B) Meaning Emoji

1 view

Skip to first unread message

Fe Kittner

unread,

Aug 4, 2024, 3:38:11 PM8/4/24

to highbartdele

Apparentlythere are more than a few people who interpret that as meaning the other person suddenly hates you and never wants to see you again or hear from you again because they are so angry over what you just texted regarding, say, composting.

Steve Pokin writes the Pokin Around and The Answer Man columns for the Springfield Daily Citizen. He also writes about criminal justice issues. He can be reached at spo...@sgfcitizen.org. His office line is 417-837-3661. More by Steve Pokin

The mission of the Springfield Daily Citizen is to inform our community and be a catalyst for good. Our vision is to reinvent local news in Metro Springfield by telling the stories of our community, bringing issues to light, encouraging discourse and inspiring citizens to vote and take action.

Copyright: 2020 Holtgraves, Robinson. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The research was funded by a grant from the National Science Foundation (BCS-1917631) which was awarded to TH. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

In a related study, Filik and Thompson [15] examined perceptions of literal and sarcastic messages as a function of whether the message included emoticons, various types of punctuation, or neither. The interpretation of an intended sarcastic message was not impacted by emoticons or punctuation when there was contextual support and the meaning clear. However, intended literal messages were more likely to be perceived as sarcastic when the message included emoticons. In contrast, when the intended meaning was ambiguous, including an emoticon (wink emoticon) increased the likelihood of sarcastic intent being recognized. Research conducted by Riordan [16, 17] suggests that these effects are not limited to face emoji. In several studies she demonstrated that the inclusion of nonface emoji in a message can reduce perceived ambiguity, and increase interpretation confidence, in addition to influencing judgments of conveyed affect, especially positive affect.

More recently, Weissman and Tanner [18] demonstrated that emojis (wink) used to convey irony generally elicit the same neural responses as irony conveyed with words. Specifically, they observed enhanced P200s (an early indicator of attention-related processes) and P600s (reflecting reanalysis or integration processes) to wink emoji. Importantly, judgments of irony correlated with the occurrence of a P600; i.e., participants who interpreted the message as ironic were more likely to display a P600. There does appear to be a processing cost associated with emoji, at least in some situations. Cohen et al. [19] examined comprehension processes when emoji were substituted for words. They examined both nouns (e.g., Fig A in S1 File for pizza) and verbs (e.g., Fig B in S1 File for love). They found a processing cost for the use of emoji (for both nouns and verbs) as indexed with self-paced reading times, a finding consistent with the enhanced P600 reported by Weissman and Tanner [18]. Interestingly, they found no differences between words and their emoji substitutes in terms of their comprehensibility.

The purpose of this experiment was to examine the impact of emoji on the endorsement of indirect interpretations. Participants read scenarios, questions, and indirect replies and then judged the meaning of the reply. We expected emoji to facilitate recognition of the indirect meaning (increased accuracy and decreased response time) of a reply, relative to the same reply without an emoji.

The experiments presented in this manuscript were approved (exempt status) by the Institutional Review Board of Ball State University. The title of the project for which the approval was granted was "Understanding Conversations" (IRB protocol 1216210). An informed consent signature waiver was requested and granted by the IRB. Participants engaged in the consent process and provided oral consent to the experimenter.

There were 36 critical scenarios. Due to an error, one of the critical interpretations (request refusal) was positively rather than negatively worded and was dropped from all analyses, resulting in 35 critical scenarios.

Twenty-four of the scenarios (situations, questions, replies, and paraphrases) were adopted from earlier research [19,22]. Twelve new (but very similar) scenarios were written for the current research. Each scenario consisted of a brief description of two people followed by a question-reply exchange. There were three different types of situations based on the question which was either a request for an opinion (e.g., "What did you think of my presentation?"), a request for a disclosure (e.g., "How did you do in chemistry?"), or a request for action (e.g., "Can you type my term paper for me?"). There were 12 of each situation type. A sample opinion scenario is presented in Table 1 (all materials are available in the S1 Appendix). The reply always violated the relation maxim by not providing the requested information or action. These replies were all excuses and functioned by providing a reason for why the opinion or disclosure might be negative, or why one could not comply with a request. On one-third of the trials, the reply included an emoji, on one-third of the trials the reply did not contain an emoji, and on one-third of the trials the reply contained only an emoji (i.e., no words). Three stimulus lists were created so that participants saw an equal number of replies with an emoji, without an emoji, and with only an emoji, and across the experiment an equal number of participants saw each of the three versions of each reply.

After reading the remarks, participants judged a potential interpretation of the reply. This paraphrase was always a face-threatening interpretation of the reply (see Table 1). To keep participants from believing that the paraphrases were always reasonable interpretations, 36 filler trials were included in which the paraphrase was clearly an incorrect interpretation. In all other respects, the filler trials were similar to the critical trials: One-third of the replies included an emoji, one-third did not contain and emoji, and one-third contained only an emoji. Presentation order of the 72 scenarios was randomized for each participant.

A pretest was conducted in order to select the emoji to be used in this study. Participants (N = 41) in the pretest (who did not participate in the subsequent two main studies) were presented with 24 of the scenarios (the set adapted from [21]). Following each reply, participants were asked to choose an emoji (from 20 common emoji that were presented) that they would use to help the recipient understand the meaning of the reply. The emoji most frequently chosen for each of the three situation types was chosen for use in the present experiments. A grimace Fig C in S1 File was selected for use with the opinion and request refusal scenarios; a sad emoji Fig D in S1 File was used for the negative disclosures. Even though we used the emoji selected most often, it should be noted that, consistent with research demonstrating the underlying ambiguity of emoji [29]), consensus regarding emoji selection was relatively low. All emoji appearing in this manuscript were obtained from the Open Emoji library. Stimuli used in the experiments were Apple Emoji.

Paraphrase accuracy and paraphrase reaction time were analyzed with a linear mixed-effects model that included Situation type (disclosure vs. opinion vs. request refusal) and Emoji condition (No emoji vs. Emoji and text vs. Emoji alone) as fixed effects and the intercepts for participants and stimuli as random intercepts. Initial analyses that included random slopes for participants and stimuli failed to converge and hence were not included in the model. For paraphrase reaction times only trials for which participants endorsed the indirect reply were included. Extreme reaction times (less than 100 ms and greater than 4000ms) were treated as outliers and excluded from the analyses (In general, then, participants were more likely to endorse an indirect, face-threatening interpretation of a reply, and to do so more quickly, when the reply contained an emoji than when it did not contain an emoji. This effect varied over situation type, and was most clear for opinions and disclosures, and less clear for request refusals. There are differences in the pragmatic mechanisms involved in these different situation types that may account for these differences, a topic that we return to in the General Discussion.

The purpose of this experiment was to replicate Experiment 1 and to examine the possible effect of emoji location. Rather than placing the emoji before the text as in Experiment 1, in this experiment the emoji was placed after the text (for the text plus emoji condition).

Participants (N = 49; 7 males) were undergraduate students enrolled in Introductory Psychology courses who participated for partial course credit. None of these participants took part in Experiment 1.