!" !# $ ! %&!!! # !" ’! (! &! The interaction techniques that are used in tabletop groupware systems (such as pickanddrop or pantograph) can affect the way that people collaborate. However, little is known about these effects, making it difficult for designers to choose appropriate techniques when building tabletop groupware. We carried out an exploratory study to determine how several different types of interaction techniques (pantograph, telepointers, radar views, draganddrop, and laser beam) affected coordination and awareness in two tabletop tasks (a game and a storyboarding activity). We found that the choice of interaction technique significantly affected coordination measures, performance measures, and preference – but that the effects were different for the two different tasks. Our study shows that the choice of tabletop interaction technique does indeed matter, and provides insight into how tabletop systems can better support group work. : H.5.3 [Information Interfaces and Presentation]: Group and Organization Interfaces— Tabletop groupware, interaction techniques, coordination Tabletop groupware systems combine realworld work surfaces with computational interaction techniques, allowing people to collaborate over digital artifacts while still maintaining a faceto face working style. Tabletop groupware is now becoming more common, and tables have been proposed for a variety of situations, including meetings, design work, and leisure activities. Users of tabletop groupware manipulate computational artifacts using interaction techniques provided by the application. These techniques can vary along several dimensions—some use virtual cursors [4,21], some use physical reaching [20,31,32,28], and others use threedimensional pointing [15,5]. This wide variety suggests that there may also be differences in the collaboration that results – for example, some techniques communicate more information about others’ actions, and so could result in different levels of group awareness. Little research has been carried out, however, to investigate how the choice of interaction technique affects collaboration and coordination on tabletops. Ha and colleagues [9] considered the effects of different input devices (mouse and stylus) on group work, but the effects of interaction techniques are still unclear. Without this knowledge, it is difficult for designers to make informed decisions about what techniques to select. To explore the role that interaction techniques play in tabletop collaboration, we first developed a framework that classifies current techniques according to three dimensions: location of input, location of feedback, and embodiment. We then selected representative techniques from the framework (pantograph, telepointers, radar views, draganddrop, and laser beam) and implemented them in two tabletop groupware systems (a cooperative game and a storyboard application). We then compared the techniques in an experiment where ten groups carried out tasks in the game and storyboard systems, using each of the techniques. Our findings show significant and sometimes unexpected differences across techniques in terms of conflicts, reaching patterns, object transfer, group performance, and user preference. The main result is that no one interaction technique is best for all tasks and all tabletop situations. We found that directtouch techniques such as draganddrop reduce resource conflicts, but at the price of being much less effective when used for reaching distant artifacts. Worldinminiature views such as the radar view were surprisingly effective for a game task, although in a design task the radar did not provide enough awareness information to be successful. There were also strong differences in the ways that users viewed the pantograph, telepointers, and laserbeam techniques, even though these techniques are quite similar on the surface. This paper makes two contributions: first, the framework provides dimensions on which techniques can be compared and evaluated, and differentiates between a large number of interaction techniques; second, the study provides empirical evidence on how those differences result in changes to group coordination and collaboration in tabletop tasks. The study also provides initial recommendations about where and when the different techniques will be effective, and builds a foundation for the design of novel techniques that can support a wider range of table settings and tasks. Group awareness is the understanding of others’ presence, locations, and current activities in the shared workspace. A good sense of awareness can help people simplify communication, find opportunities to help one another, and coordinate activities and access to shared resources [14]. Group awareness in tabletop collaboration is maintained through three main mechanisms: territoriality, artifact feedthrough, and consequential communication. Territoriality is a useful organizing principle for interaction on tabletops. It is based in the physical realities of a person’s presence and reach, and the physical extents of the table surface. Scott et al. [25] found that groups divide the space on the tabletop into three territories: personal, group, and storage. These different regions result in different types of shared and individual activity, requiring that awareness information be interpreted in light of the territory in which it was produced. Feedthrough is the public feedback produced by artifacts when they are manipulated by an interaction technique. Although the feedback is usually intended for the person performing the action, it can also inform others who are watching [6]. In addition, feedthrough can be manipulated to provide better awareness information: for example, delete actions can be drawn out to make them more obvious [8]. One important type of feedthrough on tables is artifact orientation. Collaborators use the orientation of A 0.5 inch (1.18 cm) space must be left at the bottom of the left column on the first page (i.e., blank the text from this box before submitting). 191 7 Graphics Interface Conference 2007 28-30 May, Montréal, Canada Copyright held by authors. Permission granted to CHCCS/SCDHM to publish in print form, and ACM to publish electronically.