Detecting Deception (wk 10)

Question 1

What is deception?

Answer 1

“A successful or unsuccessful deliberate attempt, without forewarning, to create in another individual a belief which the communicator knows to be untrue.”

Important points
-Doesn’t need to be successful
-Has been done with intent i.e. the liar knows that the information they are providing is untrue

Question 2

Why Do We Lie?

Answer 2

-Self-Oriented Lies: Lies told to protect or enhance the liars psychologically, or to advantage or protect the liars’ interests. (usually what we are dealing with in the criminal justice system)

-Other-Oriented Lies: Lies told to protect or enhance another person psychologically or advantage or protect the interests of others.

Question 3

Means of detecting deception

Answer 3

-Non-verbal
-Verbal
-Physiological

Note: difficult in the criminal justice system as dealing with individuals largely that the interviewer does not know- gap in knowledge about usual mannerisms/ no baseline)

Question 4

Non-Verbal Cues to Deception

Answer 4

-Most striking finding is that a typical nonverbal response during deception does not exist i.e. there is nothing that happens exclusively when you are lying and not when you aren’t.

Some nonverbal behaviours are likely to increase/decrease during deception due to 3 processes:
>Emotion
>Cognitive Load/Content Complexity
>Attempted Behavioural Control

These 3 processes then effect body language and we can likely pick up on it.

Question 5

Emotion

Answer 5

-A person’s emotional state is likely to influence their nonverbal behaviour

-The 3 most common types of emotions associated with lying are fear, guilt, and ‘duping delight’

Note: duping delight refers to lying giving you joy

Question 6

Cognitive Load

Answer 6

-Telling a convincing lie is not easy!

-Liars need to think of plausible answers, avoid contradictions, tell a lie that is consistent with the interviewer’s present and future knowledge,
remember what they’ve said…..

-What happens to your body language when you are required to think really hard about something? Tend to freeze/ move around less: therefore stereotypes about liars fidgeting a lot are not true according to cognitive load literature at least

Question 7

Attempted behavioural control

Answer 7

-Liars engage in “impression management” to avoid being caught

-This is likely to lead to “overcontrol” (can come across a bit odd)

Question 8

The Verdict
DePaulo et al., 2003; Vrij, 2000

Answer 8

• Meta analyses

4 behaviours are more likely to occur when lying (change in rate not absolute there/not) than when telling the truth:
-Increase in Voice pitch
-Increase in Speech errors
-Decrease in Illustrators
-Decrease in Hand/Finger movements

Note: need to know the liars norm to be able to relate and compare accurately and reach the correct verdict.

-Combined accuracy data from available reviews and studies:
based on non-verbal cues could spot truth tellers 55% of the time and liars 55% of the time.
Not every good (only just above chance)

Question 9

Reasons for Failure (in detecting deception)
1. People hold false beliefs about indicators of deceit

Answer 9

Vrij (2000)
-Examined people’s beliefs regarding deceptive body language
-Most cited increases in gaze aversion, fidgeting, and latency periods (latency: delay before answering questions)
-None of these have been shown to increase in liars

Vrij, Semin, & Bull (1996)
-Examined people’s behaviour when they were lying
-Then asked them how they thought they behaved
-Responses were consistent with stereotypes of liars, not actual behaviour
-People are poor judges of their own behaviour let alone others!

Question 10

Reasons for Failure (in detecting deception)
2. People can be taught the wrong cues to deceit

Answer 10

Kassin & Fong (1999)
* Half of the observer participants were trained on Inbau et al.’s nonverbal cues to deceit
* This group of participants actually performed worse than controls when
asked to detect lies

Question 11

Reasons for Failure (in detecting deception)
3. Empirical research may send us mixed messages

Answer 11

Emotion increases blinking rate
Cognitive load decreases blinking rate

So will liars exhibit an increase or a decrease in blinking?

Question 12

Reasons for Failure (in detecting deception)
4. Inadequate Comparisons

Answer 12

• What constitutes “more” or “less” of a behaviour?
• Obvious that the comparison should be a within subjects comparison
• But when should the “baseline” measurement be taken?

Question 13

List the four reasons for failure in detecting deception

Answer 13

1. People hold false beliefs about indicators of deceit
2. People can be taught the wrong cues to deceit
3. Empirical research may send us mixed messages
4. Inadequate Comparisons

Question 14

Verbal Cues to Deception

Answer 14

-Research = less extensive than for nonverbal cues

Most research has examined the use of Statement Validity Assessment (SVA):
- Structured Memory Interview
-Interview Coded
-Evaluation of Outcome (is the person lying or not?)

-CBCA (Criterion-Based Content Analysis) was originally used for assessing the validity of child sexual abuse statements

-Undeutsch Hypothesis –statements derived from memory are qualitatively and quantitatively different than those derived from invention or fantasy

-19 criteria used in credibility assessment

Question 15

How accurate is CBCA (Criterion-Based Content Analysis) i.e. using verbal cues to detect deception?

Answer 15

Memon, Vrji & Bull (2003)
-Truth tellers identified 76% of the time
-Liars identified 68% of the time

Question 16

Reason for failure for using verbal cues (CBCA scores) to detect deception?

Answer 16

CBCA scores may be variable due to factors other than deception:
-Increase Age
-Interviewer’s Style- increase in open ended questions
-Increase verbal and Social Skills

Question 17

Physiological Cues to Deception

Answer 17

Polygraph:
-A scientific measuring device that can display a direct and valid representation of various types of bodily activity

-Most commonly measured activities are sweating of the fingers, blood pressure, and respiration

-Used in many countries (not including NZ) as part of forensic investigations

-Highly controversial (are really good at measuring physiological responses but can we always infer lying from this?)

Question 18

The control question test. How good is it?

Answer 18

-The CQT compares responses to relevant questions with responses to control questions

e.g., “On March 4, did you murder Mr Jones?”
(relevant question = question to which only guilty suspect lies)

e.g., “Have you ever hurt someone in order to get revenge?” (control question = question to which all examinees are expected to lie)

-CQT can tell truth tellers 72% of the time and liars 87% of the time

Question 19

CQT: reasons for failure

Answer 19

1. False positives are likely when innocent suspects are nervous
2. Countermeasures:
   * When suspects try to influence polygraph outcomes in order to “pass”
   * Attempts to increase arousal (e.g. foot tensing and increasing cognitive load) are more successful

Question 20

Countermeasures success in the control question test

Answer 20

Control questions will freak truth tellers out because they have done them. Higher rate of physiological l arousal to control as opposed to relevant.

Liars show the opposed pattern. Not as high arousal for control, more for relevant

Liars using counter measures (try and increase arousal artificially for control questions) show the same trend line as truth tellers i.e. they have greater arousal for control questions as opposed to relevant questions

Specific point doesn’t matter, the slope matters i.e. everyone have a different baseline physiological response

Question 21

Honts, Raskin, & Kircher (1994)

Answer 21

-Simulated theft

-Three groups of participants:
Innocent
Guilty Untrained
Guilty Trained in Countermeasures (when you spot control questions, increase arousal)

-All participants subjected to a CQT polygraph

-Although innocent suspects still came out on top, training resulted in increased performance in the CQT test

-Compared to untrained participants, over twice as many trained participants passed the test

-Compared to untrained participants, just under half as many trained participants failed the test

Question 22

The guilty knowledge test + how does it compare to the CQT

Answer 22

-The GKT examines whether suspects possess knowledge that they do not want to reveal about a particular crime

For example:
-If murder weapon was knife…
innocent and guilty suspects will deny the murder
only the guilty suspect will recognise the knife

-Lykken (1998) described how the GKT could have been used in the OJ Simpson case

-When compared to the CQT, the control question test identifies more liars, but the guilty knowledge test identifies more truth tellers

Question 23

The guilty knowledge test: reasons for failure

Answer 23

1. Limited Applicability
   * The test designer needs to know the answer
   * Innocent suspects should not know the answer (could know details about case through media- does not mean you committed the crime)
   * The guilty examinee needs to know the answers (not necessarily going to remember everything about what they did e.g. may not know what they left at the scene)
   * Podlesny (1995) – Only 9% of CQT cases could have used the GKT
   * Suspect could have guilty knowledge but deny guilt

Question 24

General problems with lie detection: there is no Pinocchio test…

Answer 24

Although some behaviours are more (or less) likely to occur when a person is lying, there is no reliable giveaway clue that lie catchers can use

Question 25

General problems with lie detection: Research difficulties

Answer 25

Research can be divided into two categories:

Field Research Laboratory Research
-Ground truth problems
-Serious crimes
-High stakes

Laboratory research
-No ground truth problems
-Non-serious crimes
-Generally lower stakes

Straddling the boundary between research and practice is very difficult

Question 26

General problems with lie detection: Othello error

Answer 26

-Liars and truth tellers may experience the same processes (emotion, cognitive load, and attempted behavioural control), so they might respond similarly to measures aimed at detecting deception

-e.g., should fear be interpreted as a sign of guilt or fear of being wrongly accused?

-Verbal tests are the least likely to be affected

Question 27

General problems with lie detection: false beliefs

Answer 27

“A successful or unsuccessful deliberate attempt, without forewarning, to create in another individual a belief which the communicator knows to be
untrue”

-An examinee who has a false belief is unlikely to display physiological or non-verbal signs of deception

-Verbal tests could be the most promising of the three techniques discussed so far

Question 28

List the four general problems with lie detection as identified in this lecture

Answer 28

-There is no Pinocchio test
-Research difficulties
-Othello error
-False beliefs

Question 29

The New Wave of Lie Detection Research: Targeted Interviewing

Answer 29

-Aims to elicit diagnostic cues to deception, rather than just assuming that they will arise (interview people in a way that dispproprionatley disadvantages liars)

-Example 1:
Assumption = deception is more demanding than truth telling
Solution = place suspects under additional cognitive load
e.g., ask them to tell their story in reverse order
e.g., make them maintain eye contact

-Example 2:
Assumption = liars often prepare themselves for an interview
Solution 1 = ask suspects unanticipated questions
Solution 2 = ask anticipated questions in an unanticipated format (draw, sketch: won’t have prepared for that)

Question 30

A new wave of lie detection research: Brain scanning

Answer 30

-Rather than looking at measurements of autonomic activity, this approach measures blood flow to various parts of the brain using fMRI

-Hypothesis = brain activity when lying could differ from brain activity when telling the truth

-Two research groups have devised experimental paradigms and statistical methods that, they claim, allow identification of neuronal activity patterns consistent with lying

-Two main areas of increased activation when lying, relative to when telling the truth:
1. Prefrontal cortex
2. Anterior cingulate gyrus

These regions may be recruited for inhibiting a prepotent response (i.e., giving a true answer) = holding back the truth

-This has been proposed as one of the major cognitive differences between lying and telling the truth

-Findings have resulted in two commercial enterprises, launched in 2006, each with the goal of bringing these techniques to the public for use in:
- Legal proceedings
– Employment scrutiny
– National security investigations

Question 31

Reasons for caution when comes to using brain scanning in lie detection…

Answer 31

• Generalisability:
  – Population generalizability
  – Being instructed to lie vs. free choice
  – The stakes
• Specificity of measurement:
  – Suppression could be associated with things other than lying
• The ‘CSI Effect’ (sciency evidence is impressive and very persuasive to jurors)
• Ethical concerns
  – Premature adoption of a scientifically immature technology
  – Cognitive Liberty – the right to privacy of one’s thoughts
• Legal Concerns
  – Admissibility
  – Does fMRI constitute a search?
  – Right to avoid self-incrimination