Abstract
We examined the Devil’s Advocate lie detection method which is aimed at detecting lying about opinions. In this approach, participants give reasons for why they hold an opinion in the eliciting-opinion question and counter-arguments to their opinion in a devil’s advocate question. Compared to Mann et al. (2022), we extended the number of eliciting-opinion questions (from one to three) and simplified the devil’s advocate question. Truth tellers (n = 55) reported their true opinion about protestor actions, whereas lie tellers (n = 55) reported the opposite of their true opinion. In the interview, participants answered the eliciting-opinion questions and the devil’s advocate question about their alleged opinion. The answers were coded for number of arguments and plausibility, immediacy, clarity and scriptedness of the answers. Data were analysed with analyses of variance with Veracity being the sole factor. Supporting the hypothesis, truth tellers provided more pro-arguments than lie tellers and their answers sounded more plausible, immediate and clear than lie tellers’ answers. We found this for all three eliciting-opinion questions. The opposite pattern was predicted for the devil’s advocate question but not found, likely caused by the simplification of the question. Neither was being scripted a diagnostic veracity indicator.
Original language | English |
---|---|
Journal | Psychiatry, Psychology and Law |
Early online date | 19 Sept 2023 |
DOIs | |
Publication status | Early online - 19 Sept 2023 |
Keywords
- Devil’s Advocate Question
- Lie detection
- Interviewing
- Protestor actions
- Verbal lie detection
- opinions
- clarity
- plausibility
- immediacy
- arguments
- UKRI
- ESRC
- ES/N009614/1