Writing and driving: An assessment of handwriting recognition as a means of alphanumeric data entry in a driving context

This paper reports on two studies conducted within a right-hand drive fixed-based driving simulator both of which investigated the use of handwriting touchpads as a means of entering alphanumeric data whilst driving. In Study 1, 12 right-handed participants drove a ten-minute route and on three occasions were instructed to enter an address into a navigation system (half with a conventional on-screen keyboard system; half with a prototype handwriting recognition system) using their non-preferred left hand. There was evidence that the time taken to enter a destination while driving was reduced and driving performance (speed variability) was improved when participants utilised handwriting versus the on-screen keyboard. Furthermore, participants had a strong preference for the handwriting system. Study 2 focused on the impact of the location of a handwriting touchpad. Fourteen right-handed participants drove two routes each of approximately ten minutes duration, and at pre-determined points were instructed to use either their nonpreferred (left) or preferred (right) hand to write a range of single characters. In this study, there was evidence that using the non-preferred hand for data entry was more demanding (increased glances, perceived workload) and the quality of handwriting was poorer, as compared with use of the right hand. With respect to driving performance, speed variability was found to be greater when drivers were using their left compared to right hand. It is concluded that handwriting touchpads do offer particular advantages over conventional alphanumeric data entry systems in cars, but should be capable of being placed in a range of positions within the vehicle to suit the handedness and comfort requirements of the driver.