A black box test consists of a series of small-scale experiments that can be distributed to labs that are responsible for forensic facial examination in crime prevention and criminal justice applications. The current project developed and implemented tests that measure skills across a broad range of facial identification tasks with image and video data. The project had three components: 1) the collection of normative performance data on potential image and video stimuli, using untrained human observers and facial identification software; 2) performance of extensive item analyses on the individual stimuli used in a recent test of 27 facial identification experts from the Facial Identification Scientific Working Group; and 3) the evaluation and refinement of the black box test through an iterative release to participating labs. The researchers believe the project results will widely impact the fields of forensic psychology and criminal justice policy and practice in the United States. This research shows for the first time the high skill levels of forensic facial examiners and reviewers compared with control groups of forensically trained fingerprint examiners and the general population. This project also shows that state-of-the-art face recognition algorithms compare favorably with trained human experts.