Rates of smoking among populations with psychiatric disorders are more than twice as high as among the general population, yet there are few controlled studies of the mechanisms of this relationship. This trial examined the effect of nicotine by way of cigarette smoking on both self-report and autonomic parameters of anxiety among individuals with PTSD who were also heavy smokers. Half of the participants were randomized to an anxiety-eliciting condition, whereas the other half were assigned to a condition of neutral emotional content. Within each of these conditions, participants were randomized into a nicotine condition (high-yield nicotine cigarette) or a placebo condition (denicotinized cigarette). The final layer of randomization involved assigning participants to either an attention-demanding task or a no task condition. We examined the interactive effects of nicotine and attentional demand on parameters of anxious responding. Nicotine exerted an anxiolytic effect relative to placebo on self-report measures; however, nicotine did not interact with attentional demand in producing this effect. In contrast to the findings on the self-report measures of distress, nicotine enhanced autonomic reactivity to trauma cues, thereby suggesting a decoupling of self-reported anxiety and autonomic arousal associated with responding to trauma-relevant cues. The implications for understanding smoking and psychiatric comorbidity are discussed.