Coastlines Online, UC Santa Barbara Alumni Association

Departments

Winter 2017

  • around storke tower
  • authors
  • arts
  • research
  • milestones
  • Features

The Detectable Art of Deception

By Ariel Reed '17


Dr.Duncan Mellichamp

The use of deception has been an art that has been mastered by many and has evolved throughout the ages. Various types of species utilize forms of deception for the purpose of survival. For example, chameleons change colors to blend with their surroundings in order to avoid predators. Likewise, phasmids or “walking-sticks” resemble twigs as a defense mechanism for survival. Many argue that deception first began within the human species for this very reason: survival.


If basic survival was the initial purpose of deception, then why does it exist today in various complex forms? One might argue that survival is still the premise for which it still exists: social survival, financial survival, or the survival of a relationship. Dr. Norah Dunbar, Chair of the Department of Communication at UCSB, has continually conducted research in regards to modern day deception.

Traveling to Arizona, Dr. Dunbar orchestrated research with Judee Burgoon, Director of Human Communication Research at the University of Arizona and world-famous researcher in non-verbal communication. “I had a large grant a few years ago and I included her as one of the co-PI’s and her research team at Arizona,” says Dr. Dunbar. “So we still do a lot of collaborative work together.”

Dr. Norah Dunbar

How exactly does one go about detecting deception? According to Dr. Dunbar, non-verbal cues have been the hot topic of study since the early 90’s. This method of research focused more on body language, rather than the actual words one uses. With the passage of time, however, this method of observation has proven to be unreliable. Researchers realized that individuals can mask their facial and body expressions very well. Thus in the stream of research that Dr. Dunbar does today, she focuses on verbal analysis, which includes what words one uses, what tense one speaks in, the utilization of first person pronouns, and a nimiety of other cues that exist within verbalization.

Specifically focusing on print, Dr. Dunbar and other linguistic analysis began studying text messages and emails in order to determine what types of linguistic patterns arise in deceptive print. Researchers began implementing a program called L.I.W.C. (short for Linguistic Inquiry Word Count), developed by James Pentabaker at the University of Texas. In the analysis of a written transcript, it identifies how many first person pronouns, emotion words, or other words similar to this nature are used. One might question the reason as to why studying print is more useful than face-to-face contact. The reason for this is because it is more concrete. Researchers find that in text based email, people are less likely to lie because there is a physical copy of it. In contrast, one is more likely to find deception in face-to-face interaction because there is no physical proof of what one said. “You can always deny later that you said it,” says Dr. Dunbar.

Where, then, is the practicality in analyzing all this data? The information obtained from this research could be useful in detecting scams that swindle hundreds and thousands of dollars from individuals and companies that fall victim to email scams. With the assistance of this research, people will be able to predict when fraud is going to happen. In a study conducted on Enron emails researchers looked at the emails to detect the difference between those that were honest and those that were deceptive. In a similar study conducted by Jeff Hancock at Stanford University, researchers also looked at fraudulent journal articles in order to expose certain qualities of deception in print. In this study, he found that articles that made fraudulent statements were either retracted or revised with corrections.

This was seen as especially significant in this past presidential election. Both social media as well as verbal rhetoric on behalf of the candidates were utilized to convey information to the American voters. It is difficult to detect deceit within tweets, status updates, or compact statements given their brevity. Thus in this situation the main method of deception detection simply remains fact-checking. Organizations such as Politifact and Factcheck.org exist primarily for the purpose of fact-checking. Yet, when one comes across speeches and longer statements that contain hesitant promises and unsteady allegations, there really is no sure-fire way to detect whether the statement made is legitimate or not. As a result, the research that Dr. Dunbar and her colleagues conduct find ground in this arena and arenas similar to such.

Collectively, researchers have found a few universal linguistic markers that are characteristic of verbal deception. For example, one speaking in a passive voice or never using first person is common characteristic associated with deceit. Likewise, fabricators often do not take ownership of their story, as a way to avoid consequences. Though these factors do not solely indicate an imposter, it can be enough to raise a red flag and allow for investigation. “There’s no Pinocchio’s nose, unfortunately,” says Dr. Dunbar. But the existence of enough of these patterns can be enough to stir examination of the matter.