The recent development of motorcycle simulators has made it possible to study rider behavior in safe conditions. However, their use still raises validity issues. Our study examined how riders’ steering and gaze behaviors and subjective experience are influenced by motorcycle roll tilt and reverse steering, which are considered to be essential factors in real-life motorcycle riding. The results revealed that tilting the motorcycle in the roll plane did not lead to significant changes in rider behavior, gaze sampling, or perceived realism. The steering control strategy adopted by riders did, however, significantly influence these results. A direct steering control strategy meant that riders took a racing path and scanned the road far in advance. When reverse steering was implemented, however, riders chose to take a “safety path”, as recommended by training manuals. Reverse steering also received the highest realism score. However, steering control was more difficult, as shown by the larger number of lane departures recorded and a change in the trade-off between guiding and look-ahead fixations. This suggests that although reverse steering matches riders’ real control behavior and improves the subjective experience of simulator riding, it is hindered by an inadequate internal model of vehicle dynamics.