The motor cortex is causally related to predictive eye movements during action observation Claudia Elsner a,n , Alessandro D’Ausilio b , Gustaf Gredeb ¨ ack a , Terje Falck-Ytter a,c , Luciano Fadiga b,d a Department of Psychology, Uppsala University, Uppsala, Sweden b Robotics, Brain and Cognitive Sciences Department, Italian Institute for Technology, Genova, Italy c Center of Neurodevelopmental Disorders at Karolinska Institutet (KIND), Karolinska Institutet, Stockholm, Sweden d DSBTA – Section of Human Physiology, University of Ferrara, Ferrara, Italy article info Article history: Received 29 September 2012 Received in revised form 27 November 2012 Accepted 14 December 2012 Available online 23 December 2012 Keywords: Prediction Eye movement TMS Biological motion Motor cortex Mirror neuron Direct matching abstract We examined the hypothesis that predictive gaze during observation of other people’s actions depends on the activation of corresponding action plans in the observer. Using transcranial magnetic stimulation and eye-tracking technology we found that stimulation of the motor hand area, but not of the leg area, slowed gaze predictive behavior (compared to no TMS). This result shows that predictive eye movements to others’ action goals depend on a somatotopical recruitment of the observer’s motor system. The study provides direct support for the view that a direct matching process implemented in the mirror-neuron system plays a functional role for real-time goal prediction. & 2012 Elsevier Ltd. All rights reserved. 1. Introduction When observing other people interacting with the environ- ment, humans systematically fixate action goals ahead of time. Such predictive eye movements emerge early in human develop- ment and continue to play a crucial role for collaboration and competition throughout life (Falck-Ytter, Gredeb ¨ ack, & von Hofsten, 2006; Flanagan & Johansson, 2003; Kanakogi & Itakura, 2011). To date, the mechanisms behind predictive goal-directed eye movements in action observation are unknown. According to a motor-cognitive view, goal-directed eye move- ments reflect the activation of effector-specific action plans in the observer. On a general level, we know that others’ action percep- tion is dependent on motor system activation (D’Ausilio et al., 2009; Pobric & Hamilton, 2006; Stadler et al., 2012; Urgesi, Calvo- Merino, Haggard, & Aglioti, 2007). More specifically, the idea that action plans generate predictions in real time receives support from behavioral data demonstrating similar patterns of eye movements during execution and observation (Flanagan & Johansson, 2003). In line with this finding, several studies have demonstrated that observers take advantage of their own motor abilities (Aglioti, Cesari, Romani, & Urgesi, 2008) and specific motor cues like hand pre-shaping (Ambrosini, Constantini, & Sinigaglia, 2011) to predict other people’s actions. In addition, it was shown that predictive gaze behavior is temporarily affected when observers’ hands are tied behind their backs while obser- ving reaching-grasping but not touching actions, suggesting that effective action prediction also depends on the observer being in a position to perform the same actions or not (Ambrosini, Sinigaglia, & Constantini, 2012). Moreover, Constantini, Ambrosini, and Sinigaglia (2012a) found that object reachability influences obser- vers’ predictive gaze performance, i.e. observers are more pre- dictive when an observed agent is able to reach for a goal object than when an object is out of the agent’s reach. Besides, they demonstrated that compatibility between observed and executed prehension selectively impacts gaze behavior (Constantini, Ambrosini, & Sinigaglia, 2012b), showing that the readiness of the observers’ own motor representations effects proactive eye movements as well. Despite plenty of studies indicating that action prediction is driven by a recruitment of the observer’s own corresponding action plans, available data are not able to rule out alternative hypotheses, e.g. that predictive eye movements are driven by inferential processes, implemented by non-motor brain areas (Eshuis, Coventry, & Vulchanova, 2009; Falck-Ytter, 2012). Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/neuropsychologia Neuropsychologia 0028-3932/$ - see front matter & 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropsychologia.2012.12.007 n Corresponding author. Tel.: þ46 184712157. E-mail addresses: claudia.elsner@psyk.uu.se (C. Elsner), Alessandro.Dausilio@iit.it (A. D’Ausilio). Neuropsychologia 51 (2013) 488–492