Effects of Emojis on Perceived Meaning of Text MessagesMertkan Alacahan1Abstract—Emojis are widely used in text messaging in orderto emphasize a notion and augment the meaning of a particulartext. In context of modern informal text communication, inclusionor lack of certain emojis in a text message is likely to elicitdifferent emotions in receivers than intended by the sender. Inthis study, this possibility is hypothesized and explored. Varioustexting scenarios with and without the usage of emojis areprepared. Participants were asked to evaluate these scenarioswith help of a rating scale. Results show that presence and lackof emojis cause people to interpret text messages differently.I. INTRODUCTIONRapid increase in social media usage over the years madetext communication more sophisticated than simple e-mailexchanges. About %80 of application users on smartphonesuse them to reach social networks, instant messaging applicationsand blogs 1. Users of messaging applications areable to enhance their texts with various other media such asphotos, videos, voice recordings and emojis.An emoticon is a “typographic symbol that appears sidewaysas resembling a facial expression” 2 like :). Emoticonsare being used in text communication since its first adoptionin 1982 by Scott Fahlman, a computer scientist at CarnegieMellon University 3. Emojis, however, are different thanemoticons although the words are often used interchangeably.Emojis are “picture characters or pictographs” that representnot only facial expressions but also feelings, ideas andconcepts. 4.II. RELATED WORKIn face-to-face communication, people use non-verbal cueslike facial expressions to support their verbal efforts. Intext communication, emoticons are utilized to replace thesenon-verbal cues in an effort to achieve the same effect 5.Additionally, emoticons can affect the interpretation of a textso dramatically that “the same text can be perceived as eitherhappy or sad depending on which emoticon accompanies it”6.Regarding computer-mediated communication, a significantamount of research has been dedicated to explorethe effects of not only emoticons but punctuation too. Amethodology-wise similar study by Kenneth et al. 7 hasshown that some text messages can be “perceived as morenegative, or less enthusiastic, with a period”. By using asimilar approach, this study aims to test the hypothesis thatlack of emojis appropriate for the context can make message*This work was not supported by any organization1Mertkan Alacahan is with Informatics Institute, GameTechnologies, Middle East Technical University, Ankara, Turkeymertkan.alacahan at metu.edu.trreceivers interpret the messages in a different way thanintended.III. METHODOLOGYScreenshots of text messages were prepared with helpof an iPhone text generator website 10 (See AppendixA). Each texting scenario consists of four text messages intotal; two for each side of the conversation. There were twoversions of screenshots; with an emoji appropriate for thecontext in the last message of the conversation and withoutany emojis. A total of six emojis representing commonreactions were tested (Happy, sad, affectionate, silly, angry,surprised. See Appendix B). Text messages were preparedin Turkish language in order to make participants comfortablewhile interpreting the messages by using their nativelanguage since all participants were Turkish.A total of 16 participants (6 female, 10 male) answeredthe questions associated with each scenario using a Likertscale from 1 to 7. Half of the participants answered thequestions related to conversations without any emojis and theother half were presented with conversations with emojis. Inorder to make participants relate more easily and strongly toconversations, they were asked to imagine that they were theone side of conversations (See Appendix A).IV. RESULTS AND DISCUSSIONOut of the six scenarios associated with six differentemojis considered by participants, there were noticable differencesin results regarding four of the scenarios/emojis(Happy, sad, affectionate, angry) whereas lack or inclusion oftwo emojis in two scenarios (Silly, surprised) did not affectthe result significantly.The texter who ended their message with a happy emojiwas perceived as more willing (M = 6.875, SD = 0.33, SEM= 0.125) to attend to a party than the texter who did notinclude any emojis in their text (M = 5.25, SD = 0.66, SEM= 0.25. Some participants in the experiment stated that theyinterpreted the one-word answer without emoji as the senderpossibly being reluctant.The texter who ended their message with a silly emojiwas not perceived as if they were joking (M = 4.25, SD =0.96, SEM = 0.36) significantly more than the texter whodid not include any emojis in their text (M = 4.125, SD= 0.78, SEM = 0.29. This indifference might have occuredbecause the context of conversation clearly suggested thatthe last message of the conversation was most likely a joke,thus, inclusion or exclusion of emoji might not have made adifference.The texter who ended their message with an angry emojiwas perceived as more furious (M = 6.625, SD = 0.48, SEM= 0.18) about a previous conversation between participantsthan the texter who did not include any emojis in their text(M = 5.5, SD = 1.11, SEM = 0.42. Most participants in theexperiment who evaluated the conversation without emojiremarked that they couldn’t tell if the person is angry orsimply giving an attitude or not really furious but just slightlyannoyed.The texter who ended their message with a surprised emojiwas not perceived as more fascinated (M = 5.25, SD = 0.96,SEM = 0.36) about an interesting fact than the texter who didnot include any emojis in their text (M = 5, SD = 0.7, SEM =0.26. Usage of another fascination indicator, the expression”Wow”, might have influenced the result so that the presenceof emoji did not make a difference.The texter who ended their message with a sad emoji wasperceived as more upset (M = 6.625, SD = 0.48, SEM =0.18) about a recent occasion witnessed by participants thanthe texter who did not include any emojis in their text (M= 5.5, SD = 0.86, SEM = 0.32. Some participants who wereassigned the conversation without emoji pointed out that thesender of the last text might just be sarcastic and not reallyunhappy about the situation.The texter who ended their message with an affectionateemoji was perceived as more adoring (M = 6.375, SD = 0.69,SEM = 0.26) towards their partner than the texter who did notinclude any emojis in their text (M = 5.25, SD = 0.96, SEM= 0.36. Interestingly, three female participants stated that thisemoji is what they use when they want to express affectionand this real-life experience affected their judgment.V. CONCLUSIONSTexting is a form of communication which lacks linguisticand non-verbal cues. Emojis are one of the tools that peopleuse in place of these cues in order to express themselvesbetter. This conducted experiment can serve as a startingpoint to show the differences in people’s perception of textmessages based on the usage of emojis.A limitation of this experiment is that participants werenot actually involved in texting, but only observed an alreadyconcluded one. This might have made the participants lessinvolved and less able to empathize with their role in theconversation. Conducting this experiment in an online settingwith real-time stimuli has the potential of yielding moreconvincing results.Various factors might influence a texter’s behavior in atexting session incluing age, sex, topic of the conversationand the dynamics of the relationship between participants.Hancock et al. already have found evidence that texterstend to assume a less formal mindset and express theiremotions more when their texting partner is someone theyknow 8. The factors mentioned were not accounted for inthis experiment and thus it can be expected that consideringthese variables in a more extensive study would lead to morein-depth results.Over the years, people have engaged and continue toengage in text-based online communication via differentmedia such as e-mail and messenger applications on variousdevices like cell phones, tablets and computers. Nature ofthese communications is also subject to change with thepossibility of using elements other than plain text as supportivecues. For example, in addition to emojis, usage of theGraphics Interchange Format (GIF) is becoming more andmore ubiquitous 9. Even the usage or lack of punctuationin certain contexts makes a difference in the perception ofmeaning, as mentioned in “Related Work” section. Whatother supportive media will rise to popularity and be usedin text-based communication remains to be seen. However,there is room for more studies focusing on the effects of theusage of media that is already present, some of them beingvoice and video clips.VI. ACKNOWLEDGEMENTSI would like to thank my family, friends and co-workers fortheir participation and advice throughout this study, as wellas Professor Cagiltay for his contributions via the course.REFERENCES1 Quorus Consulting Group Inc. (2012). 2012 cell phone consumerattitudes study report. Ross, C., Orr, E., Sisic, M., Arseneault,J., Simmering, M., & Orr, R. (2009). Personality and motivationsassociated with Facebook use. Computers in Human Behavior, 25,578586.2 Walther, Joseph B., and Kyle P. DAddario. 2001. The Impacts ofEmoticons on Message Interpretation in Computer-Mediated Communication.SSCR 19 (3): 32447.3 Joon Young Lee, Nahi Hong, Soomin Kim, Jonghwan Oh, JoonhwanLee, Smiley face: why we use emoticon stickers in mobile messaging,Proceedings of the 18th International Conference on Human-ComputerInteraction with Mobile Devices and Services Adjunct, September 06-09, 2016, Florence, Italy.4 Zareen, Nusrat. (2016). Psycho emotional Impact of Social mediaEmojis. Isra Medical Journal. 8. 257-262.5 Derks, Daantje, Agneta H. Fischer, and Arjan E.R. Bos. 2008. TheRole of Emotion in Computer-Mediated Communication: A Review.Computers in Human Behavior 24 (3): 76685.6 Lo, Shao-Kang. 2008. The Nonverbal Communication Functions ofEmoticons in Computer-Mediated Communication. CyberPsychology& Behavior 11 (5): 59597.7 Kenneth J. Houghton, Sri Siddhi N. Upadhyay, Celia M. Klin, Punctuationin text messages may convey abruptness. Period, In Computersin Human Behavior, Volume 80, 2018, Pages 112-121.8 J.T. Hancock, C. Landrigan, C. Silver, Expressing emotion in textbasedcommunication, Proceedings of the SIGCHI: ACM conferenceon human factors in computing systems (2007, April), pp. 929-932.9 Kate M. Miltner, Tim Highfield, Never Gonna GIF You Up: Analyzingthe Cultural Significance of the Animated GIF, Social Media + Society,2017.10 iPhone Text Generator at http://ios.foxsash.com/APPENDIXAppendix A: A sample text conversation with and withoutemojis and related questionAppendix B: Emojis used in the study