Emojis and emoticons have been demonstrated to be capable of many complex linguistic functions, beyond serving as simple markers of emotion. One specific conventional use of emojis and emoticons is to signal sarcasm or irony (Dresner & Herring 2010, Filik et al. 2016). While there are no ERP studies investigation the on-line processing of linguistic emojis, there are ERP studies investigating irony processing; these studies have found increased P200 and P600 effects to ironic vs. non-ironic utterances (Regel et al. 2011, Spotorno et al. 2013). This study bridges these two strands of research and investigates ERP responses to irony-producing emojis, the question being whether emoji-generated irony is processed similarly to word-generated irony.
In the experiment, positively- or negatively-valenced sentences were followed by either a congruent, incongruent, or ironic (winky-face) emoji:
You are such a jerk ☹
You are such a jerk :)
You are such a jerk ;)
ERPs were time-locked to these delivered-in-context emojis. An analysis of behavioral data revealed that only some participants treated the wink emojis as ironic in this study, and further analysis reveals that the traditional irony brainwave response (P600 effect) is present only in those participants who registered the ironic interpretations of these emojis. This is consistent with previous studies of irony, as well as with more general accounts that suggest the P600 effect reflects some sort of re-processing. This study is (maybe?) the first to investigate the neural correlates of how emojis are processed in linguistic contexts, and moreover provides a link between individual differences in interpretation of commonly-used graphical-linguistic pragmatic markers and on-line neural processes.