Over the past few decades, surgical procedures have made enormous progress, shifting from traditional open procedures to less and less invasive approaches, with the promise of smaller incisions, less complications, better cosmetic results and shorter recovery times. With these developments came a reduced dexterity and more complex control through the fulcrum effect and modified eye-hand coordination. These complications were greatly mitigated by the recent introduction of surgical robots, allowing the surgeon to sit in a comfortable posture, and restoring natural visuomotor coordination. To date, the issue of the lack of haptic feedback, which initially allowed the surgeon to intervene without vision, identify pathological tissue, feel arteries, etc., has not been resolved, despite the fact that it is crucial in certain interventions. Besides, the added value of haptic feedback is subject to controversy. Surgeons experienced in robotic surgery have adapted to the lack of haptic feedback and learned how to rely only on vision and proprioceptive cues as compensation. Nevertheless, previous studies have shown that substituting haptic information through various sensory channels can increase surgeons' performance. As a complete restoration of the sense of touch is extremely challenging in minimally invasive surgery, the advantages of haptic feedback have to be demonstrated and quantified to justify the additional cost and complexity. This is the aim of the present work. We hypothesized that tactile feedback as well as force is crucial in surgery and we investigated the haptic information involved in several representative surgical tasks. Dedicated hardware and a virtual reality environment to simulate suturing and palpation tasks were developed to address these questions. Ergonomic design guidelines were established based on an in-depth literature review and surgeons' comments gathered in a survey. These guidelines were then used to design and benchmark an ergonomic haptic handle featuring active grasping feedback and additional safety features. The results of the first two studies suggest that the benefits of haptic information highly depend on the surgical task in question. During a suturing task, force feedback significantly increased users' accuracy whereas torque feedback did not result in any significant improvement. In a palpation task, a higher recognition rate was achieved with tactile feedback than with visual sensory substitution. The performance of the haptic device integrating the ergonomic handle was assessed and compared with the standard omega.7 haptic device. Results showed that the index of performance of the original device was not degraded with the additional hardware. The index of performance was slightly increased for a manipulation task involving orientations. The ergonomic assessment of the handle showed a slight decrease of the tension in the adductor pollicis and flexor digitorium muscles, and therefore a potential decrease of fatigue. Although just a subset of surgical tasks could be investigated, the results indicate that haptic feedback and sensory substitution would greatly benefit teleoperated robotic surgery. Providing the surgeon with tactile information can potentially restore the feeling of "contact with the patient" and other surgeries that are highly reliant on the sense of touch could become possible again. This thesis presents a novel multidisciplinary approach to systematically analyzing surgical tasks in order to improve safety in two dimensions. Firstly, in the area of patient safety by providing the surgeon with haptic information to increase performance and reduce the risk of errors and secondly, in the area of surgeon safety by providing a more ergonomic master console. We expect this work to be a first step in establishing a general design method for more ergonomic surgeon consoles and trust that it will inspire research to investigate the sensory mechanisms underlying highly dexterous tasks such as those performed in surgery.