The number of wireless phone subscribers in the U.S. is constantly growing. Studies have shown that use of wireless phones while driving contributes to crashes. Efforts to pass legislation allowing only hands-free wireless phone use while driving are widespread and based on the assumption that a hands-free interface is safer than a hand-held one. The National Highway Traffic Safety Administration conducted a driving simulator study to examine the effects of phone interface type on driving performance and drivers' ability to perform phone tasks. Participants' phone interface preferences were also recorded. Fifty-four participants drove a simulated freeway route with each of three phone interfaces: hand-held, headset hands-free (voice dialing and headset), and voice dialing hands-free (voice dialing with cradle-mounted speaker phone). Although post-drive questionnaire results show that participants rated the hand-held interface to be most difficult to use, this interface was associated with the fewest dialing errors and significantly faster dialing times than the two hands-free interfaces. Drivers answered the phone more quickly when using the voice dialing hands-free phone interface than when using the hand-held or headset hands-free interfaces. Younger drivers answered the phone significantly more quickly than the older drivers. Hang up times were significantly faster in the voice dialing hands-free condition and were slowest in the headset hands-free condition. Conversation task performance did not differ as a function of phone interface. Thus, although drivers considered them easier to use than hand-held phone interfaces, hands-free interfaces were more time-consuming to interact with while driving. Further analyses of these data will investigate how other aspects of phone task performance are related to eye glance behavior and driving performance.