45 DECEPTION DETECTION
Group Decision and Negotiation 13: 45–59, 2004
© 2004 Kluwer Academic Publishers. Printed in the Netherlands
Individual and Small Group Accuracy in Judging
Truthful and Deceptive Communication
MARK G. FRANK AND NICOLE PAOLANTONIO
Rutgers, The State University of New Jersey, USA (E-mail: mgfrank@scils.Rutgers.edu)
THOMAS H. FEELEY AND TIMOTHY J. SERVOSS
State University of New York at Buffalo, NY, USA
Abstract
We examined accuracy in detecting the truths and lies of 10 videotaped students who offered their opinions on
the death penalty or smoking in public. Student lie detectors were randomly assigned to either the individual
condition, where they reported their veracity judgments and confidence independently, or the small group con-
dition, where they recorded their judgments privately and then deliberated with 5 other students before making
a consensus judgment of lie, truth, or hung. Results indicated that small group judgments were more accurate
than individual judgments when judging deceptive but not truthful communication. Small group individuals also
reported greater confidence in their abilities after the task. Finally, groups with a greater number of hung judg-
ments were more accurate, likely due to their employing hung judgments for the most difficult to judge stimulus
communicators. These results raise implications for real life group judgments, particularly in light of the increasing
availability of technology.
An interesting contradiction exists between laypersons’ confidence in their ability to catch
liars and the mounting number of studies in Social Psychology and Communication that
clearly indicates evidence to the contrary. Consider that the Supreme Court clearly assumes
that people, in the form of trial juries, can detect the lies of a witness. In Mattox v. United
States (1895), the Court wrote, “. . .the accused has an opportunity, not only of testing the
recollection and sifting the conscience of the witness, but of compelling him to stand face
to face with the jury in order that they may look at him and judge by his demeanor upon
the stand and the manner in which he gives his testimony whether he is worthy of belief”
(cited in Coy v. Iowa , 1988, p. 1026).
Unfortunately, over four decades of social science research yield one conclusion: Hu-
mans are unable to identify deception when it is present in human communication at rates
much greater than chance (for reviews see Feeley and Young 1998; Kalbfleisch 1992; Kraut
1980). Experiments typically produce deception detection accuracy rates in the 55–58%
range. It is important to note that many of these studies employ stimulus materials that should
serve to inflate lie detection rates. Take for example, Feeley and deTurck (1995, 1997a)
who raised the suspicion of student judges in two ways. First, their studies are advertised
as lie detection studies, which may draw for those who believe themselves good lie catches,
and second, the experimenter artificially raises the suspicion of students by alerting them
that deception may be present in one or more of the videotaped messages, whereas in the
real world this possibility is typically not telegraphed (e.g., Park, Levine, McCornack,
Morrison, and Ferrara 2002).