Available online at www.sciencedirect.com Int. J. Human-Computer Studies 70 (2012) 156–168 Juggling on a high wire: Multitasking effects on performance Rachel F. Adler a , Raquel Benbunan-Fich b,n a The Graduate Center, City University of New York, United States b Baruch College, City University of New York, United States Received 24 January 2011; received in revised form 15 August 2011; accepted 21 October 2011 Communicated by S. Wiedenbeck Available online 29 October 2011 Abstract In this study, we develop a theoretical model that predicts an inverted-U relationship between multitasking and performance. The model is tested with a controlled experiment using a custom-developed application. Participants were randomly assigned to either a control condition, where they had to perform tasks in sequence, or an experimental condition, where they could discretionarily switch tasks by clicking on tabs. Our results show an inverted-U pattern for performance efficiency (productivity) and a decreasing line for performance effectiveness (accuracy). The results of this study indicate that the nature of the relation between multitasking and performance depends upon the metric used. If performance is measured with productivity, different multitasking levels are associated with an inverted-U curve where medium multitaskers perform significantly better than both high and low multitaskers. However, if performance is measured with accuracy of results, the relation is a downward slopping line, in which increased levels of multitasking lead to a significant loss in accuracy. Metaphorically speaking, juggling multiple tasks is much more difficult while balancing on a high wire, where performance mishaps can have serious consequences. & 2011 Elsevier Ltd. All rights reserved. Keywords: Memory for goals; Multitasking; Task switching; Performance 1. Introduction Is it really possible to juggle while walking on a high wire? Human multitasking has reached new heights these days and its effects on performance are not yet clear. Some researchers underscore the negative impacts of multitask- ing on task performance, particularly when multiple tasks interfere with each other and this juggling produces obstructions or distractions (Aral et al., 2006). Alterna- tively, others argue that multitasking can be conducive to better outcomes by promoting a more productive use of time, allowing ideas to mature or encouraging healthy breaks from complex tasks (Madjar and Shalley, 2008). Despite this debate, juggling multiple tasks with and without technological devices is a common practice at home, at school, at work and even during meetings (Benbunan-Fich and Truman, 2009; Gonzalez and Mark, 2004; Hembrooke and Gay, 2003; Mark et al., 2005; Wasson, 2004). To a great extent, modern technology platforms enable and amplify multitasking. For example, contemporary operating systems in personal computers are designed to accommodate multitasking by allowing users to work on more than one task concurrently. Even current web browser interfaces provide tabbing capabilities to facilitate the performance of simultaneous web-based activities. Given the pivotal role of Information Technology in multitasking, our research is focused on computer-based multitasking behavior. We are particularly interested in the consequences of performing several unrelated computer- based tasks with a single technological device and in a specific period of time. By concentrating exclusively on computer-based tasks, and excluding all other concurrent tasks, such as for example eating or watching TV while using the computer, we seek to shed some light on the patterns and consequences of multitasking behavior. Investigating multitasking behavior is particularly impor- tant for Human–Computer Interaction (HCI) researchers. www.elsevier.com/locate/ijhcs 1071-5819/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhcs.2011.10.003 n Corresponding author. Fax: þ 1 646 312 3351. E-mail addresses: RachelFAdler@gmail.com (R.F. Adler), rbfich@baruch.cuny.edu (R. Benbunan-Fich).