Two experiments examined the effect of subliminal affective priming on compliance using the foot-in-the-door (FITD) paradigm. Prior to the target request, participants were exposed to subliminal emotional expressions. FITD (presence vs absence of initial request) was crossed with Priming (positive, negative, neutral, and absence of prime-blank screen) in a between-subjects design. 180 students volunteered as participants (M=22 years). 20 participants (10 females) were assigned to each of eight experimental conditions plus the control condition that neither involved the initial request nor the priming experiment. Participants were asked to judge whether target sentences were relevant or not for road safety instruction. In Experiment 1, emotional valence of prime stimuli affected both endorsement rate and time devoted to the target request but not participants' attitude. Affective priming effects did not interact significantly with the FITD effect. In Experiment 2, in 180 more students, the attitude measure was replaced by an implicit recognition task. Results showed that regardless of priming condition, in the absence of FITD, participants recognized target sentences better than in the presence of FITD. Conversely, in the presence of the FITD, participants recognized more accurately previously seen sentences that were primed by positive emotions relative to other priming conditions. The latter result suggests that the presence of the FITD involves a significant amount of cognitive resources so that only stimuli emotionally relevant to the task's goal (i.e., positive) tend to be processed. Together, these results could explain how, contrary to helping behavior, compliant behavior that has no direct association with the prime stimuli was not easily influenced by the affective subliminal priming.