The purpose of the study was to evaluate the dynamic nature of heart rate baroreflex under closed-loop conditions. We applied pressure perturbation with a random-interval cuff inflation to parametric system identification of the baroreflex. In 10 healthy men (six young, 27 +/- 3.9 years old, and four older, 50 +/- 3.8 years old), blood pressure variations were produced by a random-interval (0.06-0.50 Hz) cuff inflation around bilateral thighs to identify characteristics of the baroreflex with an autoregressive moving average model with exogenous input. The model faithfully described the dynamic relationship between mean blood pressure and RR interval. A theoretical 1 mmHg step input of mean blood pressure caused a rapid increase in RR intervals toward the steady state within 10 s in younger subjects. In older subjects, however, the step response showed a delayed and gradual increase in RR interval over 20 s. The steady-state gain obtained from the autoregressive moving average model with exogenous input was significantly greater than that determined from the slope of the pressure-RR relationship. The conclusion was that the dynamic baroreflex gain was quantified using a parametric system identification technique and blood pressure perturbation by means of a random-interval cuff inflation. This method would potentially be of great use in helping us gain insights into a closed-loop system of complex and dynamic cardiovascular regulation in humans.