Sigitov Anton, Hinkenjann André, Kruijff Ernst, Staadt Oliver
Institute of Visual Computing, Bonn-Rhein-Sieg University of Applied Sciences, Sankt Augustin, Germany.
Institute of Computer Science, University of Rostock, Rostock, Germany.
Front Robot AI. 2019 Nov 26;6:128. doi: 10.3389/frobt.2019.00128. eCollection 2019.
Large display environments are highly suitable for immersive analytics. They provide enough space for effective co-located collaboration and allow users to immerse themselves in the data. To provide the best setting-in terms of visualization and interaction-for the collaborative analysis of a real-world task, we have to understand the group dynamics during the work on large displays. Among other things, we have to study, what effects different task conditions will have on user behavior. In this paper, we investigated the effects of task conditions on group behavior regarding collaborative coupling and territoriality during co-located collaboration on a wall-sized display. For that, we designed two tasks: a task that resembles the information foraging loop and a task that resembles the connecting facts activity. Both tasks represent essential sub-processes of the sensemaking process in visual analytics and cause distinct space/display usage conditions. The information foraging activity requires the user to work with individual data elements to look into details. Here, the users predominantly occupy only a small portion of the display. In contrast, the connecting facts activity requires the user to work with the entire information space. Therefore, the user has to overview the entire display. We observed 12 groups for an average of 2 h each and gathered qualitative data and quantitative data in the form of surveys, field notes, video recordings, tracking data, and system logs. During data analysis, we focused specifically on participants' collaborative coupling (in particular, collaboration tightness, coupling styles, user roles, and task subdivision strategies) and territorial behavior. Our results both confirm and extend findings from the previous tabletop and wall-sized display studies. We could detect that participants tended to subdivide the task to approach it, in their opinion, in a more effective way, in parallel. We describe the subdivision strategies for both task conditions. We also detected and described multiple user roles, as well as a new coupling style that does not fit in either category: loosely or tightly. Moreover, we could observe a territory type that has not been mentioned previously in research. In our opinion, this territory type can affect the collaboration process of groups with more than two collaborators negatively. Finally, we investigated critical display regions in terms of ergonomics. We could detect that users perceived some regions as less comfortable for long-time work. The findings can be valuable for groupware interface design and development of group behavior models for analytical reasoning and decision making.
大型显示环境非常适合沉浸式分析。它们为有效的同地协作提供了足够的空间，并允许用户沉浸在数据中。为了在可视化和交互方面为实际任务的协作分析提供最佳设置，我们必须了解在大型显示器上工作期间的群体动态。其中，我们必须研究不同的任务条件将对用户行为产生哪些影响。在本文中，我们研究了任务条件对在墙式显示器上进行同地协作时关于协作耦合和领地行为的群体行为的影响。为此，我们设计了两项任务：一项类似于信息搜寻循环的任务和一项类似于连接事实活动的任务。这两项任务都代表了视觉分析中意义构建过程的基本子过程，并导致不同的空间/显示器使用条件。信息搜寻活动要求用户处理单个数据元素以查看细节。在这里，用户主要只占据显示器的一小部分。相比之下，连接事实活动要求用户处理整个信息空间。因此，用户必须查看整个显示器。我们观察了12个小组，每个小组平均观察2小时，并以调查问卷、现场记录、视频记录、跟踪数据和系统日志的形式收集了定性数据和定量数据。在数据分析过程中，我们特别关注参与者的协作耦合（特别是协作紧密程度、耦合方式、用户角色和任务细分策略）和领地行为。我们的结果既证实了也扩展了先前桌面和墙式显示器研究的结果。我们可以检测到参与者倾向于细分任务，以便他们认为以更有效的方式并行处理任务。我们描述了两种任务条件下的细分策略。我们还检测并描述了多种用户角色，以及一种不属于松散或紧密这两种类别的新耦合方式。此外，我们可以观察到一种在先前研究中未被提及的领地类型。我们认为，这种领地类型会对两个以上协作者的群体协作过程产生负面影响。最后，我们从人体工程学角度研究了关键显示区域。我们可以检测到用户认为某些区域长时间工作不太舒适。这些发现对于群件界面设计以及用于分析推理和决策的群体行为模型的开发可能具有价值。
