Since the 1960s, researchers have been reporting that stress reduces Stroop interference. This is puzzling, as stress and anxiety typically have deleterious effects on cognitive control and performance. The traditional explanation is that stress reduces 'cue utilisation': it withdraws attentional resources from less-relevant stimuli (including the distracter word), meaning that the target colour is left with a stronger influence over response-selection. However, it could also be that stress somehow boosts distracter inhibition, or some other aspect of executive control. To test these two accounts, 59 students completed a Stroop task featuring occasional startlingly-loud sounds (high stress), or the same sounds at a lower, comfortable volume (low stress). Alongside standard Stroop interference, two measures of executive control-negative priming and conflict adaptation-were calculated from the Stroop data. Stress produced a clear reduction of Stroop interference, but it did not influence negative priming, and no conflict adaptation effects were detected at all. These findings support the cue utilisation account. Furthermore, for the first time stress was shown to reduce Stroop interference in a task with no congruent trials, showing that the effect does not result from stress's modulating any strategy changes participants might make in response to congruent trials.