Mobile phone use at the wheel has a proven negative impact on driving. This paper aims to assess whether using an answerphone instead of a normal phone reduces this negative impact. The answerphone makes communication asynchronous and splits it into three disconnected and successive phases: interaction with the voice interface, listening to the message and answering, which have been evaluated separately. The experiment was conducted on a fixed base simulator, with 30 participants (half male and female, aged from 18 to 50 years, driving at least 5000 km per year). The results show better scores for correct responses to stimuli for answerphone communications than for phone communications, although response times were higher in both communication conditions than in the driving alone condition. When the three phases of answerphone use were compared, interacting with the answerphone and listening to the message were found to be significantly less disturbing than answering, in terms of correct responses and response times. By making the conversation asynchronous, the answerphone avoids direct interaction between driver and caller. The fact that communication is under the driver's control allows him/her to pace the interaction better. Lastly, splitting up the conversation into different phases decreases the overall task difficulty.
