L13 Lie detection

Question 1

Define deception

Answer 1

• “A successful or unsuccessful deliberate attempt, without forewarning, to create in another a belief which the communicator considers to be untrue” (Vrij, 2000)

1. Must be intentional deception
2. Without forewarning (magicians aren’t liers because we expect to be deceived by them
3. Depends on the perspective of the deceiver (child ate cookies in cookie jar. Then those in cupboard. Says mum I want more cookies. She says there are no more cookies. She is lying to the child because she BELIEVES there are more, but there are not.
4. Involves two people, doesn’t include telling lies to ourselves

Question 2

Types of Lies (DePaulo et al., 1996)

Answer 2

1. Outright lies (guilty suspect says I didn’t do it)
2. Exaggerations (facts are overstated or info is conveyed in a way that exceeds the truth. E.g. forensic suspect embellishes the remorse to get a lighter sentence.)
3. Subtle lies (literal truths that are designed to mislead. E.g. concealing info, evading a question or omitting relevant details.)

Question 3

Reasons to Lie

• 5 reasons (Vrij, 2000):

Answer 3

–To gain personal advantage (business man knows company is not doing well. Don’t tell your stockholders)
–To avoid punishment (Say I didn’t commit the crime to avoid conviction)
–To make a positive impression on others (say how great we are)
–To protect themselves from embarrassment/ disapproval (don’t admit to mistakes)
–For the sake of social relationships (e.g. the movie “the invention of lying”. If you don’t lie, you would have to say outright,” I hoped you would be better looking” on a blind date”

Self- Oriented vs others oriented. (saying you did it so your son won’t go to jail)

Question 4

Frequency of Lying; Who Lies?

Answer 4

Children lie as soon as they have theory of mind, animals can lie (Koko the gorilla)
• American diary study: college students told 2 lies/day and community members told 1 lie/day. Most lies were self-serving (DePaulo, 1996)
• Frequency of lying depends on:
1. The personality and gender of the liar
2. The situation in which the lie is told
3. People to whom the lie is told

Question 5

Which personality types and gender lies more?

Answer 5

– Extroverts lie more than introverts
– Frequency of lies similar between men and women
–Women tell more social lies
–When dating, women lie to improve physical appearance, men lie to exaggerate earning potential

Question 6

How many people lie to jet a date or job?

Answer 6

– 90% lie to prospective date

– 83% lie to get a job

Question 7

Who do people tell lies to?

Answer 7

3. People to whom the lie is told:
   – Lowest rate of lying with spouses
   – Highest rate of lying with strangers
   – College students lie frequently to their mothers (in almost half of their conversations)

Question 8

Three ways to catch a liar

Answer 8

1. Observe their verbal and nonverbal behaviour*
2. Analyse the content of what they say*
3. Examine their physiological responses (lie detector tests)

Question 9

Some verbal and nonverbal cues are more likely to occur during deception than others, depending on:

Answer 9

1. Emotion (paul ekman’s approach)
2. Content complexity
3. Attempted behavioural control

Question 10

Paul Ekman’s emotional approach: Behavioural indicators of deceeption

Answer 10

1. – Deception results in different emotions: guilt (we tend to overt our gaze when guilty) , fear, excitement (duping delight = getting excited about the prospect of telling a lie, Leads to arousal = increase in limb movements, increased speech fillers ums and ahs, increased speech errors, incomplete sentences, higher pitched voice.)

However, sometimes people telling the truth leads to high arousal (if you’re concerned someone will falsly believe you’re lying)

– Strength of emotion depends on personality of liar and circumstances of lie
– Emotions may influence a liar’s Nonverbal Behaviour (NVB)

– NVB during deception should show signs of stress compared to baseline of typical NVB

Question 11

Why does Content complexity make lying can be difficult to do

Answer 11

• People engaged in cognitively complex tasks exhibit different nonverbal behaviors

Avoid contradicting evidence out there, remember evidence etc.
When we engage in complex tasks, we have more speech errors, pause more to think of the response, we neglect our body language, move our limbs less.

We avert our gaze to an emotionless point because we don’t want to get distracted by things.

In these ways content complexity can be a behavioural indicator of deception.

By creating highly difficult questions for people, you can distinguish between liars and truth tellers. E.g. “can you sketch out the area”. “tell things in reverse order” tell subject to look into your eyes

Question 12

What is the impact and consequence of liars attempting to control their behaviour?

Answer 12

• When liars do this, they sometimes over-control themselves, resulting in rehearsed and rigid behaviour
• Nonverbal behaviour is more difficult to control than verbal behaviour, a lot of people do attempt to control it.
- There are more automatic links between words and behaviour
- We are more practiced at controlling behaviour

Question 13

Behavioural Indicators of Deception: Verbal and Nonverbal Cues to Lying Meta-analyses by Sporer & Schwandt (2006; 2007)

Answer 13

• Verbal cues:
1. Higher pitch of voice
2. Increased response latency (takes us longer to respond)
3. Increased errors in speech (ums and ahs, mixing up sentences)
4. Shorter length of description
• Nonverbal Cues:
5. Decreased nodding (still and stoic)
6. Decreased foot and leg movements
7. Decreased hand movements
The literature is a bit mixed and this is quite hard to determine

Question 14

• Microexpressions

Answer 14

A fleeting facial expression discordant with the expressed emotion and usually suppressed within 1/5 to 1/25 of a second
• It is difficult to control facial communication and it can betray a deceiver’s true emotion to a trained observer (Ekman, 1992)
• Inconsistent emotional leakage occurred in 100% of participants at least once. Negative emotions were more difficult to falsify than happiness (Porter & ten Brinke, 2008)

Question 15

What is a Content Indicator of Deception

Answer 15

Statement Validity Assessment (SVA)

Question 16

HIstory of the Statement Validity Assessment (SVA)

Answer 16

• Developed in Germany to determine the credibility of child witnesses’ testimonies in trials for sexual offences
• Extended to adults and other types of cases
• SVA accepted in other European courts, but not UK courts. Opinion in US is divided.
• Has been presented in expert testimony in US, but main role in guiding police investigations and decisions of prosecutors (not permissible in court in Australia)
From: Vrij, 2005

Question 17

Statement Validity Assessment

• Consists of three major elements:

Answer 17

1. Semi-structured interview
2. Criteria-based content analysis (CBCA) of transcribed version of statement given during the interview
3. Evaluation of the CBCA outcome via a set of questions (validity check-list)

Question 18

Describe the CBCA: The Criteria -based Content Analysis

Answer 18

• Based on the “Undeutsch hypothesis”:
–A statement derived from memory of an actual experience differs in content and quality from a statement based on invention and fantasy (Undeutsch, 1987)
• Trained evaluators judge the presence or absence (or strength) of 19 criteria
• The presence of each criterion strengthens the hypothesis that the account is based on genuine experience
• But, absence of a criterion does not necessarily mean the statement is fabricated (Vrij, 2005)

Question 19

CBCA Criteria • General Characteristics

Answer 19

1. Logical structure
2. Unstructured production (often when people are telling the truth they are unorganised, shows lack of rehearsal)
3. Quantity of details (the more details you have the more likely it is you’re telling the truth)

Question 20

CBCA • Specific Contents

Answer 20

4. Contextual embedding
5. Descriptions of interactions (between people involved)
6. Reproductions of conversation (exact wording)
7. Unexpected complications during the incident (unexpected interruptions and abrupt endings e.g. the phone rang, didn’t answer it. Indicating of truth telling, liar wouldn’t put it in)

Question 21

CBCA • Peculiarities of Content

Answer 21

8. Unusual details
9. Superfluous details
10. Accurately reported details misunderstood (child says “he peed white on me” the interviewer understands but the child doesn’t)
11. Related external associations (telling a story, at ease saying external things. Child sexual assault interview, she says she went horse-riding once, unrelated)
12. Accounts of subjective mental state (are they talking about their feelings on the day, talk about the perpetrators feelings, how they thought the perpetrator felt)
13. Attribution of perpetrator’s mental state

Question 22

CBCA • Motivation-Related Content

Answer 22

14. Spontaneous corrections
15. Admitting lack of memory
16. Raising doubts about testimony (liars don’t want to draw any attention to their lack of insight)
17. Self-deprecation (blame themselves)
18. Pardoning the perpetrator (they weren’t as bad as they could have been, Stockholm syndrome)

Question 23

CBCA • Offence-Specific Elements

Answer 23

19. Details characteristic of the offence

Question 24

Why Are These CBCA Criteria usually Absent?

Answer 24

1. Lack of imagination in inventing relevant characteristics
2. Do not realise judgements based on these characteristics, so don’t include them
3. Lack knowledge to incorporate certain criteria
4. Difficult to incorporate some criteria
5. Wary of including details in case they forget
6. Wary of including details that can be checked
7. Wary of including certain characteristics in case their stories sound less credible
   From Vrij (2000)

Question 25

What is the Validity Check List

Answer 25

Part of the statement validity assessment (SVA) Finally, to standardise CBCA findings, evaluators consider alternative interpretations: – Psychological characteristics (age, verbal and social skills) – Interview characteristics (types of questioning) (first stage of SVA was an interview with open ended answers) – Motivation to report (how was it originally disclosed? Is this a case when there’s a divorce going on at the time? – Investigative questions (consistency with other evidence)

Question 26

• Vrij (2005) reviewed first 37 experimental and field studies on CBCA What are the pros and cons of field and experimental studies?

Answer 26

• Vrij (2005) reviewed first 37 experimental and field studies on CBCA Field studies: • Statements made by persons in actual cases of alleged sexual abuse • Clear forensic relevance, but difficult to establish ground truth Experimental studies: • Statements of participants who lied or told the truth for the experiment • Easy to establish accuracy of statement, but labs differ from real-life situations

Question 27

Does SVA Work?

Answer 27

Vrij (2005) reviewed first 37 experimental and field studies on CBCA • Criterion 3 received the most support: in 80% of studies truth tellers included more details • Criteria 4 and 6 also received strong support: in 69% of studies truth tellers included more contextual embedding and reproductions of conversation • Cognitive criteria (1-13) received more support than motivational criteria (14-18) In 92% of experimental studies, truth tellers received higher CBCA scores than liars • Trained evaluators often achieve above chance classifications • For experimental studies, Vrij (2005) reported: – Overall accuracy of 55%-90% – Accuracy for truths of 53%-91% – Accuracy for lies of 35%-100% • Truth bias: CBCA is “truth verifying method” not “lie-detection technique”

Question 28

Some Concerns About SVA

Answer 28

* No formal decision rules, profiles for truth or deception, or cut points * Criteria should be given different weight (Sporer) * Different types of lies (from subtle to outright) may yield different levels/kinds of characteristics * SVA assessments are subjective and inter-rater reliability can be low, even after extensive training * CBCA assessments of written statements are time-consuming & even training may not improve accuracy (Akehurst et al., 2004)

Question 29

Accuracy Rates of Professional Lie Catchers (Memon, Vrij, & Bull, 2003)

Answer 29

Nonverbal Behaviour 55% Truth (55% Lie) CBCA 76% Truth (68% Lie)  more a truth verification method Chance rate = 50% Nonverbal behaviour: Average scores of 9 studies (mostly lab studies) • CBCA: Based on 13 lab studies • Recent evidence suggests that higher accuracy rates by combining nonverbal and CBCA

Question 30

A Whole Approach to Detecting Deception

Answer 30

• DePaulo et al. (2003) examined 158(!) cues to deception. Found that: – Liars are less forthcoming – Liars tell less compelling tales – Liars less positive and pleasant – Liars are more tense – Lies include less ordinary imperfections and unusual details • Many behaviours showed no discernible links, or only weak links, to deception • Also, no clear cut points or profile for lies and liars. Out of 158 cues, these are the only ones they found evidence for.

Question 31

What are some reasons why it is difficult to detect deception with Behavioural Indicators of Deception

Answer 31

* Liars do not seem to show signs of nervousness such as gaze aversion & fidgeting * Professional lie detectors’ ability to accurately classify truth and lies is about 55% (so we are really not much better than chance. Police officers etc. trained have a bit better, 68% truth accuracy) * Analyses of nonverbal behaviour are not accepted as evidence in criminal courts.

Question 32

Difficulties in Detection of Deception (9)

Answer 32

1. Lie detection is difficult and there is no give-away cue 2. Othello Error: Truth tellers may show similar behaviour to liars because they, too, may experience emotions, may have to think hard, or may have to control themselves. 3. Adequate comparisons between truth-telling and lie-telling are not made (e.g., small talk vs. interrogation). 4. Observers seem to have incorrect beliefs about how liars behave and people, including police officers, are taught wrong cues 5. Liars can use countermeasures (e.g., can train themselves to beat techniques). 6. Deception research is often conducted in university labs and the stakes aren’t high enough. 7. The Brokaw hazard: Individual differences in emotional expression, vocal and body movement characteristics 8. Individual differences in ability to control: 9. Cultural differences in nonverbal behaviour.