PubMed · 15433920
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
J V WILKINSON. 1950. Appearances can be deceptive.. https://pubmed.ncbi.nlm.nih.gov/15433920/
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
M. O'Sullivan and P. Ekman (2004) claim to have discovered 29 wizards of deception detection. The present commentary offers a statistical critique of the evidence for this claim. Analyses reveal that chance can explain results that the authors attribute to wizardry. Thus, by the usual statistical logic of psychological research, O'Sullivan and Ekman's claims about wizardry are gratuitous. Even so, there may be individuals whose wizardry remains to be uncovered. Thus, the commentary outlines forms of evidence that are (and are not) capable of diagnosing lie detection wizardry.
Bond and Uysal (this issue) complain that expert lie detectors identified by O'Sullivan and Ekman (2004) are statistical flukes. They ignore one class of experts we have identified and misrepresent the procedures we use to identify the others. They also question the psychometric validity of the measures and protocol used. Many of their points are addressed in the chapter they criticize. The fruitfulness of the O'Sullivan-Ekman protocol is illustrated with respect to improved identification of expert lie detectors, as well as a replicated pattern of errors made by experts from different professional groups. The statistical arguments offered confuse the theoretical use of the binomial with the empirical use of the normal distribution. Data are provided that may clarify this distinction.
Explore the source record for details and available documents.