Date created: | Last Updated:

Category: Project

Description: There are two competing approaches to the semantics of face emoji. The pictorial approach (Maier 2021) proposes that face emoji are stylized pictures of faces and that they are interpreted to represent the face of the author when typing the emoji. On this account, small differences between emoji that do not correspond to differences in action units (AU) of human faces (Ekman & Friesen 1978) should not be semantically relevant. By contrast, the lexicalist approach (e.g. Grosz et al. accepted) treats emoji as linguistic expressions. It allows for the possibility that the minimal units an emoji consists of (e.g. eye or mouth shapes) carry conventionalized meaning. Accordingly, small non-AU differences in emoji can be meaningful. To assess the explanatory power of the two approaches, we compare two groups of emoji pairs in an experimental study: Emoji pairs with a visual difference that corresponds to an AU-difference and emoji pairs with a visual difference that does not correspond to an AU-difference. The annotation of AUs in emoji is based on a FACS coding rubric for emoji by Fugate & Franco (2021). For each emoji pair, we create two contexts, which are each fitted to correspond to a prominent usage/meaning of one of the two emoji. The assumed usage/meaning of the emoji is based on their description on Emojipedia.org and based on the results of a norming study with speakers of German. In a forced-choice task, we ask German-speaking participants to choose between the two emoji of a pair for each context. If context-matching emoji will be chosen more often than at chance-level for emoji of pairs with an AU-difference and more often than for emoji of pairs without an AU difference, this would be in line with the pictorial approach. In contrast, if the rate with which context matching emoji are chosen is above chance level for both types of emoji and without a significant difference between them, the result would be in line with the lexicalist approach.