Visual comfort models based on coloured text and neutral background combinations

Abstract

Reading on mobile phones can cause visual discomfort, negatively affecting visual health. Most studies have focused on neutral text-background combinations, with limited validation for coloured text-background combinations. This study investigates the impact of coloured text on neutral backgrounds and the colour difference between text and background on visual comfort during digital reading. A psychophysical experiment was conducted, where 230 images of coloured text on neutral backgrounds were evaluated by 20 participants using a 6-point scale for visual comfort. Results showed that reading coloured text on a black background generally provided higher comfort compared to a white background. Additionally, visual comfort decreased as the text colour approached that of the background. The effect of text hue on comfort was not significant. Furthermore, several visual comfort models for mobile displays were developed and compared. The VC_1-LAB model is based on Bern’s attributes, while the VC_2-LAB model focuses on the lightness of text and background. The VC_3-LAB model includes both lightness and chroma attributes. Comparisons revealed that VC_3-LAB outperformed the others in predicting visual comfort, highlighting the importance of lightness and chroma in improving predictive accuracy. Therefore, the VC_3-LAB model is useful for evaluating the visual comfort of coloured text on neutral backgrounds.