Evaluation of Information System Mental Model Based on Eye Movement Data

As the process of mental model matching evaluation is highly subjective and difficult to quantify and compare, this paper adopts the eye tracking method to optimize it and compares the actual gaze order of the user when operating the interface with the ideal gaze order in the "representation model." The eye-movement evaluation metrics of "mental model initial variance," "mental model reconstruction," and "mental model operational resistance" are proposed in this paper, and their effects are validated. Taking the electronic reconnaissance task interface as an example, the evaluation object indicates that the above indexes can effectively distinguish the interfaces with different mental model matching degrees. Moreover, it can be combined with user interview data to locate the specific locations and causes of a mental model mismatch in the interface, which can help provide reliable data support for interface evaluation and assist in interface optimization design.