Movies

Larval zebrafish exhibit a sophisticated locomotive repertoire which depends on neural control signals that descend from brainstem to spinal cord (Budick and O'Malley, 2000; Borla et al., 2002). Because of the complexity of the descending motor control system in brainstem (O'Malley et al., 2003) and the complexity of the spinal networks that receive and respond to descending signals (Hale et al., 2001), it is difficult to learn the nature of the controls that shape larval locomotive behaviors (which include: routine turns, escape behaviors, prey-tracking and prey capture). We have therefore turned to modeling of both the kinematics of the animal's behaviors and the underlying neural networks. Our goal is to understand how the complex locomotive maneuvers exhibited by these larvae are generated.

The functioning of spinal networks that underlie locomotion in fishes and tadpoles has been extensively modeled (see e.g. Roberts and Tunstall, 1990; Dale, 1995; Grillner, 2003; Dale, 2003). While swimming in lamprey has been well studied, swimming in Xenopus tadpoles may better match larval zebrafish swimming, in terms of body form and locomotive repertoire. We therefore created a model based on Xenopus spinal network models that incorporate known properties of the oscillators underlying swimming (Tunstall et al. 2002). We first explored the control of tail-beat frequency (TBF), and found that by altering the synaptic strengths of AMPA, NMDA and Glycinergic-like synapses (all known to be present in Xenopus spinal cord), we were able to generate TBFs that spanned the range of speeds observed during burst and slow swimming behaviors. We then extended this to a multi-segment model. While regular rhythmic patterns could be generated, several marked deviations from an idealized undulatory behavior were observed. We also created a simple mechanical or "kinematic" model to visualize how network activity might be transformed into larval behaviors. This kinematic model could be driven by either neural network activity or alternatively by a set of descending control parameters.