Main results indicated no significant differences between the polygraph group that did confess and the polygraph group that did not confess. Although no significant differences were found, the authors maintain that the confession criterion is suspected when it leads to samples of cases with non-representative data, such as scores that are more extreme than the population as a whole would produce. Data were gathered from the U.S. Army Criminal Investigation Detachment Polygraph Division on all polygraph cases for which ground truth confirmation could be established. Cases that were conducted between January 1995 and February 1997 and used common testing formats were analyzed, which consisted of a sample of 3,349 polygraph examinations. All polygraph examinations used the Axciton computer polygraph. The analysis relied on an objective scoring system developed at the U.S. Department of Defense Polygraph Institute which uses physiological tracing features previously demonstrated to be the most diagnostic: respiration line length, electrodermal response amplitude, and blood volume amplitude. Figure, references