This study examined the parameters under which risk compensation in driving can occur due to the use of safety belts. Risk compensation theories hypothesize that if individuals wear safety belts, they will . drive in a more risky manner than if they do not wear safety belts due to the increased perception of safety they provide. Although much of the current literature has debated the existence of risk compensation in driving for many years, until the current study an experimental analysis of the effect has not yet been conducted that permits a controlled examination of both between-subject and within-subject effects. Risk compensation was not found in the between-subject analyses of the present research, however the within-subject analyses demonstrated the risk compensation effect. Subjects drove significantly faster when they switched from not wearing a safety belt to wearing a safety belt than subjects who either did not switch belt use or drivers who switched from safety belt use to safety belt non-use. The study also suggested that the mechanism by which risk compensation occurs is that safety belt use makes drivers feel safer when they can compare the sensations wearing a safety belt vs. those when not wearing a safety belt. The risk compensation effect probably did not manifest itself in the between-subject studies because this comparison did not (and could not) take place. The implications of this study to driving real automobiles on multi-user roadways is discussed. Suggestions and examples of possible research to further expand the knowledge about how and when risk compensation occurs are also provided. / Ph. D. / incomplete_metadata