The Accuracy and Consistency of Polygraph Examiners’ Diagnoses

Field study with criminal cases finding examiner accuracy of 92.4% for deceptive and 95.5% for truthful subjects using structured scoring.

Field study of criminal cases with confirmed outcomes, using structured scoring methods.

Hunter and Ash examined real criminal cases with confirmed outcomes, finding examiner accuracy of 92.4% for correctly identifying deceptive subjects and 95.5% for correctly identifying truthful subjects when using structured scoring methods. The high accuracy for truthful subjects is particularly noteworthy for protecting the innocent.

The 95.5% accuracy for truthful subjects demonstrates strong protection against false accusations in criminal investigations.

Background & Context

The Hunter and Ash (1973) study emerged during a pivotal era in polygraph research when the field was transitioning from purely subjective chart interpretation to more standardized, structured scoring approaches. Published in the inaugural volume of the Journal of Police Science and Administration, this research represented one of the early attempts to scientifically validate polygraph accuracy using real criminal cases with confirmed outcomes.

The study was conducted using case files from John E. Reid & Associates, a prominent polygraph firm, and examined various criminal offenses with verification through confession. This work was part of a broader movement in the 1970s to establish the empirical foundations of polygraph testing, addressing growing concerns about the technique's scientific validity and its role in criminal justice.

The research was later grouped with other Reid organization studies that faced criticism for failing to use control question techniques accepted by federal agencies and for relying on examiners trained in behavioral symptom observation methods. Despite these later critiques, the study provided important early field data on examiner accuracy in real-world criminal investigations.

Research Design & Methodology

This was a field validity study that tested polygraph accuracy by comparing blind evaluations of polygraph examiners against a criterion of verification by confession using files from a polygraph testing firm. The research examined actual criminal cases where ground truth had been established through post-examination confessions, allowing researchers to retroactively assess the accuracy of polygraph decisions.

Cases came from the firm of John E. Reid & Associates and involved various criminal offenses. The study used blind examiner evaluations, with evaluators coming from the same Reid organization. The key methodological approach involved having independent examiners review polygraph charts without knowledge of the confession-verified outcomes, then calculating accuracy rates by comparing their diagnoses to the established ground truth.

The study employed structured scoring methods—a relatively advanced approach for its time—marking a shift away from purely clinical, impressionistic chart interpretation. Cases were selected based on having confirmed outcomes through confession, though like most field studies of this era, neither cases nor examiners were selected randomly, and the cases of only certain examiners were sampled.

Results & Key Findings

The study reported 92.4% accuracy for identifying deceptive subjects and 95.5% accuracy for identifying truthful subjects when using structured scoring methods. These accuracy rates represented some of the highest figures reported in early polygraph field research, particularly notable for the exceptionally high accuracy in correctly classifying truthful examinees.

Correct guilty detections in field studies reviewed by the OTA ranged from 70.6% to 98.6% across different conditions, with one condition in the Wicklander and Hunter study reaching the upper limit. Correct innocent detections showed even greater variability across field studies, ranging from 12.5% to 94.1%. The Hunter and Ash study's 95.5% accuracy for truthful subjects placed it at the high end of this distribution.

Key statistical outcomes included:

• Deceptive detection rate: 92.4% accuracy in correctly identifying confirmed deceptive subjects
• Truthful detection rate: 95.5% accuracy in correctly identifying confirmed truthful subjects
• Significantly higher protection for innocent subjects compared to many contemporaneous studies
• Demonstrated value of structured scoring over purely clinical interpretation

Discussion & Significance

The Hunter and Ash findings contributed to early optimism about polygraph accuracy in criminal investigations, particularly regarding the technique's ability to protect innocent suspects from false accusations. The 95.5% accuracy for truthful subjects was especially significant, as false positive errors—incorrectly labeling truthful people as deceptive—have greater consequences in criminal contexts than false negatives.

However, the study faced subsequent criticism along with other Reid organization research for failing to use control question techniques accepted by federal agencies and for relying on examiners trained in behavioral symptom observation methods rather than numerical chart evaluation. The reliance on confession as ground truth created selection bias concerns, as studies using confessions may examine only a select sample of examinations—illustrated by one contemporary study obtaining only 16 verified cases from 92 total cases.

The OTA report noted that results among Reid studies did not vary substantially, though the greatest deviation occurred when examiners received additional behavioral and demographic information about suspects. The consistency of findings across Reid organization studies suggested either robust methodology or systematic biases inherent to their particular approach and training methods.

Limitations & Considerations

Serious sampling problems affected most field studies of this era: neither cases, examiners, nor evaluators were selected randomly, and nonrandom selection raised questions about whether results represented polygraph testing in general or only a subgroup of practitioners. The use of cases from a single commercial firm with a specific training philosophy limits generalizability to other polygraph approaches and examiner populations.

• Confession criterion bias: Cases verified by confession may differ systematically from unverified cases
• Lack of random sampling: Cases likely came from specific examiners rather than random selection
• Methodological transparency: Limited information about inconclusive rates and decision thresholds
• Reid technique specificity: Results may not generalize to federal agency methods or other polygraph approaches
• Evaluator training effects: Blind evaluators shared the same training background as original examiners

When random sampling was attempted in contemporaneous studies, high rejection rates of selected cases created additional sample bias problems, and the magnitude of selection bias from using confession criteria remained unknown.

Practical Applications

For criminal investigators and polygraph practitioners, this study provided early evidence supporting the use of structured scoring methods over purely subjective chart interpretation. The high accuracy for truthful subjects offered reassurance about polygraph's potential role in exonerating innocent suspects—a critical consideration given the consequences of false accusations in criminal contexts.

However, modern practitioners should interpret these findings cautiously. Contemporary polygraph standards emphasize validated numerical scoring systems, randomized quality control procedures, and recognition that accuracy rates vary significantly based on technique, examiner training, and case characteristics. The study's historical importance lies more in demonstrating the value of systematic field research and structured scoring than in providing definitive accuracy benchmarks applicable to current practice. Today's evidence-based approach requires consideration of multiple field studies using diverse methodologies, with recognition that no single study—particularly from this era—can definitively establish polygraph validity across all applications.