Are you feeling 🙂 today? Or perhaps 🙁 ? In the modern digital world, we are all conversant with emoticons and emojis. But when you come across the ? emoji, do you know what it means?

According to a new article on the Lisbon Emoji and Emoticon Database (LEED), published in the Psychonomic Society’s journal Behavior Research Methods, this emoji is among the least meaningful and most divisive emojis. Fewer than half of the people tested to create the database could assign it a meaning, with most of those who did, believing the emoji means “speechless”.

The internet is rife with speculation on its intended meanings. Yes, the emoji can be used to mean “speechless” or “silent”. But, it can also allegedly mean “vapid, without a thought”, be used to mark sadness, despair, or isolation from others­, or act as a threat.

On Twitter, ? is also sometimes used to mark subtweets – tweets specifically about someone who doesn’t use that person’s handle in the tweet itself:

Meanwhile, the official intended meaning of the emoji in Unicode, the standardized system used to index characters across operating systems, is simply “face without mouth”.

The LEED project, led by David Rodrigues, is the first large-scale attempt to measure how people interpret emoji and emoticons and to compare how the same emoji from across different operating systems (iOS, Android) are perceived. While designers and researchers may use emoji and emoticons believing that they will be interpreted with their intended meanings, the results of the LEED project demonstrate that people’s interpretations can often be quite different.

The emoji and emoticon ratings and meanings, which are all freely available, may be particularly useful to psychologists who want to study emoji and emoticon use in the wild or to use them as stimuli in studies. Outside the field, they could be of interest to designers trying to get across a clear emoji message.

A brief history of emoticons and emoji

The invention of emoticons (a combination of emotion + icon), like 🙂 or :-(, is attributed to Scott Fahlman, a computer scientist, who, in 1982, proposed their use to distinguish joke posts from non-joke posts in online bulletin boards. Their use spread because they could facilitate communication in written text. As he explained, these kinds of markers were needed because “using text-based online communication, we lack the body language or tone-of-voice cues that convey this information when we talk in person or on the phone.” Emoticons became widely adopted as text-based communication spread online.

The late 1990s brought the next big breakthrough in ideograms, with Japanese designer Shegataka Kurita creating emoji (a combination of e [picture] + moji [character] in Japanese) to improve communication on mobile devices. This now-iconic original set of 176 emoji, shown in the figure below, was recently added to the permanent collection at the Museum of Modern Art in New York.

Since their introduction, emoji popularity has skyrocketed. New standardized sets of emoji are periodically released by Unicode, with the most recent release numbering over 2600. Operating systems, such as iOS and Android, develop their own versions based on these standards. Social media platforms like Facebook have created their own versions, while big brands, including Disney, Domino’s, Pepsi, and Ikea, have developed their own non-standard additions.

Emoji have made it into advertising (including advertising targeted at users who use emoji), music videos (both low-brow and high-brow), cookies, and clothing. We may have recently hit peak-emoji with the world’s first emoji translator and the critically unacclaimed Emoji movie.

Emoji and emoticon research

There has been a corresponding uptick in research to understand people’s emoticon and emoji use. For example, there have been big data efforts to understand people’s use of emoji to convey positive and negative sentiment, use of emoticons in text messages, and use of emoji and emoticons to describe their reactions to food. However, previous research has been limited by having to rely on coder interpretations of emoji and emoticon meanings based on their surrounding text, rather than on the meanings of the characters themselves.

Psychology studies that have used emoji as stimuli have had to rely on researchers’ own interpretations of the selected emoji, to extensively norm the subset of emoji to be used, or to use computer-based analyses, with the bias associated with reliance on text-based interpretations.

Creating the LEED

LEED is the first large-scale database of people’s perceptions and interpretations of emoji and emoticons. To build the database, researchers David Rodrigues  and colleagues collected people’s reactions to 238 stimuli: 85 emoticons and 153 emoji (77 iOS, 63 Android, 9 Facebook, and 4 potential new emoji from Emojipedia). Most of the stimuli represent facial expressions. Emoji were selected to have similar Unicode meanings to the emoticons. When possible, the same emoji from Android and iOS were tested.

505 Portuguese participants each evaluated a subset of 20 of the stimuli. Most of the participants were women, and most used the Android operating system. For each stimulus, participants rated its aesthetic appeal, familiarity, visual complexity, clarity, valence (negative vs. positive), arousal (passive vs. arousing), and meaningfulness. They also wrote down its meaning. Individual ratings and interpretations of emoji and emoticons are all available online.

Several interesting patterns emerged from the data.

First, emoji were rated as more appealing, positive, arousing, meaningful, and clearer than emoticons. This may be related to the recent increase in emoji use as emoticon use has declined.

Second, the operating system of emoji matters. iOS emoji were rated as more appealing, familiar, clear and meaningful than Android emoji. This was likely not due to participants’ increased familiarity with iOS emoji, since most were Android users.

Finally, women rated emoji as being more familiar, clear, and meaningful than did men. Women tend to use more emoji than do men, and so they may be more familiar with their intended meanings.

Generalizing the findings

Some degree of caution is needed in generalizing the ratings and interpretations collected in LEED across populations. All the stimuli were evaluated by Portuguese participants sometime in the 2010s, which is important because emoji use can vary by culture and across­ time.

While emoji are used around the world, their frequency of use and the particular emoji that are popular differ across countries (for an accompanying world emoji atlas, see here). Interpretations of emoji can also vary across cultures. Since emoji originated in Japan, many of the objects and scenes pictured are Japanese and may be interpreted differently in other cultures – foods like oden and dango, a Japanese symbol for beginner drivers, a moon viewing ceremony, and a name badge that looks like fire to people unfamiliar with it.

Emoji that have straightforward meanings can also acquire more complex, metaphorical meanings over time. For example, the nail painting emoji and woman tipping hand emoji have come to express sass or nonchalance, the 100 emoji to represent agreement or perfection, and the upside down smiling emoji to mark silliness or sarcasm.

Apart from these expected variations, LEED offers a great resource for people’s standard reactions to emoji that are likely to be much more informative than taking “face without a mouth” or “smiling face with open mouth and cold sweat” at face value.

Article focused on in this post:

Rodrigues, D., Prada, M., Gaspar, R., Garrido, M. V., & Lopes, D. (2017). Lisbon Emoji and Emoticon Database (LEED): Norms for emoji and emoticons in seven evaluative dimensions. Behavior Research Methods, DOI 10.3758/s13428-017-0878-6.