clusive. A judgment of a significant response is normally followed by further questioning and possibly further testing with single-issue format polygraph tests.

The stimulation test, sometimes called the stim test or acquaintance test, is used by examiners in some test formats either during the pretest or between charts. Examinees are presented with a question set of very similar items and directed to lie about one. The examinee may be asked to pick one of several playing cards (card test) or to pick a number between three and seven (numbers test), and then to deny having picked each of the cards or numbers while connected to the polygraph machine. The main purpose of the procedure is to induce or strengthen in examinees the expectation that the polygraph can accurately determine the truthfulness of their answers.

Concealed information tests (more often called guilty knowledge or concealed knowledge tests) present examinees with sets of very similar items, much in the manner of stimulation tests, except that the similar items include one true and several (usually, four) false details of some aspect of an incident under investigation that has not been publicized, so that the true answer would be known only to the investigators and to those present at the incident. In a burglary, examinees might be asked about several possible points of entry into the house, one of which the burglar actually used. (For more detail about question construction and administration of concealed information tests, see Nakayama [2002].) When an examinee is asked whether he or she used each of these routes, the answer is expected to be negative regardless of the examinee’s innocence or guilt. Guilty examinees are expected to reveal their concealed knowledge by responding more strongly to the true item than to the others.

Concealed information tests are applicable only under restricted conditions: when there is a specific incident, activity, or thing that can be the subject of questioning and when there are several relevant details that are known only to investigators and those present at the incident. Thus, these tests are not applicable in typical screening situations in which the only possible relevant questions concern generic events, such as unspecified acts of espionage that may or may not have occurred.