The sense of agency during skill learning in individuals and dyads Robrecht P.R.D. van der Wel a,b,⇑ , Natalie Sebanz b,c , Guenther Knoblich b,c a Department of Psychology, Rutgers University, Camden, NJ, United States b Donders Institute for Brain, Cognition, and Behaviour, Center for Cognition, Radboud University Nijmegen, The Netherlands c Department of Cognitive Science, Central European University, Budapest, Hungary article info Article history: Received 23 December 2011 Available online 25 April 2012 Keywords: Sense of agency Joint action Skill learning Forward model Mental causation abstract The sense of agency has received much attention in the context of individual action but not in the context of joint action. We investigated how the sense of agency developed during individual and dyadic performance while people learned a haptic coordination task. The sense of agency increased with better performance in all groups. Individuals and dyads showed a differential sense of agency after initial task learning, with dyads showing a min- imal increase. The sense of agency depended on the context in which the task was first learnt, as transfer from joint to individual performance resulted in an illusory boost in the sense of agency. Whereas the quality of performance related to the sense of agency, the generated forces to achieve the task did not. Our findings are consistent with a predic- tive model account at the perceptual level, such that the sense of agency relies most strongly on sharable perceptual information. Ó 2012 Elsevier Inc. All rights reserved. 1. Introduction People perform an immense variety of actions together with others. For example, they carry heavy objects, play music and team sports, and build houses together. Much research has recently investigated how people plan and coordinate such joint actions. In this article, rather than focusing on the processes that may support successful joint actions, we focus on the expe- riences actors have when engaging in joint actions. We ask how the experience of being in control (i.e., the sense of agency) develops when people learn a new task alone versus together, and whether the sense of agency differs for individual actions versus joint actions. In addition, we examine how developing a sense of agency individually transfers to sensing agency over joint actions, and vice versa. The sense of agency has often been studied in tasks that vary along two dimensions; an actor (a) objectively either causes (or is subjected to) or does not cause (or is not subjected to) the action (Haggard & Tsakiris, 2009; Metcalfe & Greene, 2007) while (b) an alternative cause for the action is either present or absent (e.g., Sato, 2009; Sato & Yasuda, 2005). A central ques- tion in this approach is under what conditions people experience agency over actions they do not produce themselves, or fail to experience agency over actions they do in fact cause. Additional studies have studied the sense of agency in relation to whether movements are actively or passively generated, and in relation to the timing of the feedback accompanying these movements (e.g., Tsakiris, Longo, & Haggard, 2010). But how do people experience agency over actions they intentionally produce together with somebody else? Before we delve into this question, it is useful to first consider potential differences between individual and joint actions in terms of requirements on action planning and control. We will then outline the major theoretical frameworks concerning the sense of agency, and extract predictions from them for the sense of agency for joint actions. 1053-8100/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.concog.2012.04.001 ⇑ Corresponding author at: Department of Psychology, Rutgers University, Camden, NJ, United States. E-mail address: r.vanderwel@rutgers.edu (R.P.R.D. van der Wel). Consciousness and Cognition 21 (2012) 1267–1279 Contents lists available at SciVerse ScienceDirect Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog