Considering all the elements related to the occurrence of accidents, an important aspect is to identify the drivers most prone to accidents. This study aims to evaluate the behavior of drivers concerning gender and age. For this, a stretch of highway with a dangerous curve was modeled in a simulator using an immersive virtual environment and 59 people experienced the simulation. The car speed was registered along the route and from a segmented regression model, the behavior for both groups was evaluated regarding average speed. The results showed that older drivers (26 to 35 years old) of the female gender slowed the car before the other profile groups. Also, younger drivers (18 to 25 years old) took longer to slow down and had more accidents in the simulations. These results show that there are differences in driver’s behavior concerning gender and age.