Reversal theory provides a new approach to understanding ex-smokers' behavior during highly tempting situations. Hypotheses derived from the theory were tested in 2 studies of the highly tempting situations of ex-smokers drawn from community smoking cessation programs. Study 1 consisted of interviews with 55 Ss (25 men and 30 women, mean age 37 years) conducted 3 months after cessation. Study 2 consisted of interviews with 104 Ss (45 men and 59 women, mean age 41 years) conducted 6, 9, 12, and 15 months after cessation. All interviews were coded using reversal theory constructs. Results supported the hypotheses that individuals in paratelic or negativistic states were more likely to lapse than individuals in telic/conformist states and that cigarette availability was related to lapses in paratelic, but not telic, states. Reversal theory constructs accurately classified 93% and 85% of the cases, suggesting that the theory provides an improved model for understanding behavior during highly tempting situations.