TAN, EW; COWAN, NJ; FORTUNE, ES; Johns Hopkins Univ.: Whole-animal Closed-loop Locomotor Performance of Weakly Electric Gymnotiform Fishes

Eigenmannia and Sternopygus actively maintain their positions within refuges and will track the motion of salient objects in their environment during daylight. We developed a mechanism to assess the closed-loop, whole-animal sensorimotor performance in these and other Gymnotiform species, using a procedure similar to Rose and Canfield (1993). We constructed a pen-plotter driven refuge (shuttle) with two parallel slotted polycarbonate walls (20cm X 5cm, separated by 5cm). Each wall has six 1.5cm wide slots (1.5cm spacing) filled with electrically transparent ceramic. To assess the contributions of sensory systems to whole-animal performance, we degraded aspects of the sensory information available to individuals. To modulate electrosensory information, we used electrodes in the tank to deliver biologically relevant electrical signals. Ambient light levels and reduced visual cues on the shuttle were used to regulate the contribution of the fish’s visual system. A 30Hz video camera system attached to the shuttle captured the fish’s behavior. The location of the fish and shuttle were tracked using MATLAB. Test stimuli included ~10cm peak-to-peak fixed-frequency sinusoidal shuttle motions ranging from 0.05Hz to 1Hz for 4 to 8 cycles per trial. We measured the relative magnitude and phase of the responses of individual fish. As a first step in modeling the whole-animal closed-loop tracking dynamics, we fit a second order, linear “mass-spring-damper” style model to the data. Under this model, the closed-loop response of Eigenmannia virescens appeared to be either critically- or slightly over-damped. Sternopygus macrurus showed better tracking at all frequencies. The closed-loop bandwidths for these species were in the range of ~0.2 to ~0.5 Hz.