In order to increase the realism of medical simulators, haptic interfaces could be used to simulate the patient's body behavior. It is especially interesting to reproduce the stiffness of different soft tissues with corresponding haptic behaviors. In this paper, two control laws, impedance control and backstepping associated with a closed-loop stiffness tuning, are introduced and applied to a pneumatic actuator. Both controllers have been obtained by using the AT transform which is suitable to model the behavior of a pneumatic system, in a strict-feedback form. Both control laws allow to tune the system stiffness. A comparison of their performances is presented, based on experimental results. I. INTRODUCTION Medical staffs require continuing hands-on training on medical methods. During their education, they usually train on cadavers or animals (when available), and more recently on passive and active simulators, before training on real patients. It has been proven that computer-based haptic training simulators lead to an efficient training for advanced tasks [1]. They offer residents a risk-free training in order to improve their experience before performing real procedures. However, to be efficient, this kind of training needs to be realistic. It is therefore necessary to simulate correctly the human body behavior, that is to say to obtain a realistic haptic feedback. A way to provide a realistic haptic rendering is to generate coordinated forces and motions on the haptic interface close to the ones measured on the real tool. Another way is to render an impedance (a stiffness and a damping) close to the one of the patient's body. Nowadays, for each simulation need, the corresponding haptic interface has to be dedicated to the application because generic commercial haptic devices are not always suitable [2]. For practical reasons, in commercial simulators, electric actuators are commonly used, in order to reproduce the force feedback mimicking the response of the human body behavior to medical tool interaction. Indeed, the control laws for electric actuators are quite well mastered and easy to set up. However they have drawbacks such as a low power to weight ratio and difficulties to provide at the same time a high torque at high speed, and mechanical limitations in their backdrivability. This limits their performances to render a variable stiffness. On the opposite side, pneumatic