The head-neck system has multiple degrees of freedom in both its control and response characteristics, but is often modeled as a single joint mechanical system. In this study, we have attempted to quantify the perturbation parameters that would elicit nonlinear responses in a single degree-of-freedom neuromechanical system at small amplitudes and velocities of perturbation. Twelve healthy young adults seated on a linear sled randomly received anterior-posterior sinusoidal translations with +/- 15 mm and +/- 25 mm peak displacements at 0.81, 1.76, and 2.25 Hz. Head angular velocity and angular position data were examined using a nonlinear phase-plane analysis. Poincare sections of the phase plane were computed and Lyapunov exponents calculated to measure divergence (chaotic behavior) or convergence (stable behavior) of system dynamics. Variability of head angular position and velocity across the entire phase plot was compared to that of the Poincare sections to quantify spatial-temporal irregularity. Multiple equilibrium points and positive Lyapunov exponents revealed chaotic behavior at 0.81 Hz at both amplitudes whereas responses at 1.76 and 2.25 Hz exhibited periodic oscillations, clustered phase points, and negative Lyapunov exponents. However, intersubject variability increased at the lowest frequency and a few subjects presented chaotic behavior at all frequencies. An inverted pendulum with position and velocity threshold nonlinearity was adopted as a simplistic model of the head and neck. Simulations with the model resulted in features similar to those observed in the experimental data. Our principal finding was that increasing the perturbation amplitude had a stabilizing effect on the behavior across frequencies. Nonlinear behaviors observed at the lowest stimulus frequency might be attributed to fluctuations in control between the multiple sensory inputs. Although this study has not conclusively pointed toward any single mechanism as responsible for the responses observed, it has revealed clear directions for further investigation. To examine if changing the sensory modalities would elicit a significant change in the nonlinear behaviors observed here, further experiments that target a patient population with some sort of sensory deficit are warranted. (C) 2005 American Institute of Physics.