The Neuroscience Of Emoticons

Yes, there is one. 🙂

It’s hard to say for sure when the first emoticon appeared–some see one as far back as 1862, in the transcript of a speech by Abraham Lincoln–but its digital birth is typically dated to September 19, 1982. That’s when computer scientist Scott Fahlman of Carnegie Mellon unceremoniously proposed using the character scheme 🙂 to mark forum messages meant to be taken in jest. Whenever the emoticon was invented, it’s as much a part of modern communication as the colon, hyphen, and parenthesis themselves.


Today emoticons are so pervasive that behavioral science has taken an active interest in how people use them. Among the evidence (recently surveyed by Roni Jacobson at the great new Science of Us blog), we find that women use more emoticons than men, that using emoticons too soon can creep out a new acquaintance, that deploying too many :-(‘s might makes us less popular, and that emoticon configuration varies by culture. While 🙂 indicates a smile to Americans, for instance, ^_^ does so for the Japanese.

Neuroscience has started to embrace emoticons, too–a line of research we humbly suggest using (@):-) to symbolize hence forth. “Emoticons are a part of communication for a great many people,” psychologist Owen Churches of Flinders University, in Australia, tells Co.Design. “Social neuroscientists should study the way we socialize now.” Though in its infancy, such work has already produced an intriguing insight: Our brains might be adapting to an emoticon-filled world by processing them differently.

Take one small study conducted by researchers from Tokyo Denki University in Japan. Test participants placed in a brain imaging machine looked at images of happy and sad faces as well as Japanese emoticons, with (^_^) for happy and (T_T) for sad (the T’s are tears). In both cases, participants showed increased activity in the right inferior frontal gyrus, an emotional processing areas. But only when they viewed facial images did they show an increase in the right fusiform gyrus, the brain’s key face-processing region–not when they viewed emoticons. “Remarkably, emoticons convey emotions without cognition of faces,” the researchers, led by Masahide Yuasa, wrote in a 2006 paper.

Shaded areas indicate the activity areas in (1) right fusiform gyrus (2) right inferior frontal gyrus (3) right middle frontal gyrus, and (4) right inferior parietal lobule.Adapted from Yuasa et al, CHI 2006.

As a follow-up study, Yuasa and company collected brain images from participants reading sentences with emoticons placed at the end. Once again they found strong activity in the emotional area and none in the face-processing center. This time, they also found increased brain activity in Broca’s area, a region involved in text comprehension. In other words, as far as the brain is concerned, emoticons seem to straddle the line between verbal and non-verbal communication.

More recent work, led by Churches, conducted a similar test using a different scientific approach. With electrodes hooked to their scalp, 20 test participants looked at pictures of actual faces and Western-style emoticons–with 🙂 for a smile and 🙁 for a frown. They also looked at these same items in inverted form: an upside-down face, for instance, or an emoticon like (-: facing the wrong way.


The reason for this setup requires some technical explanation. Rather than measuring brain images, Churches and company measured electrical brain responses known as event-related potentials, or ERPs. The brain produces distinct ERPs for various events. When it sees faces, for instance, a reliable negative ERP occurs 170 milliseconds later–known as the N170. When it sees inverted faces, the amplitude of that N170 gets larger (below).

For inverted faces, the N170 wave increases in amplitude, compared to normal faces, in both the left (above, left) and right hemispheres.via Churches et al (2014), Social Neuroscience.

In simple terms, the researchers wanted to know if emoticons, both normal and inverted, produce a similar N170 effect. Sure doesn’t seem like it. While normal emoticons did produce a large N170 signal, inverted emoticons produced a smaller one (below), the researchers report in a 2014 issue of Social Neuroscience. They suspect that flipped emoticons lose their cohesive symbolic meaning–becoming just a loose collection of punctuation.

For inverted emoticons, the N170 wave decreased in amplitude, compared to normal emoticons.via Churches et al (2014), Social Neuroscience.

With so many variables, from emoticon type to the scientific method used, it’s tough to say for sure why the early studies showed no brain activity in the face-processing region while the recent one showed some face-related activity. The answer might simply be that emoticon use has increased so much even in the few years separating the research that we now process them more as faces than we did before. As Churches points out, our brains are “amazingly dynamic” in response to new environments.

“Thirty years ago, this activation of face-selective areas in the brain would not have been seen when 🙂 was presented,” he says. “But our world has changed and so have we.”


About the author

Eric Jaffe is an editor at CityLab, where he writes about transportation, history, and behavioral science, among other topics, through the lens of urban life. He's also the author of The King's Best Highway (2010) and A Curious Madness (2014)