A Video-Based Screening System for Automated Risk Assessment Using Nuanced Facial Features

Microexpressions Are Not the Best Way to Catch a Liar

“…Two experiments (Pentland et al, 2017 ) tested the RE directly with videotapes from high-stakes experiments in which emotional and microexpressions were captured with the Computer Expression Recognition Toolkit. In the first experiment, guilty subjects completing a CIT (which controls for cognitive load) showed far less variance in four deception-relevant emotions (disgust, fear, sadness, and surprise) than did innocent subjects when responding to target questions.…”

Design Principles for Signal Detection in Modern Job Application Systems: Identifying Fabricated Qualifications

“…Facial movement was one candidate selected to serve as a signal. We expected overall facial movement to decrease during deception, based on prior research identifying facial freeze as an indicator of a deceptive response during short answers [58]. While some research has shown decreases in movement that are noticeable to a trained eye [e.g., 62,69], this recent IS research identified decreases in facial movement that, because of their small scale, must be detected through analysis of data generated via computer vision.…”

“…This minute behavioral difference was discovered in the context of one-word answers to structured questioning. While SIGHT systems prescribe a similar question structure, the responses are expected to be considerably longer than one word, creating potential for noise in a signal that previous research reported to be statistically significant but minor in terms of scale [58]. However, the mechanism driving facial freeze appears to be either a natural fight-or-flight response or a natural tendency to try to appear truthful [66]-and both of these mechanisms should be expected to persist throughout a response, regardless of the response duration.…”

“…This new generation of decision support systems is especially meaningful in situations where human behaviors cannot be manually examined, or where behavioral deviations are so slight that they require high-precision instruments to measure. IS scholars are increasingly leveraging day-to-day sensing instruments such as human-computer input devices [33] and standard video cameras [58] to capture and analyze human behaviors. The use of ubiquitous sensing instruments allows for remote collection and analysis and is essential in cases involving large volumes of human traffic.…”

Separating the Wheat From the Chaff: Guidance From New Technologies for Detecting Deception in the Courtroom

“…The aforementioned research frequently points to liars reducing postural, gestural, head and facial activity to the point of crossing a thin line between appearing composed and appearing wooden, rigid and unnatural (31). This generalized inhibition and rigidity across trunk, limbs, head, and face may reflect overcontrol of felt arousal and negative emotions (32, 33). Even when told about this trend, liars fail to increase their movement (34).…”