Text Me If You Can: Investigating Text Input Methods for Cyclists

Cycling is emerging as a relevant alternative to cars. However, the more people commute by bicycle, the higher the number of cyclists who use their smartphones on the go and endanger road safety. To better understand input while cycling, in this paper, we present the design and evaluation of three text input methods for cyclists: (1) touch input using smartphones, (2) midair input using a Microsoft Hololens 2, and (3) a set of ten physical buttons placed on both sides of the handlebar. We conducted a controlled indoor experiment (N = 12) on a bicycle simulator to evaluate these input methods. We found that text input via touch input was faster and less mentally demanding than input with midair gestures and physical buttons. However, the midair gestures were the least error-prone, and the physical buttons facilitated keeping both hands on the handlebars and were more intuitive and less distracting