We discuss a two-handed user interface designed to support three-dimesional neurosurgical visualization. By itself, this system is a “point design,” an example of an advanced user interface technique. In this work, we argue that in order to understand why interaction techniques do or do not work, and to suggest possibilities for new techniques, it is important to move beyond point design and to introduce careful scientific measurement of human behavioral principles. In particular, we argue that the common-sense viewpoint that “two hands save time by working in parallel” may not always be an effective way to think about two-handed interface design because the hands do not necessarily work in parallel (there is a structure to two-handed manipulation) and because two hands do more than just save time over one hand (two hands provide the user with more information and can structure how the user thinks about a task). To support these claims, we present an interface design developed in collaboration with neurosurgeons which has undergone extensive informal usability testing, as well as a pair of formal experimental studies which investigate behavioral aspects of two-handed virtual object manipulation. Our hope is that this discussion will help others to apply the lessons in our neurosurgery application to future two-handed user interface designs.