Friends of the Nonverbal Communication Blog, this week we present the paper “Intentional-Deception Detection Based on Facial Muscle Movements in an Interactive Social Context”, by Dong, Z.; Wang, G.; Lu, S.; Dai, L.; Huang, S. and Liu, Y. (2022), in which authors carry out a study using the facial electromyography technique to study the facial muscles that may be associated with deception.
Deception detection has been a social topic of interest throughout human history, in which facial expressions have played a key role.
The face can be used as a clue to interpret the mental activity of a person and, therefore, it may be useful to know if we are facing an honest speech or not.
We have already mentioned that there are two types of facial expressions, depending on their duration: macroexpressions and microexpressions. Macroexpressions are more frequent, last longer and are, therefore, easier to control and suppress. Microexpressions, on the other hand, are brief, subtle and more discreet. They are born from a failed attempt to hide or suppress emotions. Thus, they are believed to be the most reliable clues for detecting lies and dishonesty. Ekman argued, on this subject, that uncontrollable and quick muscle movements in the forehead area could be important clues to detect lies.
One of the techniques with the highest success rate for the study of microexpressions and their relationship with lying is the analysis of facial expressions on video. However, in this type of analysis an algorithm is needed to classify expressions, for instance, using facial action units (AUs), making manual annotation somewhat necessary.
The truth is that the accuracy of humans in deception detection does not usually exceed randomness, based on previous research, reaching just over 50% accuracy. But what happens if we use computerized means? According to experts, accuracy would increase to approximately 70%.
These computerized means, such as the polygraph, normally focus on analyzing physiological responses like facial temperature, pulse, heart rate, blood pressure… It is understood that the liar will suffer moments of emotional stress because he/her will be frightened, nervous and anxious when lying, and that is precisely what these indicators relate to.
The problem is that innocent people who tell the truth may also be frightened and nervous in a situation where their honesty is being judged. Therefore, this method is not completely reliable.
Other lie detection studies have relied on brain imaging techniques, such as electroencephalography, and have had very positive results. However, this method requires the use of inaccessible sensors and machinery.
In recent years, facial electromyography has been proposed as a method to investigate facial muscle movements and their association with lying. It has had positive results, although its usefulness in this field needs to be further explored.
This is precisely the method of analysis tested by the authors of the article, who carried out an experiment with 22 volunteers who were divided into pairs.
The activity consisted of a role-play: one of the two people would be the informant and the other would be the detective. The detective would ask a series of simple autobiographical questions, and others more extensive about personal preferences. In a second stage of the experiment, the roles were exchanged and finally, in a third stage, the subjects tried to recognize who was lying and when. While this was going on, the participants underwent facial electromyography.
Authors obtained several interesting insights. First, it looks like humans tend to use the zygomatic muscle for expressions associated with positive emotions and to hide emotions while lying.
On the other hand, the corrugator muscle was associated with expressions related to negative emotions (for example, frowning).
Besides, most interestingly, it seems that those who lied in this experiment had a higher activity of the zygomatic muscle.
Therefore, since this muscle is associated with expressions of positive emotions, the authors infer that the liars may be experiencing some joy due to having succeeded in their lie.
One idea that would support this proposal of the authors is the Duchenne smile, which, with zygomatic muscle activation, is an indicator of happy emotions.
Both the zygomatic muscle and the corrugator muscle are located in the upper facial area and the muscles in this area are subject to less volitional control by the motor cortex that is responsible for their movement. As a result, some researchers believe that when people lie, their upper facial muscles filter out emotions more easily. This idea would be supported by the results of the authors’ experiment, which, while inviting further research on the subject, yield revealing information that should be taken into account.
If you want to know more about nonverbal behavior and how it affects personal relationships, visit our Master of Science in Nonverbal and Deceptive Behavior, which you can take in English or Spanish, with special grants for readers of the Nonverbal Communication Blog.