Screen Magnification for Readers with Low Vision: A Study on Usability and Performance
Meini Tang, Roberto Manduchi, Susana Chung, Raquel Prado · 2023 · Proceedings of the 25th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS 2023) · doi:10.1145/3597638.3608383
Summary
This study investigates how people with low vision use two common screen magnification modes — full screen and lens — when reading documents and exploring web pages on a laptop computer. The researchers recruited 20 participants with a range of low vision conditions (including macular degeneration, glaucoma, retinitis pigmentosa, nystagmus, and others) from a university optometry clinic. Participants read text documents aloud and located titles on web pages using both magnification modalities, provided through the native macOS screen magnifier on a MacBook Pro with a 13.3-inch display. The study was designed around three core questions: whether one magnification mode leads to better reading performance, how users physically control the center of magnification via mouse movement, and where users prefer to position their area of interest on screen (measured through gaze tracking). Participants were free to choose their own magnification factor and, for the lens mode, their preferred window size. The experimental design was counterbalanced with practice sessions before each trial. Performance was measured through total reading speed, per-line reading speed (excluding retracing time), reading speed deficit due to retracing, horizontal scrolling consistency and uniformity, total path traversed by the mouse, exploration time for web pages, and subjective difficulty ratings collected via exit questionnaires.
Key findings
No statistically significant differences were found between full and lens magnification modes for any of the primary performance metrics — reading speed, exploration time, or subjective difficulty ratings. Participants were evenly split on which mode they preferred, with individual reasons varying considerably. However, the two modes produced distinctly different mouse movement patterns. When using lens mode, participants maintained more consistent vertical positioning (lower Y-coordinate standard deviation, p=0.000714) and more uniform horizontal motion (lower zero-velocity proportion, p=0.000545), indicating tighter, more disciplined mouse control. In full mode, participants made longer and more frequent pauses and traversed shorter total path lengths, suggesting a more parsimonious movement strategy. The preferred angular print size correlated linearly with visual acuity (slope near 1.0 on a log scale), and magnification level and age were the main factors affecting reading speed. Retracing — moving from the end of one magnified line to the start of the next — accounted for an average 18% reading speed deficit, with central vision loss increasing this penalty. Gaze tracking data in full mode revealed that most participants positioned their area of interest in the middle-to-right portion of the screen horizontally, and in the central third vertically, when reading text documents.
Relevance
This research has direct implications for both screen magnification software developers and accessibility practitioners. The finding that neither full nor lens mode is objectively superior reinforces that magnification tools should support both modalities and let users choose based on personal preference and task context. The detailed mouse movement analysis provides concrete design guidance for automatic magnification control systems — for example, gaze-based magnifiers should mimic the consistent patterns seen in lens mode and the parsimonious motion of full mode. The 18% reading speed penalty from retracing highlights a specific area where assistive technology could improve, such as cursor-locking features (like ZoomText offers) that constrain vertical mouse drift during horizontal scanning. For web developers, the study underscores that screen magnification users navigate pages very differently from sighted users, spending substantial time and effort finding content — a strong argument for clear visual hierarchy and predictable page layouts.
Tags: screen magnification · low vision · reading performance · usability · mouse interaction · gaze tracking · assistive technology