A study was conducted to investigate the effects of color CRT misconvergence and display luminance on reading performance and perceived image quality. The objectives of this study were (1) to determine the effects of color CRT misconvergence on reading time and errors, (2) to determine the effects of color CRT misconvergence on subjective ratings of image quality, and (3) to examine the interaction between color CRT misconvergence and display luminance upon reading performance and subjective ratings of image quality.

Ten participants performed a Tinker Speed of Reading Test on a shadow-mask, color CRT computer workstation. Both search time to find the out-of-context word and errors were recorded. Following the reading task, each participant rated the image quality of the display using a 9-point subjective scale.

Reading performance was not affected by CRT misconvergence or character luminance. Increasing misconvergence did result in lower image quality ratings; however, it was found that participants perceived the more luminous displays to have higher image quality. Also, it was determined that image quality ratings were lowest for magenta misconvergence. However, using normalized ratings, it was found that image quality ratings decreased quickest for the yellow characters with increasing misconvergence. These findings indicate that lower levels of misconvergence (1 to 2.5 arcmin) may not affect reading performance adversely. However, since individuals perceive display quality to be poor at these levels of misconvergence, color CRTs should be used only when application absolutely requires them.