Mirroring multiple agents: motor resonance during action observation is modulated by the number of agents

Social Cognitive and Affective Neuroscience, 2016, 1422–1427 doi: 10.1093/scan/nsw059 Advance Access Publication Date: 25 April 2016 Original article Emiel Cracco,1 Lize De Coster,1 Michael Andres,1,2 and Marcel Brass1 1 Department of Experimental Psychology, Ghent University, Henri Dunantlaan 2, Ghent, 9000, Belgium and Psychological Sciences Research Institute, Université Catholique De Louvain, Place Cardinal Mercier 10, Louvain-la-Neuve, 1348, Belgium 2 Correspondence should be addressed to Department of Experimental Psychology, Ghent University, Henri Dunantlaan 2, Ghent, 9000, Belgium. E-mail: Abstract Although social situations regularly involve multiple persons acting together, research on the mirror neuron system has focused on situations in which a single agent is observed. Therefore, the goal of the current study was to explore the role of the mirror mechanism in situations involving multiple agents. Specifically, we used transcranial magnetic stimulation (TMS) to investigate whether mirror activation is modulated by the number of observed agents. Based on group contagion research, we hypothesized that multiple agents would provide a stronger trigger to the motor system and would therefore produce a stronger mirror response than a single agent. Participants observed movements performed by a single hand or by two hands while TMS was applied to the primary motor cortex. The results confirmed that activation in the motor system was stronger for two hands. This suggests that input to the motor system increases as the number of agents grows. Relating back to group contagion, our study suggests that groups may be more contagious simply because their actions resonate louder. Given that the mirror mechanism has been linked to a variety of social skills, our findings additionally have important implications for the understanding of social interaction at the group level. Key words: TMS; action observation; imtation; mirror neuron Since their discovery in the monkey brain (Gallese et al., 1996; Rizzolatti et al., 1996), mirror neurons have been studied extensively in the literature. As a result of this research, it is now well established that a shared system for perception and action does not only exist in monkeys but can be found in humans as well (Molenberghs et al., 2012; Rizzolatti and Sinigaglia, 2010). A useful technique to study the human mirror neuron system is transcranial magnetic stimulation (TMS). Numerous studies have now shown that the application of TMS to the primary motor cortex increases corticospinal excitability of the muscles involved in executing the observed movement (Fadiga et al., 1995, 2005; Maeda et al., 2002; Naish et al., 2014). Furthermore, it has been shown that these effects rely on input from regions within the frontoparietal mirror neuron network such as the premotor and intraparietal cortex (Avenanti et al., 2007, 2013; Koch et al., 2010; Catmur et al., 2011; Enticott et al., 2012). Interestingly, it has been argued that this mirror mechanism facilitates social interaction because it allows individuals to obtain first-person knowledge on the actions of others (Rizzolatti and Fabbri-Destro, 2008). In line with this argument, studies have shown that motor resonance does not only reflect the kinematics (Maeda et al., 2002), but also the intention (Cattaneo et al., 2007; Tidoni et al., 2013), the goal (Cattaneo et al., 2009) and the outcome (Aglioti et al., 2008) of an observed action. This is further supported by evidence suggesting that motor activation is facilitated when an observed action is produced by another person but suppressed when it is produced by oneself (SchützBosbach et al., 2006). Received: 22 September 2015; Revised: 12 April 2016; Accepted: 21 April 2016 C The Author (2016). Published by Oxford University Press. For Permissions, please email: V 1422 Mirroring multiple agents: motor resonance during action observation is modulated by the number of agents E. Cracco et al. 1423 Stimuli and apparatus The experiment was programmed with Tscope (Stevens et al., 2006). Stimuli consisted of frames that were extracted from video clips (Figure 1). The stimuli (1010  568 pixels) depicted two different male right hands. The hands were presented next to each other on a blue background from a first person perspective. The position (left/right) of the hands on the screen was counterbalanced. To produce an illusion of movement, a picture of the hands in their end posture was superimposed on a picture of the hands in their starting posture (see also: Catmur et al., 2007, 2011). The hands could either not move or abduct the index or little finger. Importantly, when both hands made a movement they always performed the same movement. As a result, the experiment included seven possible end postures: Static-Static, IndexFinger-Static, Static-IndexFinger, LittleFinger-Static, Static-LittleFinger, IndexFinger-IndexFinger and LittleFinger-LittleFinger. Task and procedure The experiment took about 45 min and consisted of four blocks of 105 trials each. All end postures were presented an equal number of times in each block in a random order. The experimental task required participants to monitor a cue (N, W or P) appearing at the top of the screen simultaneously with the presentation of the end posture. Participants were instructed to abduct the index finger when W (10%) was presented and to abduct the little finger when P (10%) was presented. When N (80%) was presented, no action was required. The movement (W or P) and no-movement (N) trials were distributed equally among the seven possible end postures. On the movement trials, the cue was chosen randomly so that W and P appeared an equal number of times. This resulted in 14% neutral movement trials, 44% congruent movement trials and 42% incongruent movement trials. The rationale behind the task was twofold. First, we wanted to maintain the attention of the participants. Second, we wanted to ensure that the relevant motor representations remained active throughout the experiment. Note that analyses were restricted to the N trials. As a result, motor execution processes could not influence the results. Each trial started with a picture of the hands in their starting posture and a fixation cross at the top of the screen for 500 ms. The hands were then presented in their end posture for a duration of 1000 ms together with the cue. A TMS pulse was delivered on every trial. The pulse was delivered randomly at 300, 400 or 500 ms after the presentation of the end posture. The pulses were distributed equally among the three stimulation moments. The trial ended with the presentation of a black screen for a jittered duration of 4000, 5000 or 6000 ms. Materials and methods Participants TMS and electromyography Thirty-six right-handed males (Mage ¼ 22.25, SDage ¼ 3.06) participated in the study in exchange for 25 euros. However, as described below, two participants were excluded from (...truncated)