Math structure biases attention during arithmetic problem solving: A webcam eye-tracking study [Author Accepted Manuscript]

Using lab-based eye-trackers, previous studies have demonstrated that visual presentation, such as positions of high-order operators (HOOs) and presence of superfluous brackets, influences gaze behaviors during arithmetic problem solving (e.g., Egorova et al., 2024). The current study explored the feasibility of moving eye-tracking arithmetic problem solving studies online by replicating Egorova et al.’s (2024) study using a webcam-based eye-tracker (i.e., WebGazer). Furthermore, we examined whether the proportion of gazes on the HOO over time, a gaze measure that was not analyzed in Egorova et al. (2024) but commonly used in WebGazer studies, can suggest participants’ online problem solving strategies. We analyzed gaze data from 119 college students who mentally evaluated simple arithmetic expressions where the HOO appeared in left, center, or right positions with or without superfluous brackets. Replicating Egorova et al.’s (2024) findings, participants registered their first gaze on the HOO faster when the HOO was on the left compared to on the right (a left-to-right tendency), and slower when brackets were present compared to absent around the center HOO (center brackets effects). Nevertheless, the online gaze data did not replicate the gaze difference in left versus center HOO positions. Furthermore, results of the proportion of gazes on the HOO over time confirmed the left-to-right tendency and further indicated that superfluous brackets guide gaze behaviors in both early and late stages of evaluations. Based on these findings, we discuss the feasibility and cautions of using WebGazer in online eye-tracking arithmetic problem solving studies. This research is based upon work supported by the National Science Foundation under Grant No. 2320053 to Worcester Polytechnic Institute. reviewed acceptedVersion