Socially Shared Regulation in Learning: What Does the Triangulation of Different Multimodal Data Sets Tell Us?

• In Event: What Can Multimodal Data Streams Reveal About Students’ Self-Regulated Learning?

Abstract

Objectives
This paper explores what multimodal data can tell about SRL processes in authentic collaborative learning tasks. It demonstrates how multimodal data can be used for collecting and triangulating data about the regulation of learning.
Theoretical framework
If success in self-regulation of learning (SRL) is difficult at the individual level, it is even more difficult when interacting with peers and in teams, known as co-regulation (CoRL) and socially shared regulation (SSRL) (Hadwin et al., 2017). In order to understand how regulation evolves during collaborative learning, research should inform students’ strategic adaptation to the situated challenges and strategic regulation of cognition, motivation and emotion. Multimodal data originate from different sources of learning traces to obtain more robust understanding of the learning process (Ochoa, 2017). For a SRL researcher, it is critical that multimodal data provide different channels to the learning process in order to capture the dynamic, context- and time-sensitive nature of regulation (Zusho, 2017).
Methods and data sources
In order to demonstrate how multimodal data (electrodermal activity (EDA), heart rate, video, audio) can be triangulated with videos, five data examples illustrate possible ways of using multimodal data in learning regulation research (Pijeira-Díaz et al., 2016). The examples report analysis of the high school students aged 16 who worked in groups of three for scientific inquiry by using the WeSpot learning environment.
Substantiated conclusions
Our analyses show that triangulating multimodal data can be used to identify markers that characterize successful (S)SRL and help in understanding about (a) interactions between different facets of regulation (cognition, motivation, emotion) (b) temporality and cyclical processes (planning, enacting strategies, reflecting, adapting) of regulation, and (c) the occurrence and temporality of different types of regulation (SRL, CoRL, SSRL).

Example 1. gives an example of the context around an instance of high arousal. The video data are synchronized with the standardized EDA signal. The episode demonstrates a situation in which the SCL and NSSCR frequency increase almost simultaneously when the group confronts a technical challenge (minute 65), but remain stable during a task understanding challenge (minute 64) and a motivational challenge (minute 68). The screenshot captures an exact moment when the student in the middle expresses his frustration during the technical challenge. Soon also other group members become aware of the challenge which is prohibiting their task completion. Instead of using strategies for regulating emotion, the two other group members start to echo frustration. In sum, a technical challenge in the learning environment leads to an emotional challenge and joint monitoring, which seems to be reflected also as an increase in students’ EDA. Details of the data and conclusions of the current and other examples will be discussed.
Scientific significance
Limited methods exist for making invisible mental regulation processes and the accompanying social and contextual reactions visible. Effort is needed to investigate whether and how these processes can be shown with certain additional data channels. Our data examples contribute to the progress of discussion on how multimodal data collection can advance (S)SRL research.

Authors

• Sanna Järvelä, University of Oulu

• Lars-Erik Malmberg, University of Oxford

• Marta Sobocinski

• Eetu Haataja

• Paul A. Kirschner, Open University of the Netherlands