The authors conclude that polygraph has the same scientific status as medicine, psychiatry, engineering, forensic sciences, and other fields that attempt to make critical diagnostic decisions from complex, raw data that is collected and interpreted by trained, experienced professionals. The paper first discusses some key concepts, i.e., “validity,” “reliability,” and “variance.” “Validity” refers to how well polygraph decisions match known truth. It is generally recognized that the polygraph does not directly detect lies, the criterion of interest; it merely monitors and records selected physiological functioning within the body of the examinee, from which the examiner draws inferences about the examinee’s truthfulness. The validity of polygraph is gauged by how well the underlying phenomenon (differential physiological arousal) predicts deception. “Reliability” is the measure of repeatability or reproducibility. Test-retest reliability relates to testing the same person with the same test on multiple occasions in order to assess whether the same outcome is reached on all occasions. “Variance” pertains to differences among populations and individuals regarding physiological responses to similar stimuli, which bears upon whether the polygraph has universal application as a detector of deception. Following a discussion of the aforementioned concepts, this paper discusses “levels of rules,” which focuses on empirical data relevant to the psychophysiological phenomena determined empirically to be related to deceptive responses to polygraph questions. This is followed by a discussion of assigning numbers to various polygraph readings as measures of likelihood of deception. The paper next discusses “decision rules,” which pertain to how the examiner interprets the numerical scoring of polygraph charts in order to determine whether deception is indicated. 6 figures and 32 references