LIAR?

-Samantha Mann, Aldert Vrij and Ray Bull

SUSPECT, LIES AND VIDEO TAPES: AN ANALYSIS OF AUTHENTIC HIGH-

STAKE LIARS

Background

This study is one of the very few, and the most extensive to date, which has examined deceptive behavior in a real-life, high-stakes setting. The behavior of 16 suspects in their police interviews has been analyzed. Clips of video footage have been selected where other sources (reliable witness statements and forensic evidence) provide evidence that the suspect lied or told the truth.

Aim• To determine if there are systematic

behavioral indicators to distinguish between those who are telling lies and those who are telling the truth.

• To determine if cognitive load (complex cognitive processes) causes changes in behavior relevant to lying or telling the truth

•Deceptive behaviour• Veracity

•Real life high stake

lies•Cognitive

Load•Blind testers

METHOD : laboratory experimentDESIGN:

Repeated measure

SAMPLE•4 participants were juveniles :3 aged 13 and other aged 15• Remainder were adult < 65 years•15 were Caucasian (where English was their first language)•1 participant was of Asian ethnicity (a male whose first language was Punjabi but who was fluent in English).• All interviews were conducted in English. 13 3

SAMPLE•Range of crimes:1. theft (N =9)2. arson (N =2)3. attempted rape (N =1)4. murder (N =4). •The majority of participants (at least 10 of 16) were well known to the police and had been interviewed on several occasions previously, relating to previously committed offenses.

PROCEDURE• Police detectives at Kent County Constabulary, UK, were asked

if they could recollect videotaped interviews in which they were involved where the suspect had lied at one point and told the truth at another.

• The end result was an hour-long videotape consisting of clips from 16 suspects. Truths that were selected were chosen so as to be as comparable as possible in nature to the lies (a truthful response to an easy question such as giving a name and address was not comparable to a deceitful response regarding whether or not the suspect has committed a murder. Video-footage about names and addresses were therefore not included as truths in this study).

• Total of 65 video clips (27 truth / 38 lies)• Length of clip and length of response varied but not significant

in terms of analysis of behavior.• Two observers independently coded behavior and were

compared for inter-rater reliability on a sample of the videos (not all of them)

• Behaviours recorded were: 1) gaze aversion, 2) blinking, 3) head movements, 4) self-manipulations, 5) illustrators, 6) Hand-finger movements, 7) speech disturbances 8)pauses

• Given the small number of participants and relative large number of behaviors, researchers reduced behavioral indices by clustering the three types of hand and arm movements (illustrators, self manipulations, and hand/finger movements) into one category “hand and arm movements.” This clustering had no effect on the results.

• Each of the coded behaviors was transformed into a format in order that truths and lies may be directly compared.

For example: Hand & Arm Movements X 60Total length of truth/lying

RESULTS• Data was analyzed using Multivariate

Analysis of Variance (MANOVA): a statistical test procedure for comparing multivariate (population) means of several groups.

• Veracity was the only within subject factor and the dependent variables were the six listed in the table.

BEHAVIOURTRUTHFUL DECEPTIVE

MEAN SD MEAN SD

GAZE AVERSION 27.82 9.25 27.78 11.76

BLINKS 23.56 10.28 18.50 8.44

HEAD MOVEMENTS 26.57 12.34 27.53 20.93

HAND/ARM MOVEMENTS 15.32 14.35 10.80 9.99

PAUSES 3.73 5.14 5.31 4.95

SPEECH DISTURBANCES 5.22 3.79 5.34 4.93

Individual differences occurred and there was no particular behaviour that all liars exhibited.

50% showed an increase and 50% showed a decrease in head movements and speech disturbances while lying.

56% 44% - gaze aversionMore participants (69%) showed an than in

hand movements during deception. Blinking and pausing as the majority of

participant paused longer and blinked less while lying (81%)

EVALUATION OF RESULTS

DISCUSSION

• The findings of the study give some support for the cognitive load process in explaining deceptive behaviour.

• Falsifies the stereotypical view of increased blinking; Nixon effect

• However, neither cognitive load nor nervousness were measured or manipulated before the study conclusions were speculative

APPLICATIONS

Challenges the simplistic views of professional lie catchers that a typical of deceptive behaviours exists.

Highlights the importance of establishing a baseline behaviour before attempting to detect deception.

STRENGTHS

•Control of variables in lab experiment;• inter-rater reliability;• specific behaviors identified and

measured;•quantitative data- reliable

Different interviewers were used for different participants.

Sometimes more than one interviewer was present.The total number of people present varied. Though these factors may have influenced the

person’s behaviour, the researchers controlled these factors using within factor design.

No comparison between high stake liars and high stake truth tellers.

Small sample size= not generalizable (this was because of the limited availability of appropriate tapes and the time consuming process of obtaining them) limits generalizability to criminals in police custody only.

People who have been arrested more than once might experience less fear and guilt, be more experienced liars.

LIMITATIONS

Real life settingMost participants familiar with police

interviewingRage of crimes

LOW ECOLOGICAL VALIDITYMany of the crimes used are relatively rare

ECOLOGICAL VALIDITY

THE END