PSYCH 3CC3 Lecture Notes - Lecture 3: United States Secret Service, Lie Detection, Fidgeting
Detecting Deception
Lecture 3
• US secret service, FBI, judges, police, psychiatrists, university students, and adults taking
one-day course on deceit asked to judge whether someone was telling truth/lying about
their reactions to videos. Also asked to rate their confidence in general lie-detection
ability
o Only US secret service agents judged better than chance and other groups (64%
accuracy – this is still bad b/c there’s 36% error and we only accept <5%)
o Accuracy uncorrelated w/ self-rating of confidence, except for deceit group
o Shortcomings: earlier studies found no evidence that “experts” better at lie
detection than students. Forensic relevance unclear – forensic lies are not about
currently felt emotions, people will lie about context of crime
• We have higher lie accuracy (saying something is a lie) than truth accuracy. More likely
to say something is a lie rather than true. This is especially true for officers w/ more
experience
o More likely to call a truth-teller a liar, than to call a liar a truth-teller
• Meta-analysis – combines the data of multiple studies and re-analyzes them
• Truth Bias – calling more statements as true than as false b/c we tend to believe in the
good of people. Judges have truth bias
• Higher stake lies (more negative consequences) = more likely to be detected in their lie.
Motivation makes liars easier to detect
• Facial/body cues to deception that aren’t reliable: avoiding eye contact, more
smiling/laughter, high rate of eye blinking, nervous fidgeting, more illustrative gestures,
more body/head movements, more shrugging
• Cues that may be related to deception: less forthcoming & say less, tell less compelling
stories, leave a negative impression, more tense (but don’t fidget), fewer
imperfections/mistakes, contains fewer unusual content
o Experienced more cognitive load (have to use more mental energy while lying;
can’t do other things simultaneously), expressed more negative emotion,
distanced themselves from event, used fewer sensory-perceptual words (see,
touch, hear), referred less to cognitive processes (what they believed), expressed
same amount of certainty as truth-tellers “believe me, I am telling the truth”
• Language of Deception:
o Low verbal immediacy – verbal immediacy is tendency to use present tense to
describe past events as if it’s happening right now. Liars use past tense & passive
voice “it happened” rather than “I did this”. Distances themselves from the event
o Fewer details in story – there will also be fewer unusual details
o Impression of verbal uncertainty
o Impression of nervousness – no indication where it comes from, just a gut feeling
o Lack of logical structure/plausibility in story – story doesn’t make sense &
doesn’t follow a structure, things they say happened are usually unlikely to
happen (but still possible)
o Raised pitch of voice – liars talk in a higher pitch but this differs by individuals.
Unreliable in forensic context
• Word Usage/Grammar and Deception
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Lecture 3: us secret service, fbi, judges, police, psychiatrists, university students, and adults taking one-day course on deceit asked to judge whether someone was telling truth/lying about their reactions to videos. Forensic relevance unclear forensic lies are not about currently felt emotions, people will lie about context of crime: we have higher lie accuracy (saying something is a lie) than truth accuracy. More likely to say something is a lie rather than true. Judges have truth bias: higher stake lies (more negative consequences) = more likely to be detected in their lie. Liars use past tense & passive voice it happened rather than i did this . People often lie about these questions, so this is compared to relevant questions. Innocent people have higher arousal to control q"s while guilty suspects have higher arousal to relevant q"s: procedure, pre-test explain process and hook them up to the machine.