What impact does the model statement have on future intentions for liars and truth tellers?

What impact does the model statement have on future intentions for liars and truth tellers?

Eleanor Evans

2017

Current literature states that there is a marked difference between statements given by truth tellers in comparison to liars. This difference is seemingly determined from when the cognitive load for participants is increased and liars struggle more. There is also evidence from distinctions in the linguistic make-up of the statements. Thirty-six undergraduate participants took part in a study exploring the effect of the model statement on truth tellers compared to liars when discussing a future event. All participants gave their first statement, then listened to the model statement before giving their second statement. Participants also filled out a questionnaire after completing the interview. All interviews were transcribed and analysed using CBCA, WMatrix and ANOVA. Results indicated that while there was a clear effect of the model statement, there was no significant effect of veracity from the CBCA and ANOVA analysis. On the other hand, WMatrix indicated differences in veracity. In conclusion, both truth tellers and liars were able to increase the amount of information between their first and second statements, thus providing an effect of the model statement. However, there were nevertheless distinct differences between the language used by participants under both conditions; suggesting that there are in fact marked differences between truth tellers and liars when discussing a future event.

liars
manipulation
model statement
truth tellers
veracity

The model statement (Appendix A) is a 734-word document and was replicated from Leal et al. (2015). It is known that the model statement had an effect in the Leal et al. (2015) experiment and therefore it seemed appropriate to remain consistent with using the exact same model statement for this study. The interview questions (Appendix B) were also adapted from the Leal et al. (2015) study.
Questionnaires (Appendix C and Appendix D) were fashioned for each condition. The material for the questionnaires was largely developed for this experiment with the questions specifically tailored to relate to the interview. Both questionnaires contained a total of nine questions, a mix of Likert-scale and open-ended questions.
A digirecorder was used to record all participant interviews.
This study is a 2 (within factor: the manipulation through using the model statement) x 2 (between factor: the veracity, participants are either in the truth telling condition or the lying condition) mixed effects ANOVA. The experiment was carried out in two sections, the first being the interview and the second involving a questionnaire.
Procedure
Before completing the experiment, all participants were provided with an information sheet (Appendix E). Participants met with the researcher so that they could be briefed about their task: to discuss the day in which the participants would go and collect their degree results. In addition, the participants were given a consent form (Appendix F), which they were required to fill out in order to participate. Participants took part in the experiment individually.
Once consent was given the participants were further briefed about which condition they would be participating in – either the truth telling condition or the lying condition. Participants were randomly assigned to one of the conditions. All participants were told as a baseline for the interview that the normality for collecting their degree results was that they would go to their subject department on the day, show their library card, and be given an envelope with their degree results in. Those participants placed in the Truth Telling condition (N = 16) were told to answer the questions asked by the interviewer as truthfully as possible to the best of their knowledge. Participants in the Lying condition (N = 17) were informed to lie when answering the interviewer’s questions. Participants in the lying condition were told that they could either lie about one element of the intended event or all aspects of it.
A third party member, who was unaware of the participants’ veracity status, carried out the interview; this is how the interview was conduced in the Leal et al. (2015) experiment. During the interview the participants were asked the first question of “OK, just so I can understand, I am going to need you to take me to the day that you will collect your degree results, and tell me in as much detail as possible everything that will happen from before you collect your results through to you receiving your results”. Participants then gave their first statement in response to the question. Preceding this, all participants in both conditions were exposed to an example statement after the interviewer said, “I know that sometimes people are not sure just how much detail to include. In order to give you an idea of what I am looking for I'd like to play you an example of what we consider a detailed answer”. This example statement, known as the model statement was a recording of someone dictating the details of an event that has no relevance to what the participants were asked to talk about during the experiment. Following the model statement, the interviewer asked “OK, I know that wasn’t too relevant to your story but hopefully you have an idea of the amount of detail it takes for us to get a clear rounded idea of how the event will go! Could you now please tell me in as much detail as possible everything that will happen from before you collect your results through to you receiving your results” and the participants proceeded to give their second statement.
Once the participants had completed the interview section of the experiment, they were given a questionnaire to fill out. The questionnaire was tailored to whichever condition they were allocated – there was a separate questionnaire for the Truth Tellers and the Liars. All participants in the lying condition were asked as part of their questionnaire to state the element(s) they had lied about.
Data Analysis
After the data collection, the participants’ interviews were transcribed and a primary analysis was performed using Criteria-Based Content Analysis (CBCA). There are a total of 19 possible criteria for analysing statements, however only a certain number were selected for the purpose of this study. The chosen criteria can be seen in table 1. Each statement was given a score corresponding with each individual criterion, leading to an overall CBCA score. Each criterion was scored between 0 and 2: 0, if the criterion is not found in the statement; 1 if it is a present, but only a small amount; and 2, if the criterion was found frequently throughout the statement. Following CBCA coding, the scores were analysed in SPSS using a repeated measures ANOVA, in which the overall scores for each of the statements could be compared.
Table 1
List of criteria used for CBCA coding and the descriptions
General Characteristics

1.Logical Structure
Coherency of the statement in terms of not containing logical inconsistences or contradictions
2. Unstructured Production
The presentation of the information in a (non) chronological order
3. Quantity of Details
The inclusion of specific descriptions of place, time, persons, objects and events
Specific Contents

4. Contextual Embedding
Events being placed in time and location, and actions being connected with other daily activities and/or customs
5. Description of Interactions
Information that interlinks at least the alleged perpetrator and witness
7. Unexpected Complications During the Incident
Elements incorporated in the statement that are somewhat expected
8. Unusual Details
Details of people, objects or events that are unique, unexpected or surprising but meaningful in the context
9. Superfluous Details
Details in connection with the allegations that are not essential for the accusation
11. Related External Associations
Events are reported that are not actually part of the alleged offence but are merely related to the alleged offence
12. Accounts of Subjective Mental State
Development and change in feelings experienced at the time of the incident
Motivated-Related Contents

14. Spontaneous Corrections
Corrections that are made or information that is added to material previously provided in the statement without having been prompted by the interviewer
15. Admitting Lack of memory
An unprompted interviewee admitting lack of memory either by saying “I don’t know” or “I don’t remember”
Note. This list of criteria was adapted from Vrij, A. (2005). Criteria-Based Content Analysis: A Qualitative Review of the First 37 Studies. Psychology, Public Policy and Law. 11(1), 3-41

WMatrix (Rayson, 2008) was used in addition to CBCA and SPSS. All the statements were separated into files and uploaded onto WMatrix so that linguistic analysis could commence. WMatrix is a software program that allows for corpus linguistic analysis and comparison. It provides frequencies and percentages of how the words are distributed in a given text; it also lists the concordances for reference. It produces tables from the output for each comparison and ranks the words based on their log-likelihood. The log-likelihood is an indicator based on the difference among frequencies, in this instance how often a particular word is used by participants. The word count and interview duration were also calculated and used in the analysis.

Lancaster University

data/SPSS.sav

Evans2017

John Towse

Open

English

Data

LA1 4YF

Lara Warmelink

MSc

Cognitive Psychology

Thirty-six participants (males 12 and females 24) were recruited to take part in the study. All participants were Undergraduate Lancaster University Students over the age of 18 (Mage = 20.29, SD = 1.36)

ANOVA