This article aims to contribute to the still scarce evidence available on the social nature of metacognitive regulation of joint efforts in a collaborative problem-solving process, and to offer systematic means to operationalise and analyse shared regulation. The need and the rationale for studying social regulation are based on the view that the group is a social system (Vauras, Salonen, & Kinnunen, 2008), a qualitatively different entity from individuals working side by side, but on their own (Salomon & Globerson, 1989). Social regulation cannot be reduced to the group members’ individual characteristics such as self-regulatory activities, rather inter-relational characteristics and functioning are needed in order to understand group dynamics as a complex situational interplay across different systemic levels (Volet, Vauras, & Salonen, 2009). Both the self and social forms of regulation (other- and co-regulation) are needed in order to understand regulation of actual collaborative learning processes. In other words, the manifestation of inter-individual metacognition is not equivalent to individual metacognition, and should be conceptualised differently (Iiskala et al., 2004 and Vauras et al., 2003). In addition to the generic term “co-regulation” (McCaslin, 2009, Volet et al., 2009 and Volet et al., 2009), we have introduced the concept “socially shared metacognition” (or “shared regulation”) to refer to the consensual monitoring and regulation of joint cognitive processes in demanding collaborative problem-solving situations (Iiskala et al., 2004 and Vauras et al., 2003). Socially shared metacognition can be considered the most profound social mode of regulation, because it refers to individuals’ metacognitive processes that operate as a genuine social entity, aimed at a single objective, that is, the fully shared goal of the activity (Volet, Vauras, et al., 2009).
However, despite growing agreement on understanding regulation as both an individual and a social process (see, e.g., McCaslin, 2009, Nolen and Ward, 2008 and Volet et al., 2009), empirical evidence on social regulatory processes pertaining to higher order learning is scarce and insufficient compared to the extensive conceptual analyses and empirical studies on individual metacognition for more than 30 years, starting from the seminal work of John Flavell and Ann Brown in the 1970s and 1980s (Brown, 1978, Brown, 1987 and Flavell, 1976). In these studies, sophisticated operationalisations and methods to study individual metacognition have been introduced. However, the study of metacognition as a social phenomenon is characterised by a lack of clear operationalisation and methods of data analysis. The case studies of Iiskala et al. (2004) and Vauras et al. (2003) have clearly demonstrated that it is possible to distinguish social regulation processes from individual ones in collaborative learning contexts. To understand the functions and fluctuations of self- and social regulation in collaborative contexts, reliable methods to identify and analyse socially shared metacognition within large data sets of interactions are urgently needed. The main aim of the present study was to investigate whether and how this can be done, and to present a detailed analysis of the functions and foci of socially shared metacognition. This was studied by analysing high-achieving dyads’ collaborative mathematical problem-solving processes.
