|
|
Thesis/Dissertation Information
- Degree Disciplines:
- Psychology
Notes
- Abstract:
- Lie detection plays a critical role in criminal investigations. Decades of research has suggested that human lie detection ability is poor, particularly when using behavioral indicators of deception such as nervousness. However, recent studies have suggested that cognitive theories of lie detection and verbal indicators of deception, may prove more effective when attempting to differentiate between liars and truth tellers. The purpose of the present study was to investigate the utility of linguistic cues when differentiating between liars and truth tellers when standard interview questions and questions modeled after Behavior Analysis Interview (BAI) are used in an investigation with innocent and guilty participants. The study was archival and included a sample size of 33 participants (FIU students) ranging in age from 18 to 43. In the original study, participants were randomly assigned to steal money from an office (guilty) versus not (innocent). All participants were interviewed by a trained research assistant and the interviews were recorded. For the present study, the interviews were transcribed verbatim and coded for the presence of linguistic indicators of deception: word count, third-person pronouns, hedges (i.e., maybe), “negative” content words, and overzealous expressions. It was hypothesized that the guilty participants would differ with regards to the content of their responses to the BAI-type questions (speculative) compared to direct questions. Additionally, it was hypothesized when asked to speculate, guilty participants would respond with fewer words and more linguistic indications compared to innocent participants. Thus, it was hypothesized that the question type (speculative vs. direct) would interact with the participants’ guilt when assessing the effectiveness of linguistic indicators. A 2 (participant type: guilty vs. innocent) x 2 (question type: speculative vs. direct) factorial MANOVA examined differences between liars and truth tellers as a function of the question type on word count and the combined number of linguistic indicators of deception. Analyses revealed that participants spoke more words and revealed more deceptive markers when asked speculative vs. direct questions. Contrary to hypotheses, there were no statistically significant differences between liars and truth tellers in the number of words spoken or the number of deceptive markers and no significant guilty by question type interaction. Although not statistically significant, an examination of the means revealed that the predicted interaction was in the expected direction. That is, differences between guilty and innocent individuals in word count and deceptive markers was more pronounced when speculative versus direct questions were used. Exploratory analyses on the individual linguistic indicators demonstrated that hedges, words conveying uncertainty, were the only linguistic indicator that differed significantly between liars and truth tellers, with liars conveying significantly more hedges than truth tellers. The present findings highlight the need to replicate real-world scenarios as much as possible when examining the effectiveness of deception techniques and the importance of gathering additional data on potential linguistic differences between guilty and innocent individuals when inviting suspects to speculate, as in the BAI. As participants spoke more and revealed more deceptive indicators when speculative versus direct questions were used, future research should explore this strategy to differentiate reliably between liars and truth-tellers.
- Thesis:
- Thesis (M.S.)--Barry University, 2019.
- Bibliography:
- Includes bibliographical references (leaves 47-49).
Record Information
- Source Institution:
- Barry University
- Holding Location:
- Barry University Archives and Special Collections
- Rights Management:
- Copyright Taira Medina. Permission granted to Barry University to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
- Resource Identifier:
- P128.I53 M43 2019_MedinaTaira ( BU-Local )
- Classification:
- P128.I53 M43 2019 ( lcc )
|
|
