The effects of mobility, encumbrance, and (non-)dominant hand on interaction with smartwatches
David Dobbelstein, Gabriel Haas, Enrico Rukzio
Abstract
David Dobbelstein, Gabriel Haas, Enrico Rukzio
Abstract
Smartwatches are designed for short interactions in varying mobile contexts. However little data is available on how present mobile conditions affect interaction with these devices. In this work, we investigate the effects of mobility (walking), encumbrance (by carrying items like shopping bags) and wearing the watch on the (non-) dominant hand on interaction techniques present with current devices: tapping targets, swiping, and flicking the wrist. The results showed that for tapping and swiping, the outfitted hand had the largest effect on selection time (9.41%, resp. 4.84% slower interaction when the watch was worn on the dominant hand), while for wrist flicking, encumbrance had the largest effect (11.94% slower when carrying bags). The walking condition had the largest effect on the error rate for all techniques. Swiping as an interaction technique was barely affected by any condition, both in terms of selection time and error rate, making it a robust mobile interaction technique for smartwatches.
OpenAlex reports 16 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Smartwatches are designed for short interactions in varying mobile contexts. However little data is available on how present mobile conditions affect interaction with these devices. In this work, we investigate the effects of mobility (walking), encumbrance (by carrying items like shopping bags) and wearing the watch on the (non-) dominant hand on interaction techniques present with current devices: tapping targets, swiping, and flicking the wrist. The results showed that for tapping and swiping, the outfitted hand had the largest effect on selection time (9.41%, resp. 4.84% slower interaction when the watch was worn on the dominant hand), while for wrist flicking, encumbrance had the largest effect (11.94% slower when carrying bags). The walking condition had the largest effect on the error rate for all techniques. Swiping as an interaction technique was barely affected by any condition, both in terms of selection time and error rate, making it a robust mobile interaction technique for smartwatches.
Key concepts: Smartwatch, Mobile interaction, Tapping, Computer science, Human–computer interaction, Selection (genetic algorithm), Mobile device, Word error rate