Challenging interviewees during interviews: The potential effects on lie detection

Abstract

The impact of interview styles on lie detection was examined. Thirty-six truth tellers and liars (undergraduates) were interviewed in three different ways: The interview started with an information-gathering interview style (Phase 1), which then developed into an interrogation (Phase 2), and finally transformed back into an information-gathering interview style (Phase 3). On the basis of DePaulo and Kirkendol's Motivational Impairment Effect, which suggests that when people are highly motivated to get away with their lies they actually look like they are controlling their behavior, it was hypothesized that the most prominent differences between liars and truth tellers would emerge in Phase 3. The findings supported this hypothesis, and thus suggest that specific interview styles could facilitate lie detection.

Keywords:

• Interviewing to detect deception
• Motivational Impairment Effect
• behavioral adaptation effect
• overshadowing effect

This study was sponsored by the Economic and Social Research Council (grant R000222820)

Notes

1. This three-factor model is based upon DePaulo, Stone, and Lassiter's (1985a) and Zuckerman, DePaulo, and Rosenthal's (1981) four-factor model, which also included “arousal” as an additional factor. Vrij (2000) left this factor out as it shows an overlap with the emotion factor. Zuckerman et al. (1981, p. 9) themselves already suggested this by finishing their arousal factor paragraph with the following statement: “It is possible, however, that the general autonomic responsivity to deception reflects specific emotions. If so, cues to deception may be accounted for by the particular affects that are involved rather than by general arousal.”

2. It might seem suspicious from the participant's perspective that the experimenter would know that an interview was about to ensue in order to ascertain the truth of the blackboard mishap. However, none of the participants in this study raised this issue. In other words, we have no evidence to suggest that the participants were suspicious.

3. Initially, latency period (period of time between the question being asked and the answer given, r=0.38) and pauses (frequency of noticeable pauses between words, r=0.22) were coded, but they were dropped due to low inter-rater reliability between the two coders.

4. Additionally, MANOVAs were carried out with Veracity as factor and all seven behaviors (rather than just smiles, and hand and finger movements) as dependent variables. Separate MANOVAs were carried out for each phase. At a multivariate level, only the MANOVA for Phase 3 was significant, F(7,28) = 3.81, p<0.01, η ²=0.49 (Phase 1: F(7,28) = 1.94, NS; Phase 2: F(7,28) = 1.76, NS). Univariate tests in all three MANOVAs showed no significant effects other than those mentioned in the main text.