Text Font Affects Reading Speed at a Low Luminance but Not at a High Luminance

Abstract

We are beginning a series of studies in which we shall investigate which characteristics of text fonts may be successful in allowing text to be read with different kinds of degradations that may be experienced by low-vision patients. As a first attempt to simulate one of the problems that certain patients might have, we measured reading speeds with two proportionally spaced fonts at a high photopic luminance (146 cd m−2) and at a very low luminance (0.146 cd m−2). We used the RSVP (rapid serial visual presentation) reading method, which eliminates the need for scanning eye movements. The two fonts were ‘Swiss’, a simple sans-serif font, and ‘Dutch’, a serif font similar to Times Roman, both presented at the 20/80 size; letters were white on black, with a contrast of 0.94. Subjects were young, normally sighted high-school and optometry-college students. The reading speed results from 47 subjects were as follows, for four conditions: high luminance, Swiss 531 and Dutch 540 words min−1; low luminance, Swiss 479 and Dutch 429 words min−1. At the high luminance, there is no significant difference between reading rates. There is a significant advantage for the Swiss font at the low luminance: p=0.005. There may be a significant difference in reading speed with different fonts when the patient's perceptions are degraded by disease and/or aging processes. Other parameters to be investigated with this method include contrast, size, blur, and visual field position. The simulation results will suggest parameter values to test on low-vision patients.