscillators model for 3D steering
  • Connecting the crossed-oscillators model to a mechanical environment

  • Next