Abstract : Studying heart transplanted patients records provide interesting data since modulation of the heart rate is not due to neural activity . Studying these patients reveals the so called mechanical modulation (MM) that is increased as the exercise intensity is higher. In a recent paper we have introduced the PFM model that relates the observed amplitude of the MM to the ventilation and the mean heart rate, with a set of non transplanted subjects. In this paper we use a time-frequency representation added to a t-test to show whether the HRV signal, or tachogram, contains or not power in the respiration frequency band. This approach is well adapted to the purely dynamic condition corresponding to increasing exercise. Observation models such as PFM and IPFM can be accounted in the analysis since there is no evidence of a linear interaction between observed values from the heart rate variability signal and physiological input such as the ventilation. The aforementioned models are compared regards our proposed t-test.