Deception detection and question effects: testing truth-default theory predictions in South Korea.

This article focuses on testing Truth-Default Theory (TDT) and its proposition that the way interviewees are questioned significantly affects deception-detection accuracy. Conducted with 81 South Korean university students assessing English-language interviews of American participants, the study replicated prior findings showing that certain question types can either reduce accuracy below chance (around 30%) or improve it above the typical meta-analytic average of 54%, reaching up to 65%. The results support TDT's claim that diagnostic, context-rich questions enhance lie detection, while poorly designed questions can produce anti-diagnostic answers that mislead judges, especially among non-native speakers who may rely more on demeanor than nuanced content. The study highlights the importance of language proficiency and cultural context in deception detection and suggests that question design plays a crucial role in accuracy across cultures.