Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Psychology


Cognitive Psychology

Content Description

1 online resource (xii, 120 pages) : color illustrations.

Dissertation/Thesis Chair

Heather Sheridan

Committee Members

Gregory Cox


E-Z Reader, emojis, eye tracking, Foveal Load Hypothesis, SWIFT, Eye tracking, Emojis, Reading, Eye

Subject Categories

Cognitive Psychology


Emojis are nonverbal elements used in text-based communication that may function like gestures in spoken communication (Feldman et al., 2017; McCulloch & Gawne, 2018). Like words, emojis can convey semantic information and support message comprehension (Lo, 2008; Riordan, 2017b). Nevertheless, the cognitive and perceptual processes involved in emoji recognition and integration during reading remain unclear. Specifically, it is unclear how early semantic processing of emojis begins as well as how emojis and text interact during reading. Such empirical questions have implications for contemporary models of eye movement control, such as E-Z Reader (Reichle et al., 1998, 2012) and SWIFT (Engbert et al., 2002, 2005). To investigate the time course of emoji and text processing, across two experiments I monitored participants’ eye movements while they read sentences containing a target word (e.g., coffee) with a semantically congruent (e.g., ☕️) or incongruent (e.g., ?) or no emoji. Both experiments revealed that emoji processing begins in the parafovea and has an overlapping time course with that of words. As evidence for parallel lexical processing during reading, Experiment 2 revealed a semantic emoji parafoveal-on-foveal (PoF) effect whereby a congruent parafoveal emoji facilitated foveal target word processing. Lastly, the data from Experiment 2 were reanalyzed to examine if processing difficulty at the fovea (i.e., foveal load) influences the degree of parafoveal processing (i.e., the foveal load hypothesis; Henderson & Ferreira, 1990) and subsequent processing of an emoji (i.e., spillover effects). Foveal load was not only operationalized by target word frequency, but also by orthographic distinctiveness, and semantic diversity. Although the results for the fixation time eye-tracking measures failed to document support for the foveal load hypothesis or spillover effects, the emoji skipping measure revealed greater parafoveal processing of incongruent emojis under low foveal load conditions. In summary, this dissertation demonstrates the generalizability of key assumptions of models of eye movement control during reading to emojified text.