In the ECG recordings during tilt test, changes in RR series characteristics can be observed. The purpose of the present study is to investigate to what extent these changes can be explained by changes in the atrial input and in the atrioventricular (AV) nodal characteristics induced by the autonomic nervous system (ANS). Average RR series characteristics (mean, rmssd and sample entropy) and average atrial fibrillatory rate (AFR) were obtained from 24 patients during supine, head-down tilt (HDT), and head-up tilt (HUT). Simulations were performed using an AV node model consisting of a network of interacting node; each node with a refractory period (R) and conduction delay (D) dependent on the stimulation history. In an extension to the AV node model, R and D were scaled to account for ANS induced changes and simulations were performed with the original and extended model, respectively. The atrial impulse series entering the AV node was modelled as a point process with time-varying mean and std determined by the AFR. The clinical RR mean, RR rmssd and RR sample entropy decreased from supine to HDT and decreased further from HDT to HUT. Simulation results indicate that the model must account for ANS-induced changes to replicate the observed response in RR series characteristics, while alterations in the atrial input alone are insufficient to replicate the response.