Beyond the self: The role of co‐regulation in medical students’ self‐regulated learning

Abstract Context Medical students are expected to self‐regulate their learning within complex and unpredictable clinical learning environments. Research increasingly focuses on the effects of social interactions on the development of self‐regulation in workplace settings, a notion embodied within the concept of co‐regulated learning (CRL). Creating workplace learning environments that effectively foster lifelong self‐regulated learning (SRL) requires a deeper understanding of the relationship between CRL and SRL. The aim of this study was therefore to explore medical students’ perceptions of CRL in clinical clerkships and its perceived impact on the development of their SRL. Methods We conducted semi‐structured interviews with 11 purposively sampled medical students enrolled in clinical clerkships at one undergraduate competency‐based medical school. Data collection and analysis were conducted iteratively, informed by principles of constructivist grounded theory. Data analysis followed stages of open, axial and selective coding, which enabled us to conceptualise how co‐regulation influences the development of students’ self‐regulation. Results Data revealed three interrelated shifts in CRL and SRL as students progressed through clerkships. First, students’ CRL shifted from a focus on peers to co‐regulation with clinician role models. Second, self‐regulated behaviour shifted from being externally driven to being internally driven. Last, self‐regulation shifted from a task‐oriented approach towards a more comprehensive approach focusing on professional competence and identity formation. Students indicated that if they felt able to confidently and proactively self‐regulate their learning, the threshold for engaging others in meaningful CRL seemed to be lowered, enhancing further development of SRL skills. Conclusions Findings from the current study emphasise the notion that SRL and its development are grounded in CRL in clinical settings. To optimally support the development of students’ SRL, we need to focus on facilitating and organising learners’ engagement in CRL from the start of the medical curriculum.


| INTRODUC TI ON
Self-regulated learning (SRL) is considered a core competence of physicians and one that is essential to the safeguarding of patient care. 1,2 Many medical curricula therefore support medical students in developing SRL skills. Self-regulated learning is generally described as a cyclical process, often triggered by the formulating of goals and the subsequent employment of strategies to achieve and monitor advancement towards those goals, followed by engagement in reflection and the formulation of new learning goals. 2,3 Research findings, however, indicate that students often struggle to regulate their learning in clinical learning environments as a result of the unpredictable, dynamic and messy nature of clinical workplace settings. 4,5 As early as 1989, the importance of environmental constraints and affordances to SRL was documented. 6 Since then, the notion that medical students' SRL is context-dependent and socially derived has steadily gained momentum. 7,8 Learning climates of hospital departments, available learning opportunities and social interactions with others seem to influence students' SRL, 9 and medical trainees variably use SRL strategies, depending on social and contextual factors. 10 A study by Berkhout and colleagues, for example, showed that novice students interact with peers and residents to help them formulate learning goals. 11 Elsewhere, paediatric residents have been found to benefit from interacting with supervisors when pursuing goals. 12 Similarly, surgical residents actively employ different strategies to engage their supervisors in the monitoring of performance during surgery. 13 The potential impact of social interactions on medical students' SRL is embodied in the concept of co-regulated learning (CRL). In CRL, learners regulate their cognitions, motivation and behaviour together with other individuals in the environment. 14 Essential to CRL are social interactions between students and others in their networks, through which learning processes, such as SRL development, are mediated or distributed. 15 The concept of CRL reflects sociocultural learning theories that emphasise that learners are continuously influenced by their environment when simultaneously co-producing and co-creating the contextual surroundings in which they learn and work. 16 By jointly engaging in learning activities, students and coworkers in health care settings mediate one another's metacognitive and cognitive activities, facilitating or constraining individuals' development and display of independent SRL behaviours. 17 Findings from a recent study show that medical students' engagement in SRL changes over time and is influenced by their perceptions of the roles of, and interactions with, others in the workplace setting. 11 This illustrates that processes of CRL may influence the development of students' ability to self-regulate.
Examining how CRL, with social interactions as its main element, influences SRL development in workplace-based learning environments will contribute to a deeper understanding of how students actually develop SRL in clinical settings. This may help to create clinical learning environments that effectively foster SRL development, adequately equipping future physicians for lifelong learning. However, research focusing on the developmental aspect of SRL and how social interactions influence medical students' SRL development is still scarce. The present study aims to address this gap. Specifically, we aim to explore undergraduate medical students' perceptions of CRL in clinical clerkships and its perceived impact on the development of their SRL.

| Methodology
We used a qualitative approach to explore medical students' perceptions of relationships between CRL and SRL in clinical clerkships. Data collection and data analysis were conducted iteratively, informed by principles of constructivist grounded theory. 18,19 Using constructivist grounded theory, we acknowledged the research team's theoretical (DB, EWD, DMAS, MJBG) and practical (DB, EWD, DMAS, MJBG) backgrounds in SRL and CRL, and data collection and data analysis were informed by key concepts in SRL and CRL.

| Setting
We conducted this study in the undergraduate programme (Masters degree) in medicine at Maastricht University, the Netherlands. The Age, (year), mean (range) experiences, and any information students deem valuable for monitoring their personal and professional development. 21

| Participants and sampling
Overall, 79% of students in the Masters programme are female.
We purposively sampled medical students enrolled in the Masters programme to ensure variety in year of clinical training and gender (Table 1). Students were invited to participate via e-mail. Reminders were sent 2 weeks after the initial invitation. After 10 interviews, we considered that we had achieved data saturation and had gained an adequate understanding of the constructed themes and their relationships. Participants received a gift voucher for €10.

| Data collection
Between February and June 2018, the first author (DB) conducted semi-structured interviews that lasted between 50 and 75 minutes. During interviews, the interviewer prompted students to provide detailed examples and explanations of behaviours they described.
The interview guide was updated iteratively between interviews, informed by concurrent data analysis. All interviews were taperecorded and transcribed verbatim.

| Data analysis
Data collection and analysis followed an iterative process, allowing data collection and data analysis to inform each other. The first author (DB) and a research assistant (CN) independently read and coded transcripts 1-3 line by line to capture concepts related to CRL and students' development of SRL.
In the initial coding scheme, codes described various SRL and CRL behaviours such as the formulation of learning goals, asking for feedback and feedback reception, and details of the significant others with whom students engaged in CRL processes. Also in-

| Reflexivity
Within our methodological approach, meaning is constructed through interactions with research participants, the data generated

| RE SULTS
From the data, we were able to construct three major shifts in students' SRL, which were simultaneously influenced by and exerted influence on students' engagement in CRL as they progressed through clerkships. Firstly, students expressed a shift with regard to whom they involved in CRL, bestowing an important role on peers at the start of clerkships and increasing their preferences for clinician role models towards the end of the programme. Secondly, novice students' SRL was mainly externally driven by cues in the environment, whereas experienced students' SRL was increasingly driven by intrinsic cues and selfmotivation. Finally, regulatory behaviours shifted from a task-oriented approach towards a more comprehensive approach that focuses on professional competence and identity formation. We will detail these shifts and the influences of social interactions between students and relevant others that contribute to the shifts.

| From interacting with peers to interacting with clinician role models
Students described a change with regard to whom they engaged with in CRL over the course of clerkships. At the start of the programme,

| From external regulatory drivers to internal regulatory drivers
At the start of the programme, students' SRL seemed to be largely externally driven. Novice students described complying with portfolio instructions and relying on input from others when formulating learning goals prior to a clerkship. Quite often, however, they also found that these learning goals were not well aligned with learning affordances in a specific setting, which led to feelings of detachment from their learning plans. During the first clerkships of the programme, students then depended on others in the actual workplace setting to help them reactively adjust learning goals and strategies.

| From specific tasks to professional competence and professional development
Students indicated that the focus of the regulation of learning shifted over the course of clerkships. Novices described being task-oriented.
That is, goal setting was characterised by high levels of granularity; (PAL) has furthermore been shown to contribute to students' professional identity formation. 26,27 We therefore may need to pay more attention to stimulating PAL within CoPs, by, for example, forming learning and working partnerships between novice and more experienced students. As PAL entails bidirectional support, 28 stimulating PAL within CoPs will deliver the added benefit of developing learners' coaching and teaching skills. 29 Overall, our findings support the notion of SRL as being embedded in CRL, an idea gaining momentum. 30

| Limitations
This study was conducted with a limited sample of 11 participants.
Most (nine students) were enrolled in the second or final years of clinical training. This means that these students were able to describe their behaviour and thoughts at the start of clinical training only retrospectively, which may have made them susceptible to memory bias. Coregulated learning and SRL are abstract theoretical constructs, which students may have found difficult to talk about in practical terms. In constructing the interview guide, the research team attempted to formulate the questions as concretely as possible. It may be that CRL is more prevalent than students realise, and much of CRL may be hidden or implicit. For example, some students indicated after the interview had finished that they did not usually give very much thought to the concepts discussed in the interview. In a similar vein, participants were self-selected volunteers, which may imply that only students who actively thought about their learning agreed to participate.

| CON CLUS IONS
Our study emphasises the idea that SRL and SRL development are embedded in social interactions and CRL in clinical workplace settings. We can support medical undergraduates' development of SRL when CRL is structurally embedded in clerkships, acknowledging that the ability to self-regulate implies the ability to engage in meaningful interactions with others to support learning processes. In order to facilitate medical students' transformation into lifelong learners, we therefore need to support and organise learners' engagement in CRL.
For competency-based medical education, this means that we should reconsider the ways by which we include medical students in CoPs in order to encourage workplace partnerships between novice students and more experienced students and health care professionals.
By capitalising on the opportunities for CRL in clinical settings, we may be able to better equip students with the learning skills they need to develop into health care professionals who are able to optimise their learning and provide high-quality care in rapidly changing health care systems throughout their professional careers.

ACK N OWLED G EM ENTS
The authors would like to thank Celine Notermans, medical student at Maastricht University, the Netherlands, for her assistance in coding the data, and the reviewers and journal editor for their constructive feedback.

CO N FLI C T S O F I NTE R E S T
None.

AUTH O R CO NTR I B UTI O N S
DB is the principle author of the work. All authors contributed to the conception and design of the work. DB collected the data and drafted the initial manuscript. All authors were involved in data analysis and interpretation, and contributed to revisions of the paper. All authors approved the final manuscript for publication.