Back

Evaluating Physiological Indicators in Detecting Deception using the Comparison Question Test (CQT)

Rashid, A. M. S.; Carmichael, B.; Su, C.; Shi, K.; Lim, K.; Kumar, P. S.; Wan, N. J.; Govil, E.; Yap, D.

2024-09-06 bioengineering
10.1101/2024.09.01.609623 bioRxiv
Show abstract

Despite significant advancements in deception detection, traditional methods often fall short in real-world applications. This study addresses these limitations by evaluating the effectiveness of various physiological measures -- pupil response, electrodermal activity (EDA), heart rate (HR), and facial temperature changes -- in predicting deception using the Comparison Question Test (CQT). It also fills a critical research gap by validating these methods within an Asian context. Employing a between-subjects design, data was collected from a diverse sample of 118 participants from Singapore, including Chinese, Indian, and Malay individuals. The research aims to identify which physiological indicators, in combination, offer the most robust predictions of deceptive behavior. Key innovations include the adaptation of the CQT with a modified directed lie paradigm and an expanded sample size to assess the relative importance of each physiological measure. The studys findings reveal that pupil response is the most significant predictor of deception, with EDA enhancing the models explanatory power. HR, while relevant, adds limited value when combined with pupil response and EDA, and facial temperature changes were statistically non-significant. The study highlights the need for further research into the interactions among physiological measures and their application in varied contexts. This research contributes valuable insights into improving deception detection methodologies and sets the stage for future investigations that could incorporate additional physiological indicators and explore real-world applications.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.