Week Seven: Detecting Deception and Careers-related Information

Forensic Psychology Pathways

Slides detailing pathways into forensic psychology are available in the lecture seven folder. They were produced by Geraldine and others at the BPS Careers Fair.
Forensic psychology is a broad field with various pathways, the slides detail one common pathway to becoming a practitioner.

Detecting Deception

The lecture will cover detecting deception. Topics covered previously (predicting recidivism, eyewitness identification, investigative interviewing) are distinct but have overlap.
Signal detection theory is a useful theoretical framework applicable to many psychology tasks.
Second discussion board assignment is due Sunday at midnight.

Signal Detection Theory

Signal detection theory can be applied to tasks where one must distinguish one class of items from another.
In lie detection, the strength of evidence becomes the "lying signal".
Weak signals (left of continuum) indicate truthfulness; strong signals (right) indicate suspiciousness and lying.
The left distribution comes from truth tellers, while the right comes from liars.
Liars provide stronger evidence of lying than truth tellers.
The distance between the two distributions is D prime, which measures the effectiveness of a lie detection system.
A good system has a large D prime, while a poor one has a D prime close to zero, with overlapping distributions.

How Good Are People at Spotting Lies?

Early study (1990s) by Ekman and O'Sullivan examined professionals (secret service agents, federal calligraphers, robbery investigators, judges, psychiatrists) and a university sample.
Participants watched 10 video clips and determined if the person was lying or not (5 liars, 5 truth tellers).
Mean scores represent proportion correct; this can be converted to D prime.
Only Secret Service agents performed significantly above chance (64% correct, D prime of 0.72).
Other groups had mean scores close to 50% (chance), with D prime close to zero.
Bond and DePaulo replicated the results with a larger group, finding a D prime of 0.2.

Cues People Look For

Participants from 58 countries were asked how they can tell when someone is lying.
Key indicators: averting eye gaze, appearing nervous, fidgeting, incoherent speech/mumbling, facial expressions, and inconsistencies in their verbal account.
People expect liars to avoid eye contact, change posture, scratch their head, tell longer stories, etc.
Research suggests the opposite is often true.

Are These Cues Accurate?

Meta-analysis reveals little to no association between common cues and lying.
Liars may appear more nervous/tense or less certain under some conditions, provide fewer details.
There are no consistent, reliable indicators of deception.

Arousal-Based vs. Cognition-Based Theories

Arousal-based theory: Lying is accompanied by more emotional arousal.
Liars try to hide their arousal via indicators like increased heart rate, sweat, poor emotion management, leakage of emotion, and faked emotion expressions.
Early evidence supported this theory, but less so recently.
Cognition-based theory: Lying is more demanding/cognitively burdensome than telling the truth.
Liars must concoct a story consistent with known facts, detailed enough to be based on experience, and simple enough to remember.
Indications of cognitive load may correlate with lying.

Methods for Detecting Deception

Methods focus on measuring physiological responses (polygraph, neuroimaging), nonverbal behavior, or verbal behavior.
Verbal behavior (content of the lie) is potentially the most promising method.
Early lie detection tools were based on the idea that lying is physiologically arousing. Later, tools were built to maximize cognitive load.
In the early 1990s, neuroimaging techniques were developed to look for neural patterns correlating with lying.
From the 2000s to now, the focus has shifted to verbal behavior.

Limitations

Experimental studies are low stakes; participants are instructed to play along.
In the real world, it's often uncertain whether the person is guilty or not.

Polygraph

Measures heart rate, blood pressure, respiration rate, skin conductance.
Two types of tests: controlled question test (CQT) and concealed information test (CIT).

Controlled Question Test (CQT)

Uses relevant, irrelevant, and controlled questions.
Detects jumps in physiological measures in response to relevant questions.

Concealed Information Test (CIT)

Relies on information known only to the guilty party.
Uses multiple-choice questions where only one item is relevant.
Measures responses to each question.

Problems with the CQT

Detects emotional arousal, not necessarily lying.
No convincing basis for assuming only guilty participants will show more arousal for relevant questions.
Relevant and control questions can differ in ways that increase physiological arousal in both guilty and innocent suspects.
D prime values range from 0.87 to 2.18.

Problems with the CIT

Relies on the guilty person knowing what they stole.
The guilty person may not have encoded the information or may have forgotten it.
D prime is similar for CIT and CQT.

fMRI

Some tests are used in conjunction with fMRI.
fMRI looks at general cognitive functions involved in various tasks.
No unique brain area is activated only when lying.
There is no unique pattern of brain activity specific to lying.
Physiological approaches suffer from the fundamental problem that there is no "Pinocchio's nose".
Physiological responses can occur in innocent people for various reasons.

Nonverbal Behavior

Most nonverbal cues have no significant relationship with lying (DiPaulo et al.).
Widely believed stereotypes of lying don't seem to pan out.

Factors with Significant Relationship

Fidgeting (positive correlation): More fidgeting indicates more lying.
Illustrators (negative correlation): Increase in gestures means a decrease in lying, suggesting liars are more stiff.
Liars may fidget more but are more stiff with their hands and arm movements.

Nervousness

Lie tellers may feel more nervous, but truth tellers can also feel nervous.
Both lie tellers and truth tellers are motivated to hide their nervousness.
Observable differences in nervous behavior are likely to be minimal or inconsistent.

Eye Contact

People who are being deceptive may show more deliberate eye contact.
Study: Half of participants were instructed to lie about the purpose of their trip.
No difference in gaze aversion, but liars showed a slight effect for more deliberate eye contact.

Reasons for Deliberate Eye Contact

Liars might use it to appear more convincing.
Liars might use it to actively monitor how receptive their lie is.

Microexpressions

Microexpressions happen very rarely.
They are not a clear indicator of lying.
They appear in both liars and truth tellers.

Summary of Nonverbal Cues

Nervousness and microexpressions are not reliable indicators of deception.
Gaze aversion is also not a reliable indicator.
Deliberate and exaggerated eye contact might be.

Verbal Behavior

Innocent truth tellers usually adopt a "tell it all" strategy, giving as many details as they can remember.
Liars try to keep it simple to remember the story for future recounting.
Innocent people like to tell it all, while liars try to keep it as simple as possible with minimal detail.
Liars want to avoid giving false details that can be checked.

Details Provided

Liars provide details that cannot be verified by an independent party.
Truthful stories often include more details than deceptive stories.
Truth tellers are more likely to provide additional details if asked.

Mock Crime Study

Someone took a wallet from a briefcase, others simply moved the briefcase.
Both groups tried to convince interrogators they did not take the wallet.
Liars reported using five verbal strategies to appear innocent: making their story detailed, avoiding lying where possible, being consistent, claiming to have an alibi, and trying to make their story seem unrehearsed.
Truth tellers reported only two strategies: telling the truth exactly as it happened and being cooperative.

Strategic Use of Evidence

If a suspect does not know what evidence the interviewer has, a truthful account will be more consistent with available evidence than a deceptive one.
Truth tellers will provide more detailed accounts than liars.
Liars are more likely to change their stories if later made aware of the evidence.

Study Outcome

Half of interrogators received training in strategic use of evidence; half had not.
All interrogators were provided with the same three pieces of evidence.
Deception detection accuracy was good when interrogators had strategic use of evidence training. If not, D prime was close to chance.

Liar's Dilemma

Liars want to provide lots of details because detailed accounts are more likely to be believed, but they don't want to provide false details that could be checked.
Liars will have a greater proportion of information consist of details that cannot be checked or verified.

Aiding Credibility Assessment

Encourage suspects to say more and ask unexpected questions.
Liars will provide less new information than truth tellers.
Deceptive suspects will be less able to cope with additional cognitive demands.

Asymmetric Information Management (AIM) Technique

Encourage true tellers and liars to adopt different verbal strategies.
Suspects are informed that more detailed statements are easier to assess as truthful or deceptive.
This should encourage liars to withhold details and truth tellers to be more forthcoming.

AIM Technique Result

In the control condition, there isn't much difference in the number of details provided.
When using AIM, true tellers respond by providing as much detail as possible, while liars respond by providing as little detail as possible.

Summary

Verbal cues are most important in detecting deception.
Looking at the content of the lie, the number of details discussed, and the type of detail discussed are promising indications of whether someone is lying or telling the truth.
Physiological indicators and polygraph tests have countermeasures and correlate with other things that have nothing to do with lying.
Looking at the content of what the suspect is saying is the path forward.