Is the inference rule of the control question polygraph technique plausible?

Two experiments were designed to examine if the inference rule suggested for the controversial control question technique (CQT) makes sense. In Experiment 1 a story of a hypothetical theft was presented to 20 police interrogators. Half of the participants received a story about a guilty suspect and the other half a similar story that described an innocent suspect. Interrogators were asked to evaluate the concern that this person would feel when various questions were presented to him in a polygraph test. Results supported the reasoning of the inference rule proposed for the CQT. Experiment 2 extended the plausibility question to predictions about the relative frequencies of truthful and deceptive outcomes. Opponents of the CQT argued that many innocent examinees would show larger physiological responses to relevant than to control questions and thus yield deceptive outcomes. Analysis of a random sample of actual polygraph tests showed, that truthful outcomes are more common than predicted. It was suggested that some critical views of the CQT tended to overestimate the CQT false positive error rate.

Mixed-methods study: Experiment 1 used hypothetical vignettes with experienced interrogators; Experiment 2 analyzed a random sample of actual field polygraph tests to examine outcome frequencies.

Background & Context

The Control Question Test (CQT) has long been the most widely used polygraph format in criminal investigation and law enforcement, yet its theoretical foundation remains hotly contested. The CQT inference rule predicts that guilty examinees will show greater concern about specific relevant questions than about more general control questions, while innocent examinees will manifest the opposite pattern, having more concern about control questions to which they are lying than about relevant questions to which they are responding with the truth.

The major argument against the CQT inference rule is that it is not logical, because for a suspect in a real-life crime, the polygraph test is but one part of the broader investigative process in which the relevant questions play a major role; therefore, the relevant questions that focus on the investigated crime should not be expected to yield the concern equivalent to the more general control questions; this should be true for both innocent and guilty suspects. This fundamental challenge to the CQT's rationale prompted Elaad to design two experiments testing whether the inference rule holds up under empirical scrutiny.

Research Design & Methodology

In Experiment 1, a story of a hypothetical theft was presented to 20 police interrogators; half of the participants were presented with a story about a guilty suspect, and the other half were given a similar story that described an innocent suspect. The interrogators were asked to assess the concern that this person would experience when various questions were presented to him in a polygraph test. This design leveraged the professional judgment of experienced interrogators to evaluate whether the CQT's theoretical predictions about differential concern aligned with real-world understanding of suspect psychology.

The second experiment extended the plausibility question about the CQT to predictions about the relative frequencies of truthful and deceptive outcomes. Opponents of the CQT argued that many innocent examinees would show larger physiological responses to relevant questions than to control questions, thus yielding deceptive outcomes. Analysis of a random sample of actual polygraph tests showed that truthful outcomes were more common than predicted. This field data analysis provided real-world validation beyond the hypothetical scenarios of Experiment 1.

Results & Key Findings

Results supported the reasoning of the inference rule proposed for the CQT. The experienced police interrogators' assessments confirmed that the differential attention hypothesis makes psychological sense when evaluated by practitioners familiar with criminal investigations.

The field data analysis in Experiment 2 yielded particularly compelling findings:

• Truthful outcomes were more common than predicted by CQT critics, contradicting the hypothesis that innocent examinees would frequently show false positive results
• Analysis of a random sample of actual polygraph tests showed that truthful outcomes were more common than predicted, suggesting critics may have overestimated false positive rates
• The CQT is not more or less valid than indicated by the critical view of the CQT, but the less frequent false negative outcomes (indicating that a guilty examinee is truthful) replace the more frequent false positive errors

Discussion & Significance

This research directly addresses one of the most persistent theoretical objections to the CQT. While critics like Ben-Shakhar, Lykken, and the 2003 National Research Council have argued that the CQT's inference rule lacks logical foundation, Elaad's 2003 study presents a different result from these critical assessments. The study demonstrates that when tested empirically through professional judgment and field data, the inference rule appears more plausible than critics suggest.

The findings have important implications for the ongoing debate about CQT construct validity. The fact that experienced interrogators' intuitions aligned with CQT predictions, and that field outcomes showed fewer false positives than critics predicted, suggests the technique may capture real psychological dynamics even if the theoretical mechanism remains incompletely understood. This challenges the assumption that theoretical concerns about the CQT necessarily translate into practical invalidity.

Limitations & Considerations

The first experiment relied on hypothetical scenarios rather than actual examinees, meaning interrogators were making judgments about imagined psychological states rather than observing real physiological reactions. The sample size of 20 interrogators was relatively small, limiting generalizability.

Experiment 2's field data analysis, while addressing real-world outcomes, could not control for all the factors that influence polygraph test results in actual criminal investigations. The study examined outcome distributions but did not provide information about base rates of guilt versus innocence in the sample, making it difficult to calculate precise accuracy metrics. Additionally, the analysis relied on polygraph examiners' original decisions rather than independent ground truth verification.

Practical Applications

For polygraph examiners, this research provides empirical support for the psychological rationale underlying CQT question design and interpretation. The finding that practitioners' intuitions align with CQT predictions validates the experiential knowledge base of the profession. The evidence that false positive rates may be lower than critics suggest should inform discussions about appropriate contexts for CQT use, particularly in investigative screening where the consequences of errors differ from adjudicative settings.

For consumers and policymakers evaluating polygraph evidence, this study highlights that theoretical criticisms of the CQT do not necessarily predict practical failure rates. However, it also underscores the importance of examining actual field performance data rather than relying solely on laboratory studies or theoretical arguments when assessing the technique's real-world utility.