Summary:The objective of this study was to compare driving performance whilst talking on a hands-free mobile (cellular) telephone with performance during conversations with "considerate" and "inconsiderate" passengers. Using the Leeds Advanced Driving Simulator, participants were asked to drive through a road containing four driving scenarios: (1) car following along a straight road section, (2) car following along a curved section of road, (3) a braking event, and (4) a coherence event. A working memory digit recall and sentence verification task were used to simulate conversation in three conversation conditions: (1) "considerate passenger," where the experimenter asked drivers to responsd to the working memory task before and after a driving event, (2) "inconsiderate passenger," where the experimenter asked drivers to responsd to the working memory task throughout a driving event, and "3) "mobile phone task," which involved digital presentation of the working memory task via a mobile telephone throughout the driving events. A silent condition was also used as control. The effect of the three conversation conditions on driving performance was the same during the simple car following scenarios. However, talking to an "inconsiderate" passenger was found to be as disrupting as a mobile phone conversation, and different from conversing with a considerate passenger, during the braking and coherence conditions. Therefore, the high workload imposed by conversation was only detrimental during the more difficult driving conditions, when demand for central attentional resources from both tasks was at its highest.