Through electrical measurements of the skin and three key facial muscles, researchers in Portugal have shown that we respond differently to different strong emotions expressed by laughing or crying — and the results suggest that our bodies can distinguish fake laughter, but not fake tears, from real emotions.

The research team studied how a person's facial muscles responded when they listened to audio recordings of laughter and crying that were either authentic or fake, as detailed in a paper published May 15 in Cortex. The participants could tell the difference between fake emotions and real ones. They also displayed stronger facial responses when they listened to recordings of laughter compared with when they listened to crying. "In fact, responses to crying were not statistically different from baseline," the authors noted in the paper.

César F. Lima, first author of the paper and an assistant professor of psychology at Instituto Universitário de Lisboa, in Portugal, told The Academic Times that his previous work involved studying whether the perception of vocal expressions such as laughter and tears uses the same brain systems that are used to produce those expressions. Lima has sought to understand whether our motor systems play a role in our auditory perception.

"Understanding how the mind and brain perceive the emotional information around us is a fundamental question in psychology and neuroscience, and one that is not settled," Lima said. "There is a long-standing debate over the possibility that, when we see an emotional expression, we 'read' it by simulation — we engage the body and brain mechanisms involved in producing the expression, and that helps us infer what the others are feeling." But there hasn't been much of a consensus, he said. 

"Knowing more about these issues will help us better understand how emotion and communication work. But it also [may] be useful from an applied perspective, because difficulties with emotions, and emotion recognition, are a feature of many neurological and psychiatric disorders," Lima continued.

He explained that he and his co-researchers wanted to use electromyography, a technique that measures electrical activity produced by skeletal muscles and is not typically used to study vocal emotional processing, to explore the involvement and role of sensorimotor systems while a person is perceiving nonverbal vocalizations. In this case, laughing and crying were the nonverbal vocalizations. Electromyography differs from other tests, such as functional magnetic resonance imaging, because it can measure facial-muscle activity.  

A sample of 100 adult listeners completed two tasks while their facial muscles and skin conductance were tracked. During the first task, they listened to audio recordings of men and women speakers laughing or crying. Some of the expressed emotions were genuine; the speakers watched funny videos or recalled sad memories to elicit genuine responses. Others, which the paper described as "posed laughter and crying," were fake, produced by the speakers without any external stimuli. The listeners were not told that some expressions were authentic while others were posed. 

During the second task, the listeners heard the same recordings of real and fake emotions. This time, though, they were also asked to rate how authentic the emotions sounded.

The researchers tracked the listeners' facial muscles via electromyography, which involved the placement of electrodes on participants' faces. Their electrodermal activity, which refers to the electrical characteristics of the skin, was measured from sensors attached to listeners' index and middle fingers on their nondominant hands.

The electromyography test focused on the zygomaticus muscle, which pulls up the corners of the upper lip; the orbicularis muscle, which closes the eyes; and the corrugator, which lowers and furrows the eyebrows, according to the paper. The authors noted that "the corrugator is typically associated with negative emotions, such as anger and sadness, while the zygomaticus and orbicularis are involved in smiling and are associated with positive emotions." By elevating the cheeks, the orbicularis produces the laugh lines that are a hallmark of authentic delight in what neurologists call the Duchenne smile.

"We expected facial muscle responses to be stronger for laughter than for crying, based on evidence that laughter is particularly contagious ... and that positive emotions preferentially engage sensorimotor systems," the authors said. 

On average, authentic and posed laughs were perceived as 0.6 points more authentic than both authentic and posed cries, the authors said. But within specific kinds of vocal expression, participants performed well — they rated authentic laughs and cries as 2.2 points more authentic than their respective posed equivalents. 

"As predicted, muscle responses were stronger for laughter compared to crying. During listening to laughter, there was a selective activation of the zygomaticus and orbicularis," the authors said. "Consistent with our hypothesis, this activation was higher for authentic compared to posed laughs."

Skin conductance was studied in the participants through their fingers. Measurable changes were more often elicited by laughs than by cries, and by authentic rather than posed laughs. 

There was also an association between facial-muscle activity and how the participants rated the authenticity of laughter. Activity in the orbicularis muscle around the eyes "was related to perceptions of authentic affect, regardless of whether laughs were authentic or posed, and corrugator activity was related to perceptions of posed affect," the authors reported.

Overall, and in comparison with crying, laughter evoked stronger facial responses and was more reliably associated with participant behavior across all conditions. 

"Facial responses were much weaker for crying compared to laughter in this study," Lima said. "One thing we will need to understand better in the future is whether the putative role of sensorimotor mechanisms in vocal emotional processing is a general one, or whether it varies across emotions, task conditions, or other factors."

"These findings indicate that emotional authenticity can affect peripheral nervous system responses to vocalizations. They also point to a role of sensorimotor responses when we evaluate whether others are being authentic or not," which, until now, has been mostly informed by studies on facial expressions, Lima said.     

He explained that laughter is a "fundamentally social expression" that is common and contagious to others, while crying reflects a negative, less frequent emotion. 

"We often want to share laughter with others — but when someone is crying, we might want to support them, not necessarily to cry along. These differences might explain why we saw much stronger facial reactions when participants heard laughter compared to crying," Lima said.

"However, this does not mean that we are not able to detect authenticity in crying; our behavioral explicit task shows that we can," he added. "The process by which we do that might be slightly distinct, though, and this is something we will need to explore in future work."

The study, "Authentic and posed emotional vocalizations trigger distinct facial responses," published May 15 in Cortex, was authored by César F. Lima, Instituto Universitário de Lisboa and University College London; Patrícia Arriaga, Ana Rita Pires, Sofia Frade and Leonor Neves, Instituto Universitário de Lisboa; Andrey Anikin, University of Lyon/Saint-Etienne and Lund University; and Sophie K. Scott, University College London.