Once a cheater, always a cheater?
In their paper in the Archives of Sexual Behavior, Knopp and colleagues review the literature on infidelity and report their recent work looking at predictors for having affairs . They note that despite a growing body of research, there is scant data about the role of prior affairs in predicting infidelity over time. They primarily were interested in the factors underlying serial infidelity, and understanding to what degree infidelity in one relationship predicted infidelity in the next.
Anecdotally, this is what we see — when someone has been unfaithful in the past, they are more likely to be unfaithful again. I've often observed that people who engage in infidelity often experienced infidelity in their parent's relationship. While I haven't seen any data on this, it makes sense as familiarity with a behavior in general makes it more likely to do that behavior, via social learning and modeling. This holds true for many other behaviors, where past personal history and family behavior partially predicts individual behavior, including aggression and suicide — though it is worth noting that such history is not 100 percent predictive, and many with the risk factor do not go on to repeat the behavior.