Gender Differences in Multimodal Contact-Free Deception Detection

Abstract

In this paper, we explore the hypothesis that multimodal features as well as demographic information can play an important role in increasing the performance of automatic lie detection. We introduce a large, multimodal deception detection dataset balanced across genders, and we analyze the patterns associated with the thermal, linguistic, and visual responses of liars and truth-tellers. We show that our multimodal noncontact deception detection approach can lead to a performance in the range of 60%–80%, with different modalities, different genders, and different domain settings playing a role in the accuracy of the system.