اقدامات ما در ذهن من: تصویرسازی حرکتی از اقدام مشترک

How do people imagine performing actions together? The present study investigated motor imagery of joint actions that requires integrating one's own and another's part of an action. In two experiments, individual participants imagined jumping alone or jointly next to an imagined partner. The joint condition required coordinating one's own imagined actions with an imagined partner's actions to synchronize landing times. We investigated whether the timing of participants' own imagined jumps would reflect the difference in jump distance to their imagined partner's jumps. The results showed that participants' jump imagery was indeed modulated to achieve coordination with an imagined task partner, confirming prior findings from a performance task. Moreover, when manipulating both target distance and target size, the same violation of Fitts' law reported for individual jumping was present in imagery of joint jumping. These findings link research on motor imagery and joint action, demonstrating that individuals are able to integrate simulations of different parts of a joint action.

Imagining a simple action such as pouring coffee into a cup is in many respects similar to actually performing that action except that the observable motor output is lacking. Jeannerod described motor imagery as the “ability to generate a conscious image of the acting self” (Jeannerod, 2004; p. 379) and proposed that many of the principles underlying action performance also hold in action imagery (Jeannerod, 1995 and Jeannerod, 2004). This proposal has sparked a whole line of research that investigated what is common between covert (internally simulated) action and overt (actually performed) action. Similarities in neurophysiological activity when planning, performing and imagining actions indicate that these phenomena are governed by overlapping processes and brain networks (Decety and Grèzes, 2006, Dietrich, 2008, Grèzes and Decety, 2001 and Rizzolatti and Sinigaglia, 2010; but see Dietrich, 2008 for a critical view). In particular, imagined and to-be-performed actions might be represented in a common motor format (Jeannerod, 1995 and Prinz, 1997), thereby relying on internal forward models that predict the (imagined) outcome of an action (Grush, 2004, Blakemore and Frith, 2005, Wilson and Knoblich, 2005 and Wolpert et al., 2003). Whereas researchers have intensively studied motor imagery of individual actions (e.g., Guillot and Collet, 2005 and Jeannerod, 2004), motor imagery of joint actions has not been addressed. However, investigating imagery of joint action can help us to better understand the mechanisms underlying motor simulation. The reason is that in order to imagine a coordinated joint action it is neither sufficient to simulate one's own action nor is it sufficient to simulate the other's action. Rather it is also necessary to integrate these two action simulations. This becomes clear when one considers that joint action often requires that two or more individuals adapt their actions in space and time to what the other is doing (Clark, 1996 and Sebanz et al., 2006). Examples for such joint actions range from carrying a heavy object with a friend to passing a basketball to a team-mate or dancing a tango together. Importantly, co-actors need to represent not only their own and a partner's part of a joint action, but also the shared goal resulting from their combined actions (Vesper, Butterfill, Knoblich, & Sebanz, 2010). For performance, it has been suggested that joint action coordination toward a shared goal is to a large extent achieved by internal simulations that allow co-actors to predict their own and their partner's actions using their own motor system (Keller, 2012 and Wolpert et al., 2003). We propose that the same simulation processes that support the planning and execution of joint action also support imagery of joint action. Especially when coordinating actions with others, motor simulations of one's own and a partner's action parts need to be integrated to plan one's own action with respect to achieving the shared goal. Although there is growing evidence that different motor simulations can run in parallel (Hamilton et al., 2004 and Kourtis et al., 2013), there is hardly any evidence that motor simulations can be integrated to simulate different components of a joint action. The current study attempted to test this assumption using motor imagery. Imagining performing actions is a pure form of motor simulation as imagery is not subject to any sensorimotor or perceptual influences that are present when movement is actually performed (cf. Schmidt and Richardson (2008)). If people were able to engage in imagery of joint action that constrained their own as well as their partner's action parts in the same way as during actual joint action planning and performance, this would provide evidence for an integration of motor simulations of one's own and others' actions. Previous research on individual motor imagery has compared how people actually perform actions to how they imagine performing the same actions (e.g., Jeannerod, 1995 and Jeannerod, 2004). A highly consistent finding in such studies is that constraints present in performance also govern motor imagery. For example, when people imagine walking a specific distance then the time their movement takes is similar to actually walking that distance (Decety, Jeannerod, & Prablanc, 1989). Moreover, if a to-be-imagined action is more difficult reported movement times increase systematically. In one study where individual participants were asked to imagine walking through doors of varying width, their reports of the imagined movement time scaled as a function of the distance toward the door and its width (Decety & Jeannerod, 1995), thereby complying with the speed-accuracy trade-off known as Fitts' law (Fitts, 1954). Similarly, the same biomechanical constraints determining in which way people lift objects were found in their self-reports of imagining grasping the object (Johnson, 2000). Based on these previous findings, the present study asked whether the constraints imposed by the requirement to coordinate with another person would influence imagery in the same way as when two people perform coordinated actions together. Two experiments tested whether behavioral effects previously observed in joint action coordination could be observed in a motor imagery task where participants imagined both parts of the joint action. To this end, we adapted and extended an existing joint action task (Vesper, van der Wel, Knoblich, & Sebanz, 2013) and asked participants to imagine coordinating their own action with an imagined partner. If participants' action imagery resembled actual performance this would demonstrate that participants take the same aspects of another person's task or action into account when imagining interpersonal coordination. This, in turn, would support the assumption that they can engage in an integrated motor simulation of their own and another's part of a joint action.