Activation patterns during action observation are modulated by context in mirror system areas Pascal Molenberghs a, b, ⁎, Lydia Hayward b , Jason B. Mattingley a, b , Ross Cunnington a, b a The University of Queensland, Queensland Brain Institute, Qld 4072, Australia b The University of Queensland, School of Psychology, Qld 4072, Australia abstract article info Article history: Received 21 April 2011 Revised 13 July 2011 Accepted 23 July 2011 Available online 5 August 2011 Keywords: fMRI Perception of action Mirror system Action understanding Premotor cortex Inferior parietal lobule The role of the mirror system in action understanding has been widely debated. Some authors have suggested that the mirror system plays an important role in action understanding (Rizzolatti and Sinigaglia, 2010), whereas others have claimed that direct evidence to support this view is lacking (Hickok, 2009). If mirror neurons have an active role in action understanding rather than passive visuomotor transformation during action observation, they should respond differently to the observation of actions depending on the intentions of the observer. In this fMRI study, twenty participants observed identical actions under different instruction contexts. The task was either to understand the actions, identify the physical features of the actions, or passively observe the actions. A multi-voxel pattern analysis revealed unique patterns of activation in ventral premotor cortex and inferior parietal lobule across the different contexts. The results suggest that ventral premotor and inferior parietal areas respond differently to observed actions depending on the mindset of the observer. This is consistent with the view that these regions do not merely process observed actions passively, but play an active role in action understanding. Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved. Introduction According to the ‘direct-matching’ hypothesis, people under- stand actions by directly mapping the visual representation of an action to an equivalent motoric representation of the action in their own motor system. The ‘mirror system’ has been suggested to be the underlying network responsible for this visuomotor transfor- mation (Rizzolatti et al., 2001). Mirror neurons fire both when an action is performed, and when a similar or identical action is passively observed (Rizzolatti and Craighero, 2004). These neurons were discovered during single-cell recordings in monkey areas F5, PFG and AIP (Gallese et al., 1996, 2002; Rizzolatti and Sinigaglia, 2010). Mirror neurons have also been described in humans (Mukamel et al., 2010), and it has been hypothesized that a similar parieto-frontal ‘mirror’ mechanism is present in the human brain (for recent reviews see Cattaneo and Rizzolatti, 2009; Fabbri- Destro and Rizzolatti, 2008; Iacoboni and Dapretto, 2006; Rizzolatti and Fabbri-Destro, 2010; Molenberghs et al., 2011). Several different brain mechanisms might underpin action understanding (Brass et al., 2007; Saxe, 2005, 2006), but Rizzolatti and Sinigaglia (2010) claim that the parieto-frontal ‘mirror’ mecha- nism is the only system that allows an individual to understand the action of others ‘from the inside’ by allowing the observer a first- person grasp of the motor goals and intentions of other individuals. According to this view, mirror neurons in higher motor areas play an important role in the cognitive understanding of actions (Blakemore and Decety, 2001; Rizzolatti and Sinigaglia, 2010). A number of single- cell studies in monkeys have provided evidence of a role for mirror neurons in action understanding. Umilta et al. (2001) found that a subset of mirror neurons in ventral premotor cortex responds to goal- directed actions even when the final part of the action is hidden from view. The authors concluded that the missing parts of the actions were internally generated in the monkey brain and therefore involved in action recognition. Kohler et al. (2002) found that some mirror neurons in the ventral premotor cortex not only respond to the observation and execution of an action but also to the sound of that action, supporting the view that these neurons code the goal of an action, irrespective of input modality. In a more recent single-cell study, Umilta and colleagues (2008) trained monkeys to grasp an object with ‘normal’ and ‘reverse’ pliers. With the normal pliers monkeys closed their hand to grasp the object whereas with the reverse pliers they opened their hand to achieve the same goal. Despite the different actions involved, mirror neurons in ventral premotor cortex were active in both conditions consistent with the coding of a common goal (grasping an object). These studies suggest that mirror neurons in monkeys do not simply represent the movement characteristics of an action, but in fact encode the goal of that action. In humans, fMRI studies have revealed very similar activations within the mirror system in response to robotic (Gazzola et al., 2007) and tool actions (Peeters et al., 2009), suggesting that this NeuroImage 59 (2012) 608–615 ⁎ Corresponding author. Fax: + 61 7 3365 4466. E-mail address: p.molenberghs@uq.edu.au (P. Molenberghs). 1053-8119/$ – see front matter. Crown Copyright © 2011 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2011.07.080 Contents lists available at ScienceDirect NeuroImage journal homepage: www.elsevier.com/locate/ynimg