In Neil v. Biggers, the Supreme Court considered the admissibility of an eyewitness identification based on a suggestive 'show-up' confrontation between the victim and the defendant. While conceding that the procedure used was unnecessarily suggestive, the Court held that the identification was nonetheless admissible if it could be deemed 'reliable' based on a totality of the circumstances. The circumstances affecting reliability were indicated by the Court to be (1) the witness' opportunity to view a person, (2) the degree of attention which a witness focuses on a person, (3) the accuracy of a witness' initial description of a person, (4) the level of certainty expressed by the witness in making the identification, and (5) the time that has elapsed between observation and identification. In order to test these reliability standards, an experiment was conducted in a classroom which consisted of a man entering the classroom in the midst of a lecture and forcing a woman student to leave with him. Neither the man or woman was known to the students. Students were asked immediately to write descriptions of the involved parties, and the following day, show-ups were conducted before the class. In an effort to simulate the Supreme Court standards for reliability of identification, student witnesses were classified according to opportunity to observe the parties in the incident and the accuracy of initial descriptions. Accuracy of identification according to elapsed time was not measured since all students viewed the show-up at the same time. Patterns of accuracy were not significant enough to confirm the Supreme Court standards as a measuring device for reliability of identification. 'Opportunity to observe' carried the greatest weight in identification accuracy. A significant finding was that witnesses were more likely to identify the person of their own sex. Footnotes arre provided.