**Rationale** In the real world, perpetrators may slip by potential eyewitnesses undetected, or witnesses may only orient toward an ongoing crime because of a loud noise. In the lab, many researchers try to encourage real-world-type conditions, but others warn participants of a crime before the video begins (perhaps partially due to ethical desires to not deceive participants, to give trigger warnings, etc.). Some researchers have even warned participants of an upcoming lineup task before they watch the crime video to intentionally manipulate the attention the participant would pay to the criminal rather than the details of the crime. After an initial sweep of some of the recent eyewitness identification literature, we saw that the instructions given about what to expect and what to attend to vary as we anticipated. This same sweep also convinced us that manipulating video instructions and measuring lineup performance and confidence is both not previously published and arguably critical for the field in two ways. It is critical for the field if we find that instructions change accuracy or confidence, because it seems that most studies either do not report their video instructions at all or do so in vague enough terms that reproduction would be impossible. In other words, there may be a source of variation in lineup accuracy and confidence that is not being controlled or reported. It may be the case that detailed instructions warning participants of crimes or telling them to watch the criminal inflate either accuracy or confidence, thus making our designs less externally valid. Wulff and Hyman (in prep) found that, in a long video depicting a hard-to-notice crime, instructions to watch for a crime did not generate significantly different performance on a later lineup than control instructions (i.e., watch the video), though far fewer participants noticed the crime in the control condition. Their findings may lead some to conclude that pre-video instructions will not influence lineup accuracy, but most laboratory studies of lineup accuracy show videos in which the crime is the obvious focal point. In Wulff & Hyman (in prep) the crime is not the obvious focal point of the video stimuli used in the study. In fact, some participants did not notice the crime at all. This methodological reality should be addressed in a follow-up study, though Wulff and Hyman convincingly argue that theirs is a more externally valid method of crime exposure. Further, their work addresses the problem of inattention in eyewitnessing, but it cannot speak to the possibility that a person could notice a crime occurring and try to pay attention to it to enable themselves to report its details accurately to police later. Experts in face memory have long wondered whether intentional memorization of faces differs from incidental memorization (starting with Warrington & Ackroyd, 1975), thus a study in which the participant-witness easily notices the crime would also make a theoretical contribution whether or not specific video instructions improve lineup performance. Thus, we seek to extend Wulff and Hyman's findings to more typical eyewitness identification methodology to add to the debate on intentional memorization of faces.