Parsing Pointer Movements in a Target Unaware Environment

Abstract

Analysis of the movements of the Mouse pointer could lead to valuable insights into a user’s mental status in digital environments. Previous research has yielded data showing a significant link between user mental status and pointer movements ^[1]. However, there is currently no standardized system to detect and parse out individual targeted movements of a mouse pointer by a user in an active environment. Active analysis of mouse movements could be useful in situations where the emotional state of the user is being measured. Data was collected through the Mathspring Project including results of problems solved, the facial expressions and self-reported emotions of students, and the movements of the mouse pointer, which is the focus of this work [3]. Although a connection has been shown in previous research [1], the ability to track this in a live system is held back by the manual process for splitting the motions of the pointer. The focus of this project is the development of a generalizable system to parse these movements automatically without needing much processing power or an immense amount of training data for each time.