Abstract:

Joint action is a common form of interpersonal collaboration in daily life. Successful joint action depends upon individuals' co-representation ability driven by a common goal. Also, an individual's monitoring and prediction of the actions of own and others based on this common representation ability ensures the completion of the joint task. However, researchers have different ideas on how individuals represent their own and others' behaviors during joint action. The process and the mechanism of interpersonal collaboration are still unclear. Hence, from the perspective of representation, this paper integrates the cognitive mechanism of interpersonal collaboration in joint action. Three main components of the cognitive mechanism are common goals, co-representation, and prediction. The common goal is a prerequisite for the accomplishment of interpersonal collaboration. Co-representation is the foundation of interpersonal collaboration. And during interpersonal coordination in joint action, predicting others' behavior is a core cognitive ability that triggers by co-representation capacity. Next, the authors discussed two neural mechanisms, the mirror neuron system and the theory of mind network, which play important roles in the representation process of interpersonal collaboration in joint action. It has been reported that the mirror neuron system is closely related to the goal of the acts and co-representation, and the theory of mind network is associated with co-representation and prediction.
Besides, previous theoretical research shed light on the process and content of representation during interpersonal co-representation at the theoretical level. For instance, spatial coding theory explains the interpersonal process of representation in terms of spatial dimension. Co-representation theory views the representation process in the social dimension. Reference coding theory explains the representation process along multiple perceptual dimensions (e.g., color, emotion, personal characteristics). While these theories have provided different explanations of why representations occur during interpersonal collaboration, all of these theories have some limitations in clarifying the process of representation. Up to now, no unified representation models have been provided. Therefore, this study proposes a multidimensional representation model of interpersonal collaboration based on previous theoretical explanations and empirical findings focusing on the factors influencing interpersonal co-representation.
The fundamental idea of this model is that the interpersonal representation phenomenon in joint action can be explained in three dimensions: spatial representation, embodied representation, and social representation. During the process of interpersonal coordination, individuals represent motor goals (common goals) in spatial dimension at the beginning; then they represent motor plans in both spatial and embodied dimensions; next, individuals execute their actions based on spatial representation and observe others' actions and predict others' execution by the embodied representation and social representation. Finally, embodied representation helps people integrate their behaviors with those of others and facilitates task monitoring and action feedback. There are close connections among spatial, embodied, and social representation dimensions. These three dimensions are at the same level and overlap with each other. Firstly, from the internal perspective, embodied dimension representation promotes information representation in the spatial dimension. Secondly, from a perspective of inter-individual interaction, representation involving embodied dimension helps individuals to represent social information. Finally, there is an interplay between spatial representation within individuals and social representation among people. On the one hand, spatial representation should consider the influence of representation in the social dimension. On the other hand, the clarity of information in the spatial dimension affects the representation results of social information. This model can help us understand and explain the process and content of the representation behind the collaborators' behaviors in joint action. This model can also promote the future advancement of research on intelligent human-computer interaction and improve the user experience. Meanwhile, research surrounding this model can be directly applied to designing and optimizing collaborative operation systems and ultimately provide a scientific basis for designing human-intelligent collaborative operation systems.

Key words: joint action, co-representation, multi-representative model, interpersonal collaboration, human- computer interaction