The Truth About Microexpressions


If you’ve heard of the show “Lie to Me” on the Fox network, you probably are aware of microexpressions. Claimed to be a direct link to one’s internal emotional experiences, these brief facial and involuntary expressions appear and disappear in as quickly as 1/25th of a second and have been tied to one of seven basic emotional states including:

Happiness or Joy
Disgust
Contempt
Sadness
Anger
Fear
Surprise

Although discussed most frequently by psychologist Dr. Paul Ekman (formerly of the University of California – San Francisco), several researchers have jumped on the bandwagon and have offered training or consulting in the area of microexpression detection. Some popular tools for training include the METT (Micro Expression Training Tool) and SETT (Subtle Expression Training Tool) available online through Dr. Ekman’s site, as well as the MiX™ and SubX™ product line available through Dr. David Matsumoto's Humintell site. These microexpressions have become the foundation for several programs, including the SPOT (Screening of Passengers by Observation Techniques) program used by Transportation Security Administration (TSA) officers at over 150 airports across the United States.

Although microexpressions have been discussed by Dr. Ekman and his colleagues since the early 90’s, relatively few studies have discussed the utility of microexpressions, and some studies have even questioned the application of microexpressions to real world applications. In a 2008 study published in Psychological Science, the authors noted “To our knowledge, no published empirical research has established the validity of microexpressions, let alone their frequency during falsification of emotion” (p. 509). In this study, participants were shown a variety of positive, negative, and neutral images. Researchers analyzed over 100,000 video frames of participants who reacted to the image in a genuine or deceptive manner. Of all 697 expressions identified, no microexpressions that involved both the upper and lower face were found. And, even when considering partial microexpressions, only 14 were identified (2% of all expressions). As a result, the rarity of this factor in the identification of emotion may pose a problem for real world applications. A 2010 study published in the journal Communication Research assigned 108 participants to view an episode of Lie to Me, Numb3rs, or to not view any program. Not only were individuals in the Lie to Me group the least accurate in detecting deception, but they were more likely to classify truth-tellers as liars.

A few takeaways you should remember as you further develop your ability to evaluate others. One, simply because the ability to detect microexpressions in a controlled task is related to the ability to detect deceit in a separate task, does not imply that this skill is being accessed in the deception task by participants. Studies like this one appear to imply just that. Critically evaluate not only the experimental design and analyses, but whether results can be used to practically outside the laboratory. Better yet, do you have an idea for a good study? Send in your ideas. Perhaps you could be the next expert.

Adam B. Troy, Ph.D.
Founder and President
Behavioral Research Group, LLC

This entry was posted in News. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

Upload Files

You can include images or files with your comment by selecting them below. Once you select a file, it will be uploaded and a link will be added to your comment. More than one file may be selected if appropriate.