Description of Victims.
In every experiment the particular position of each of the victims was determined randomly (by a random number generator). It was felt that the victim's location and behavior might have a bearing on his/her being found. Typically a victim might be out in the open or under cover, and, furthermore, the victim might be passive (simulating an unconscious or dead person) or active (simulating a conscious person trying to attract attention by waving, etc.). Of these four possible combinations it seemed that a victim who was unconscious and under cover would be the most difficult to detect, while a victim attracting attention in the open would be the easiest. In view of the fact that in practice the figures obtained from these experiments would be used to decide when to STOP the multiple searching of an area, it was important that the figures obtained should be on the conservative side. Consequently, five out of six experiments performed were devoted entirely to victims who were both unconscious and under cover. The remaining experiment dealt with what was (and is) thought to be the other extreme, victims attracting attention in the open. [In fact, all 8 of the victims were found in this latter experiment, which reinforced the opinion that to be conservative the tables should be derived exclusively from the least favorable situation, viz. the results of the remaining five experiments].

In all experiments victims were asked to wear "everyday" clothing (which ranged from white T-shirts and blue jeans to brown and green checked shirts and khaki trousers) and were not permitted to wear clothing with large areas of orange or bright yellow (thought to be highly visible). The use of any signaling device (e.g. mirror) was prohibited. Finally if a victim's assignment involved being under cover, he/she was instructed not to try to hide, merely to seek cover.