Neuromuscular control of trout swimming in a vortex street: implications for energy economy during the Kármán gait

Approximating the complexity of natural locomotor conditions provides insight into the diversity of mechanisms that enable animals to successfully navigate through their environment. When exposed to vortices shed from a cylinder, fishes hold station by adopting a mode of locomotion called the Kármán gait, whereby the body of the fish displays large, lateral oscillations and the tail-beat frequency matches the vortex shedding frequency of the cylinder. Although field studies indicate that fishes often prefer turbulent flows over uniform currents, the effect of hydrodynamic perturbations on the mechanics, control and energetics of locomotion is still poorly understood. In this study, electromyography is used to measure red and white axial muscle activity for rainbow trout (Oncorhynchus mykiss) holding station in a vortex street. When trout Kármán gait, they show a significantly reduced but still rhythmic pattern of muscle activity compared with that seen when they swim steadily in uniform flow. Specifically, trout selectively activated only their anterior red axial muscles and abandoned the antero-posterior wave of red muscle activity that drives undulatory locomotion in uniform flow. This supports a previously proposed hypothesis that trout are not just swimming in the reduced flow behind a cylinder (drafting). Anterior axial muscle activity was correlated to head amplitude during steady swimming but not during the Kármán gait, indicating that while activation of muscles during the Kármán gait may aid in stability and control, vortices determined overall head motion. Furthermore, anterior red axial muscle activity, the only region of muscle activity shared between both the Kármán gait and steady swimming, had a lower intensity and longer duration during the Kármán gait. At times when paired fins were active during the Kármán gait, there was no axial muscle activity measured, lending support to a passive mechanism of thrust generation in oscillating flows. Comparisons with dead trout towed behind a cylinder confirm this intriguing observation that live trout may temporarily adopt the Kármán gait with no axial muscle activity, revealing paradoxically that at times fish can passively move against turbulent flow. To Kármán gait for prolonged periods, however, trout must adapt to the demands of turbulence by eliciting a shift in neural control strategy. By decoupling motor output both down and across the body, the pattern of rhythmic Kármán gait muscle activity may reflect the entrainment of a central pattern generator to environmental vortices.