Under the guise of evaluating a head-up display in a driving simulator, participants completed scrambled sentence tasks (while waiting at stop signs) designed to prime an elderly stereotype. Driving speed and driving time between stop signs in this Elderly Stereotype condition were compared to a Control condition in which age non-specific words were substituted for elderly stereotyped words. Participants had a lower maximum speed and longer driving time in the Elderly Stereotype condition than in the Control condition. This effect was obtained even though the participants were completely unaware of the theme in the experimental condition. Theoretical, as well as applied implications are discussed.