C. Conati, K. McCoy, and G. Paliouras (Eds.): UM 2007, LNAI 4511, pp. 350–354, 2007. © Springer-Verlag Berlin Heidelberg 2007 Conversations Amidst Computing: A Study of Interruptions and Recovery of Task Activity Shamsi T. Iqbal 1 and Eric Horvitz 2 1 Department of Computer Science, University of Illinois, Urbana, IL 61801 2 Microsoft Research, One Microsoft Way, Redmond, WA 98052 siqbal@uiuc.edu, horvitz@microsoft.com Abstract. We present results from a field study investigating the influence of conversations on the multitasking behavior of computer users. We report on several findings, including the timing of the resumption of tasks following conversational interruptions and on the nature and rate of computing activities that are performed concurrently with conversation. Keywords: Interruption, disruption, recovery, conversation, cognitive models. 1 Introduction Interrupting a task at a computer user’s focus of attention often leads to a switching of attention to the source of the interruption [9]. Conversations with other people, including face-to-face conversations, phone calls, and talk through walls have been found to contribute to 15-45% of switches away from the task at a user’s focus of attention [2, 3]. Indeed, becoming engaged in conversational dialog may pose greater disruptions to users than alerts delivered within a computing system because social conventions on personal responsiveness may make it difficult to take the time and actions to prepare for the task switch [9, 10]. We report on a study exploring how conversations occurring during computing tasks affect computing activities. We employed a tool with the ability to log computing activities as well as track the occurrence of conversations by noting the acoustical fingerprint of conversations. The tool logged the start and end of conversations as well as sets of activity variables before, during, and after conversations. We describe a field study undertaken with the tool. 2 Logging Computing Activities and Conversations We developed and fielded a tool named DART (for Disruption Awareness and Recovery Tracker) for studying the influence of interruptions on computing activities. DART was constructed on top of Eve, a set of user and system monitoring components developed at Microsoft Research [4]. DART runs as a background process and logs user activities, including engagement with software applications, switches among windows, and the presence of mouse and keyboard activity. To protect the privacy of study participants, only a subset of keyboard events were recorded. The latter events