Mnemonic Techniques and Lie Detection: Accuracy of Truth and Deception Judgments in Repeated Accounts

This 2021 study examined whether observers could improve deception detection accuracy by learning about memory-based cues from mnemonic interviews. Participants (n=92) read truthful and deceptive repeated statements and made veracity judgments, with one group receiving instruction about diagnostic cues. Despite explicit guidance, informed participants performed no better than uninformed observers, revealing the difficulty of changing entrenched deception beliefs.

Between-subjects experimental design with 92 participants reading transcribed immediate and delayed interview statements from a previous mnemonic interview study, with one group receiving instruction about diagnostic cues before making veracity judgments.

Background & Context

Lie detection remains one of the most challenging tasks in investigative interviewing, with both laypeople and professionals typically achieving accuracy rates barely above chance level. Studies have shown that accuracy in detecting deception typically is not much better than chance level when assessing speech or behaviour. This persistent difficulty stems largely from reliance on unreliable behavioral cues rather than diagnostic verbal content.

This 2021 study by Izotovas, Vrij, and Strömwall represents an important extension of memory-based lie detection research examining whether the use of memory-enhancing techniques (mnemonics) in interviews can be helpful to distinguish truth tellers from liars. The research builds directly on a 2018 laboratory experiment that demonstrated memory mnemonics amplify differences between truthful and deceptive accounts, particularly in delayed interviews where genuine memory decay patterns become apparent.

The current experiment addresses a critical practical question: Can observers actually use these memory-based cues to improve their detection accuracy? This bridges the gap between identifying diagnostic cues in controlled research and applying those insights in real-world credibility assessments.

Research Design & Methodology

Participants (n = 92) were asked to read the repeated statements reported by participants in the Izotovas et al.'s (2018) study and decide whether the statements they read were truthful or deceptive. The study employed a between-subjects experimental design with two conditions: informed and uninformed.

The key experimental manipulation involved providing one group with explicit information about diagnostic cues identified in the original 2018 study before they made veracity judgments. Specifically, the informed group was told that truth-tellers reported more details than liars in immediate interviews and again after a delay, and that truth-tellers, but not liars, showed patterns of reporting indicative of genuine memory decay. Observers read transcribed interviews from both immediate and two-week delayed sessions.

The study measured both accuracy rates (correct identifications of truth-tellers and liars) and self-reported cues that participants relied upon when making their judgments. This allowed researchers to examine the relationship between stated decision-making criteria and actual performance.

Results & Key Findings

The central finding was unexpected and significant: the informed and uninformed participants were not statistically significantly different in their veracity judgements. Despite receiving explicit instruction about diagnostic cues, observers given this information performed no better than those making intuitive judgments.

Analysis of self-reported cues revealed why this occurred:

• The majority of participants relied on unreliable cues, including the informed ones
• Signs related to nervousness (speech errors, hesitations) were amongst the most prevalent cues mentioned by participants
• Although participants in the informed group indicated that they relied on the information provided in the instruction, their accuracy was not higher than participants in the uninformed group

This pattern suggests a disconnect between conscious awareness of diagnostic cues and actual application during credibility assessment. Even when told what to look for, observers defaulted to ingrained but erroneous beliefs about deception.

Discussion & Significance

The findings reveal a sobering reality about changing deception detection practices. The results for the informed group support that such views are difficult to change, even with direct instruction. This aligns with broader research demonstrating that lay people and practitioners tend to hold incorrect beliefs about deception that are remarkably resistant to modification.

The researchers propose that the instruction given to the informed group was not effective to achieve improvements in deception detection accuracy. They note that previous training in interviewing to detect deception resulted in enhanced accuracy, but it involved at least a few hours of training including theoretical information, practical examples, exercises, and performance feedback—far more extensive than the brief instruction provided in this study.

The persistence of reliance on nervousness-related cues likely stems from what researchers call the "moral explanation": the stereotypical view that lying is bad, and therefore people should feel ashamed and/or nervous about it and display signs of nervousness. This deeply ingrained belief appears to override even explicit contrary instruction.

Limitations & Considerations

The primary limitation involves the training intervention itself. The brief written instruction, while conveying accurate information about diagnostic cues, lacked the depth and practice components shown to be effective in prior training research. This raises the question of whether more comprehensive training would overcome observers' reliance on stereotypical deception beliefs.

Additionally, the study used transcribed statements rather than live interviews or video recordings. While this controlled for nonverbal behavior, it may have made the task more abstract and difficult for participants. The statements themselves came from a laboratory paradigm involving mock intelligence scenarios, which may differ from real-world investigative contexts.

Practical Applications

For polygraph examiners and investigative interviewers, this research carries an important cautionary message: simply knowing about diagnostic verbal cues is insufficient for improving detection accuracy. Awareness must be coupled with extensive training, practice with feedback, and likely ongoing quality assurance to overcome deeply entrenched but erroneous beliefs about deception indicators.

The findings suggest that effective deception detection training programs must go far beyond information transfer. They must include structured practice in applying memory-based cues, immediate corrective feedback, and perhaps decision-support tools that guide analysts toward diagnostic indicators and away from misleading behavioral stereotypes. For consumers of polygraph services, this underscores the importance of examiner training quality and adherence to evidence-based assessment protocols rather than intuitive judgments.