Object localization through the lateral line system of fish: theory and experiment.

Fish acquire information about their aquatic environment by means of their mechanosensory lateral-line system. This system consists of superficial and canal neuromasts that sense perturbations in the water surrounding them. Based on a hydrodynamic model presented here, we propose a mechanism through which fish can localize the source of these perturbations. In doing so we include the curvature of the fish body, a realistic lateral line canal inter-pore distance for the lateral-line canals, and the surface boundary layer. Using our model to explore receptor behavior based on experimental data of responses to dipole stimuli we suggest that superficial and canal neuromasts employ the same mechanism, hence provide the same type of input to the central nervous system. The analytical predictions agree well with spiking responses recorded experimentally from primary lateral-line nerve fibers. From this, and taking into account the central organization of the lateral-line system, we present a simple biophysical model for determining the distance to a source.

Estimating position and velocity of a submerged moving object by the clawed frog Xenopus and by fish--a cybernetic approach.

The lateral-line system is a unique facility of aquatic animals to locate predator, prey, or conspecifics. We present a detailed model of how the clawed frog Xenopus, or fish, can localize submerged moving objects in three dimensions by using their lateral-line system. In so doing we develop two models of a slightly different nature. First, we exploit the characteristic properties of the velocity field, such as zeros and maxima or minima, that a moving object generates at the lateral-line organs and that are directly accessible neuronally, in the context of a simplified geometry. In addition, we show that the associated neuronal model is robust with respect to noise. Though we focus on the superficial neuromasts of Xenopus the same arguments apply mutatis mutandis to the canal lateral-line system of fish. Second, we present a full-blown three-dimensional reconstruction of the source on the basis of a maximum likelihood argument.