We investigated the effect of anxiety on police officers' shooting decisions. Thirty-six police officers participated and executed a low- and high-anxiety video-based test that required them to shoot or not shoot at rapidly appearing suspects that either had a gun and "shot," or had no gun and "surrendered." Anxiety was manipulated by turning on (high anxiety) or turning off (low anxiety) a so-called "shootback canon" that could fire small plastic bullets at the participants. When performing under anxiety, police officers showed a response bias toward shooting, implying that they accidentally shot more often at suspects that surrendered. Furthermore, shot accuracy was lower under anxiety and officers responded faster when suspects had a gun. Finally, because gaze behavior appeared to be unaffected by anxiety, it is concluded that when they were anxious, officers were more inclined to respond on the basis of threat-related inferences and expectations rather than objective, task-relevant visual information.