Masked stimuli (primes) can affect the preparation of a motor response to subsequently presented target. In numerous studies, it has been shown that the compatibility effect is biphasic as it develops over time: positive (benefits for compatible trials and costs for incompatible trials) for short prime-target temporal distances and negative (benefits for incompatible trials and costs for compatible trials) for long ones. What triggers the 2nd phase is the matter of the current debate. According to the self-inhibition hypothesis, the motor response elicited by a prime is automatically followed by an inhibition phase. The object-updating and mask-triggered inhibition hypotheses assume that this phase is triggered by the mask, provided that it contains features calling for the alternative response. In the present study, the author shows that the compatibility effect is modulated on the temporal position of a nonmasking distractor presented after the prime and before the target. With a distractor possessing task-relevant features, the compatibility effect was found to be negative for short prime-distractor intervals and for moderate prime-target intervals. The consequences of these results for the 3 hypotheses are discussed.