UI Design Newsletter – May, 2000

Two of the variables most closely related to readability seem to be:

1. display polarity, where Snyder, et.al. (1990) reported that black characters on a white background elicits faster reading performance than white characters on a black background, and
2. screen resolution.

Although screen resolution is known to be an important factor, the exact relationship between display resolution and readability is not yet clear.

Martha Ziefle at the Institute of Psychology in Germany conducted two studies where she tried to better understand the relationship between screen resolution and reading performance.

Ziefle investigated the effects on reading performance using hardcopy (measured at 255 dots per inch or "dpi") and two resolutions of monitors, 1664x1200 pixels (120 dpi) vs. 832x600 pixels (60 dpi). She measured reading speed and proofreading accuracy.

Participants read from the same 19-inch monitor using black characters on light background. The subjects viewed the material from a distance of 20 inches (50 cm).

In this well-controlled study, reading from hard copy was reliably faster (200 wpm vs.180 wpm) and elicited reliably greater proofreading accuracy than reading from the monitors. Also, 80% of the participants preferred reading from paper, and the remaining 20% preferred the higher resolution screen. However, the high-resolution screen did not elicit better reading performance than the low-resolution screen.

The main difference between the second experiment and the first was the task that the participants performed. In the second experiment, they performed a continuous visual search task rather than a proofreading task. She compared their performance on the following:

High resolution screen: 1024x768 pixels (measured at 89 dpi), and Low resolution screen: 720x540 pixels (measured at 62 dpi).

Overall, she found that participants searched reliably faster using the high-resolution display. After 30 minutes using the low resolution screen, the subjects began to:

1. search slower,
2. make more errors, and
3. have more and longer fixations.

Her two studies suggest that one of the main contributors to slower reading performance, at least after 30 minutes, could be the resolution of the characters being read. Based on the fact that the lower resolution screen produced more and longer fixations, the slower reading may come from users requiring more time (per word or phrase) in perceiving the characters.