A comparison of the effectiveness of two types of deceit detection training methods in older adults

Abstract

Background In general, people are poor at detecting deception. Older adults are even worse than young adults at detecting deceit, which might make them uniquely vulnerable to certain types of financial fraud. One reason for poor deceit detection abilities is that lay theories of cues to deception are not valid. This study compared the effectiveness of two training methods to improve deceit detection among older adults: valid facial cues versus valid verbal cues to deception. Approximately 150 older adults were randomly assigned to facial training, verbal training, or a control condition. Participants completed a pre-test deceit detection task, their assigned training, and a post-test deceit detection task. Results Both training groups significantly improved at recognizing their respectively trained cues after training. However, the facial cue training group were less accurate at detecting deception post-test compared to pre-test and the control group exhibited improved accuracy of deceit detection from pre-test to post-test. Conclusions These results are consistent with the body of literature on deception suggesting people hover around chance accuracy, even after training. Older adults’ facial and verbal cue recognition can be improved with training, but these improvements did not translate into more accurate deceit detection, and actually hampered performance in the facial condition. Older adults showed the most benefit from sheer practice at detecting deception (in the control condition), perhaps because this condition encouraged implicit rather than explicit judgments of deception.

Interoceptive Accuracy Enhances Deception Detection in Older Adults

“…There were three central findings: (i) greater IAcc was associated with better lie detection; (ii) this effect strengthened with greater age among older adults; and (iii) greater IAcc was associated with better phishing email detection with increasing age in our older adult cohort. Integrating parallel lines of research on age and deception detection 14,15 and IAcc 24,25 , we provide evidence that greater IAcc enhances deception detection across the continuum of older adulthood, observed on two independent, naturalistic deception tasks (video-based lie telling and email-based phishing).…”

“…Factors contributing to older adults' susceptibility to fraud include demographics, such as income and literacy 4 , as well as psychological factors such as depression 5 , age-related changes in the brain resulting in declining executive functioning and decision-making capacity [6][7][8][9] , and increased levels of trust [10][11][12][13] . In fact, compared to young adults, older adults are typically less accurate at detecting deception 14,15 and exhibit a greater truth bias, which is the tendency to believe others are telling the truth rather than lying 16,17 . These findings are in accordance with evidence that interpersonal trust increases across the adult lifespan 18,19 .…”

Measuring the Candidates' Emotions in Political Debates Based on Facial Expression Recognition Techniques

“…As for the raw data provided by the iMotions platform on emotions, they were calculated to obtain a binary result with a threshold of 10, meaning that all those facial expressions that, with at least 10% probability, were rated as such by a human evaluator are given as valid (see Table 2 ). This criterion is the one used by e.g., Kjærstad et al ( 2020 ) and Stanley and Webster ( 2019 ).…”

Structured groups make more accurate veracity judgements than individuals

“…Statements judged with 60% accuracy were used because previous research has found that the veracity of written statements is judged at levels no better than chance (Masip et al, 2012) and therefore interventions to boost deception detection performance, such as training or measures to elicit cues, are required before group effects would be expected to emerge. Baseline detection rates were artificially increased to 60% so as to mimic the increase such successful interventions would have on performance (Colwell et al, 2012; De Turck et al, 1997; Porter et al, 2000; Stanley & Webster, 2019; Vrij, Evans, et al, 2004).…”