Neuronavigation systems are developed to support the brain surgery operations. Because of its complex anatomical structure, the neurosurgery is a risky and critical operation. The surgeon is required to perform the operation in a very small area with very restricted movements. The neuronavigation systems are developed to help the surgeon during the operation to show the current position of the surgery with respect to the 3D virtual model of the patient. In these systems, the 3D virtual model of the patient is created according to the medical data (MRI/BT) of the patient. Hence these systems work like navigations systems that are used in driving a car. The surgeon uses this system by controlling the system through a software interface and its user interface and correlates the current position of the operation with the 3D patient virtual model. In this way the surgeon checks the critical anatomical structures through this system and eliminates possible risks. Hence surgeons who will perform such operations are required to develop several skills to manage this very complicated environment. They are required to perform the operation according to the information coming from the navigation display. Additionally, in order to reach relevant information from the navigation display they have to control the navigation panel. In order to prepare surgeons to manage this very complicated environment, their required skills need to be improved during the training period. In this study, to better understand the surgeons' behaviours while managing the tasks related to the surgical navigation procedures, a simulation based environment is developed and an experimental study is conducted with 10 people. Their eye data and their performance data is recorded based on the simulated tasks. The results of the study is analysed statistically and descriptively. The results show that it is possible to control a neuronavigation display through eye movements which could be an alternative human-computer interaction option for designing the neuronavigation systems' user interfaces. Secondly, it is shown that performing a task according to the results of a second information source (neuronavigation system) lowers the general performance in terms of travelled distance with the operation tool and camera (endoscope). However the success level while performing each task and the time spent values are similar in both cases. On the other hand the number of errors is higher in the first scenario. Hence, the surgical education programs need to provide appropriate solutions to better understand and measure the skill levels of trainees on such tasks and to improve their skills through virtual practice systems.