Speech versus visual-manual interfaces in trucks: effects on driver distraction, user acceptance, and perceived efficiency
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Truck drivers often have a tight time schedule and therefore need to carry out several in-vehicle tasks while driving, such as making phone calls, writing down information and navigating to new places. Performing these tasks using a visual-manual interface can impose visual distraction which has been shown to lead to safety-critical events on the roads. Instead of using a visual-manual interface, a speech interface could be a safer alternative if designed properly. However, the cognitive load demanded by speech interfaces and the connection between cognitive load and driving behaviour is not fully understood. In this study, a speech interface and its visual-manual counterpart were evaluated and compared in terms of visual distraction, cognitive load and user efficiency and perceived acceptance. Eye tracking was used to measure visual distraction. The measurements used for cognitive load were the Tactile Detection Response task (TDRT) and the Driving Activity Load Index (DALI). Perceived acceptance and efficiency were measured using the System Usability Scale (SUS), the Subjective Assessment of Speech Systems (SASSI) and semi-structured interviews.
The conclusions were that (1) the speech interface was less visually distracting than the visual-manual counterpart, (2) the speech interface was less cognitively demanding than the visual-manual interface, especially in the navigation task, (3) the speech interface was safer to use while driving compared to the visual-manual interface and (4) the speech interface had higher user acceptance and perceived efficiency than the visual-manual interface. Further research should investigate the connection between cognitive load and driving behaviour, such as lane keeping and brake response time, by employing a variety of speech tasks with various complexity as well as including speech interfaces entirely free from visual demand. The focus should be on the difference between baseline driving and speech interaction, as opposed to speech interaction and visual-manual interaction.
Place, publisher, year, edition, pages
2016. , 61 p.
IdentifiersURN: urn:nbn:se:liu:diva-135490ISRN: LIU-IDA/KOGVET-A–17/001–SEOAI: oai:DiVA.org:liu-135490DiVA: diva2:1082143
Subject / course
Silvervarg, Annika, Lektor
Jönsson, Arne, Professor