Naturalistic walking and typing on a smartwatch: a comparison of input methods

Abstract

Despite widespread development within the technology industry, smartwatches have historically been unsuccessful in their sales worldwide. Several factors have contributed to the poor sales of smartwatches, namely the lack of an efficient means of text entry. The 2017 release of Android Wear and Apple WatchOS 4 addressed this issue as now both operating systems support some level of native text entry, albeit through different methods (tap, trace, and handwriting). Previous research has consistently shown tap and trace to be efficient means of text entry on a smartwatch, yet no studies have directly compared the two input methods to other available input methods, such as handwriting. Now that tap, trace, and handwriting input are all available for use by the mass consumer market, understanding how they compare is essential to ensuring users are receiving the best smartwatch typing experience possible. Given the ubiquitous and mobile nature of smartwatches, this study examined typing performance using three unique input methods (tap, trace, and handwriting) while standing, walking a simple route, and walking a complex route in a naturalistic setting. Results show participants typed faster with trace (30 words per minute) than with tap (20 WPM) and handwriting (18 WPM), regardless of whether they were standing or walking. The observed typing speeds with trace are among the fastest reported in the smartwatch literature. Trace also outperformed tap and handwriting across all subjective metrics, regardless of mobility. Participants subjectively rated trace as easier and less demanding to use, preferred it over tap and handwriting, and suggested they would use trace over tap and handwriting on a smartwatch regardless of their mobility. These results suggest smartwatch manufacturers should include QWERTY keyboards with trace input as a standard feature in order to provide the best overall typing experience for their customers.