The Performance of Touch Screen Soft Buttons

This paper was published by Seungyon Lee and Shumin Zhai.

In this paper, Lee and Zhai investigate how efficient soft buttons are, compared to traditional hard buttons. After an introduction, they discuss four basic properties of touch screen interaction:

• Operational mode (stylus or finger)
• Activation mechanism (contact with screen or force applied to screen)
• Feedback enhancements (audio or vibro-tactile)
• Button size

The rest of the paper is dedicated to three experiments designed to test the effects of these properties on speed and accuracy. The first experiment tested operational mode and feedback on soft and hard button devices. They found that there was little difference between audio and vibration feedback on a soft button device. Also, hard and soft buttons performed at nearly the same level of accuracy and speed, with finger-operated soft buttons being slightly worse.

The second experiment compared contact-based (capacitive) screens with force-based (resistive) screens. As with the first experiment, accuracy was very high for all types of activation, including hard buttons. While capacitive and resistive screens each have pros and cons, Lee and Zhai point out that they are essentially equivalent and actually perform slightly better than hard buttons.

The third and final experiment was the only one where variables actually made a significant difference in the results. In it, they tested button size versus activation mode. While speed did not change significantly in small versus large buttons, the number of additional characters entered to correct errors was much higher with small buttons.

I found the results of this paper to be a little surprising, as I assumed that hard buttons would always out-perform soft buttons. That being said, I feel like they could have eliminated some repetition in describing their findings. Reading the description and then the summary of each experiment would have been enough to get all necessary information, without the lengthy middle section. This is especially true on the first two experiments where the results for the different variations were statistically the same. Still, the paper was informative and provides a good basis for future work in the area of soft button design.