Alone and Together: Exploring the Relationship Between Individual and Social Metacognition in College Biology Students During Problem Solving

INTRODUCTION

Metacognition can enhance problem solving for life science students, which can help them succeed in challenging college courses. Use of metacognition is positively correlated with problem-solving ability and academic performance at the individual (Wang et al., 1990; Adey and Shayer, 1993; Ohtani and Hisasaka, 2018) and group level (Artz and Armour-Thomas, 1992; Goos et al., 2002). Students that use metacognitive skills can learn and problem solve more effectively by identifying gaps in their understanding and appraising their strategies for learning. Life science students need to be able to learn and problem solve effectively on their own and with others to be successful in their science courses and future scientific careers. However, the relationship between students’ approaches for individual and social metacognition has been undertheorized (Volet et al., 2009) and only a few studies have examined how these two types of metacognition interact (De Backer et al., 2020, 2021). Understanding how life science students use metacognition individually and socially can help inform interventions that support their success in college. We can gain insights about the potential relationship between individual and social metacognition by studying the same students’ metacognitive approaches while they solve problems on their own and in groups.

Metacognition

Metacognition can be defined as the awareness and regulation of our thinking for the purpose of learning (Cross and Paris, 1988). Schraw and Moshman's framework of metacognition conceptualizes metacognition in two major components: metacognitive knowledge and metacognitive regulation (1995). Metacognitive knowledge includes what we know about the learning task and ourselves as learners, and metacognitive regulation involves the steps we take to learn and how we assess and adapt our learning strategies as needed (Stanton et al., 2021). Our lab focuses on metacognitive regulation because these skills can result in changes in learning behavior. Metacognitive regulation includes three skills: planning, monitoring, and evaluating (Schraw and Moshman, 1995). These regulatory skills can be thought of temporally. Students use planning when they develop a plan for how they will learn (future learning). Students monitor when they check their understanding while learning (present learning). Students evaluate after completing a learning task by appraising the effectiveness of their plan (past learning).

Metacognitive regulation skills can inform one another. For example, a student preparing for an upcoming exam in their anatomy course may decide to use flashcards as a study strategy (planning) because this strategy helped them learn terms for a previous test (evaluating). While using flashcards, they may identify that they do not understand terminology related to the circulatory system (monitoring) and then review these terms more in-depth. This example also highlights how metacognition is context dependent (Kelemen et al., 2000). Study strategies that are effective in one learning setting (e.g., an anatomy course), may not work well in a different setting (e.g., a physics lab course). Utilizing metacognitive regulation skills while studying can help students identify what they do not know, and make changes to their study strategies (Stanton et al., 2021) so their learning is more effective and efficient.

Social Metacognition

While metacognition was initially conceptualized for the individual learner (Flavell, 1979), the theory of metacognition has been expanded to applications in social contexts such as group work (e.g., Jost et al., 1998; Volet et al., 2009; Kim et al., 2013). The metacognition that happens when students work in groups, or social metacognition, is defined as the awareness and regulation of other's thinking for the purpose of learning (Chiu and Kuo, 2009; Stanton et al., 2021). Social metacognition can also include the awareness of one's own thinking in the presence of a group (Halmo et al., 2022). Group work has been increasingly implemented in college science courses because collaborative learning can improve students’ conceptual understanding (Kruger, 1993; Van Boxtel et al., 2000) and scientific argumentation skills (e.g., Asterhan and Schwarz, 2009). Students also benefit from group work because group members can stimulate each other's metacognition (Volet et al., 2009; De Backer et al., 2020). Social metacognition can promote collaboration (Kim and Lim, 2018) and is related to students’ use of scientific reasoning (Lippmann Kung and Linder, 2007; Halmo et al., 2022) and improved problem solving (Artz and Armour-Thomas, 1992). During group work, students also use the metacognitive regulation skills of planning, monitoring, and evaluating (Goos et al., 2002; Halmo et al., 2022). For example, students may make plans to use a specific class resource or decide to skip a problem and come back to it (planning). While working with their peers, students can help check the understanding of their group members by offering corrections (monitoring). Additionally, students can appraise their group's solution and approach for problem solving (evaluating) and then make changes to their approaches as they continue to work together. In our past and present study, groups often use social metacognition in the form of monitoring and evaluating, but rarely use planning (Halmo et al., 2022) Understanding how life science students use metacognition on their own and in groups is reflective of how students are expected to learn in their college courses.

While many metacognition studies have focused on the individual learner (Stanton et al., 2015, 2019; Dye and Stanton, 2017; Halmo et al., 2024) and some research has explored metacognition during group work (Goos et al., 2002; Lippmann Kung and Linder, 2007; Halmo et al., 2022), little is known about how these two types of metacognition interact or inform one another. Chiu and Kuo (Chiu and Kuo, 2009) reviewed the theoretical bounds of individual and social metacognition and how these two forms of metacognition may interact. They suggested that social metacognition could facilitate individual metacognition based on the field's understanding of collaborative learning (Mugny and Doise, 1978; Brown and Palincsar, 1989; Goos et al., 2002; Chiu and Kuo, 2009). For example, elementary school students exposed to conflicting viewpoints from their peers had improved individual performance following group work (Mugny and Doise, 1978). Scholars suggest that exposure to conflicting viewpoints can influence students’ individual metacognition because when a student encounters ideas that are at odds with their own it can prompt the student to monitor or evaluate their own thinking (Goos et al., 2002; Chiu and Kuo, 2009).

Despite the potential for interaction, there is limited empirical evidence relating individual and social metacognition. From the few empirical studies exploring the relationship between these two forms of metacognition, previous work indicates that social metacognition can influence individual performance. For example, one study found that students who engaged in metacognition during group work performed better on future individual knowledge tests (De Backer et al., 2020), suggesting that social metacognition can impact individual outcomes. Yet this study only measured social metacognition, not individual metacognition.

To our knowledge, only one study has attempted to examine a potential relationship between individual and social metacognition by studying both forms of metacognition in the same students (De Backer et al., 2021). This study quantitatively assessed the individual and social metacognition that occurred in masters of education students in Belgium using think-aloud interviews and a recorded group work session. The individual and group recordings were broadly coded for the metacognitive regulation skills of planning, monitoring, and evaluating, and code frequency was used for multilevel modeling analyses. From these data De Backer et al. identified a significant correlation between students’ use of individual metacognition and social metacognition, albeit with a small effect size (2021).

While these results are promising, the learning context—a graduate education course within the European education system—is fundamentally different from an undergraduate biology course in the United States. Additionally, quantitative analysis may not fully speak to the complex relationship between individual and social metacognition, while qualitative work can illuminate potential nuances associated with these two forms of metacognition. To foster metacognition in college life science courses, we need a better understanding of how individual and social metacognition might be related in undergraduate science students in the United States.

Present Study

We hypothesize that there is a positive relationship between individual and social metacognition for college biology students. To explore the potential relationship between life science students’ individual and social metacognition, we asked:

We collected qualitative data from upper-level life science students to explore the metacognitive approaches they use to solve problems by themselves and with others and how their metacognitive approaches relate individually and socially. We conducted think-aloud interviews and audio recorded students during group work. These data collection methods capture in-the-moment metacognition, as it occurs during student learning. Use of in-the-moment methods is likely more accurate and robust than retrospective methods that ask students to remember and describe past learning, which can be subject to recall bias (Hofer and Sinatra, 2010; Schellings et al., 2013). Additionally, this study advances the field by collecting data for individual and social metacognition from the same participants so we can compare students’ approaches for metacognition across solitary and collaborative settings. By measuring life science students’ individual and social metacognition while they complete similar learning tasks, we can begin to understand how these two forms of metacognition compare during problem solving.

MATERIALS AND METHODS

Research Context and Data Collection

We collected data from a senior-level honors Cell Biology course, CBIO3400H, at a public southeastern R1 institution in Spring 2022. For this study, we recorded four groups of three students while they worked together and interviewed the same 12 upper-level students individually in order to characterize their use of social and individual metacognition. We decided to collect data from an upper-level course because students in this sample are likely to show evidence of metacognition (Stanton et al., 2019), thus making a comparison between individual and social metacognition possible.

CBIO3400H is a capstone course that integrates knowledge from biochemistry, genetics, and biology, and is representative of an upper-level biology course. The course consisted of two weekly, interactive lectures and a breakout session that occurred once a week. The enrollment size for this course was 40 students. All classes took place in a SCALE-UP classroom where students sat at circular desks in groups of three with dry erase boards positioned around the room (Beichner et al., 2007). This classroom set-up encouraged students to work with their classmates during the lecture and breakout sessions. Data collection occurred in Spring 2022 when classes were held in person and students and the instructor wore masks.

For breakout sessions, students worked in groups of three to complete guided inquiry problem sets (Moog and Spencer, 2008). Problem sets were in a pen-and-paper format, and students interpreted data from published cell biology papers that were related to the content being covered in lecture (e.g., Stanton and Dye, 2017; Pfeifer and Stanton, 2021). Breakout sessions lasted 75 min. For the first 50–60 min of the session, students worked in groups of three to discuss and complete the problem set. During this time, the course instructor and a graduate teaching assistant walked around the room to help answer questions and facilitate discussion. The remainder of the breakout session was dedicated to a whole-class discussion of the problem set. Students were assigned one of three roles when working in a group: manager, recorder, and presenter. When assigned the role of manager, students kept their group on task during the breakout session. The recorder wrote down the group's responses, which were turned in and graded. Last, the presenter discussed the group's responses during the whole class discussion. Students rotated through these roles as a group and formed new groups after completing a full rotation. Groups received written feedback on their shared problem set from the graduate teaching assistant. While problem sets were graded based on good faith effort rather than correctness, material from the breakout sessions made up 30–40% of the exam content.

We qualitatively characterized individual and social metacognition for 12 participants, which is a sample size that is aligned with or larger than past qualitative work on social metacognition (Goos et al., 2002; Lippmann Kung and Linder, 2007; Halmo et al., 2022). All participants provided written consent and were compensated $40 for their participation in this study. The classroom was predominantly women (62% women, 38% men), which is reflected in this study's sample; 75% of consenting participants were women (Supplemental Materials, Supplemental Tables 1 and 2). The number of women that consented to participate could represent a potential volunteer bias in this study (Rosenthal, 1965; Brownell et al., 2013). All names used are pseudonyms. This study was declared exempt by Institutional Review Board at the University of Georgia (STUDY00006457).

Data Collection for Social Metacognition.

Students working in groups of three were audio recorded in order to capture their use of social metacognition. We also collected each group's completed problem set for future analysis. We collected data at two different timepoints in the semester. Early in the semester we recorded two breakout sessions (week 2 and 3), and later in the semester we recorded an additional breakout session (week 10). The first recorded breakout session covered protein visualization techniques. We recorded this breakout session so participants could get comfortable being recorded during group work and did not analyze data from this recording. The second recorded breakout session covered ER import (week 3), and the last recorded breakout session covered nuclear import (week 10). For recording, students were individually microphoned using the 8W-1KU UHF Octo Receiver System equipment. Individual audio files were merged using Steinberg Cubase software and transcribed using Rev.com. All transcripts were checked for accuracy prior to data analysis.

Data Collection for Individual Metacognition.

We developed two isomorphic questions for the individual interviews that were aligned with material from the recorded breakout sessions (Kotovsky et al., 1985). Isomorphic questions are questions that maintain the same question format and assess similar concepts but in a novel way. For example, students were asked to interpret data from an ER import assay during their breakout session and provide a scientific explanation of protein translocation to the ER. For the isomorphic question used in the individual interviews, students interpreted novel data from a different ER import assay and then had to provide a scientific explanation. Use of isomorphic questions allows us to compare students’ use of individual and social metacognition for a similar problem-solving task on the same cell biology topic. We interviewed the same 12 participants using these isomorphic questions in a think-aloud protocol for capturing in-the-moment metacognition (Supplemental Materials). Think-aloud protocols have participants verbally describe their thought process while answering questions (Afflerbach, 2001; Schellings et al., 2013). This method allows for students to describe the thoughts that enter their working memory during their problem solving process (Ericsson, 1988; Schellings et al., 2013). Students were given a practice biology problem to get acquainted with the think-aloud process before solving the isomorphic questions. During the think-aloud, when students were silent for more than 10 s the interviewer used the prompt “what are you thinking now?” to encourage participants to share their thought process, although we rarely had to use this prompt. In addition to use of the think-aloud protocol, we also asked students questions from a revised version of our published semi-structured interview protocol (Dye and Stanton, 2017) to gain additional insights into their metacognition, self-efficacy, agency, science identity, and perceptions of group work. Individual interviews were audio and video recorded on Zoom. Interview transcripts were transcribed verbatim using Temi.com and checked for accuracy.

Data Analysis

We analyzed transcripts using MaxQDA 2020 qualitative software. Our unit of analysis was an utterance, which are collections of words, phrases, or statements. For example, “Okay. Why would you say that was used as a control? Because it says it's used as a control” can be broken into multiple utterances. In this example, “Why would you say that was used as a control?” would be a monitoring utterance, while the statement “Because it says it's used as a control” is an evaluating utterance related to the relevance of their reasoning. Alternatively, utterances can be multiple lines of speech or only a few words. We focused our characterization of social metacognitive utterances on the two recorded breakout sessions that have corresponding isomorphic questions in the individual interviews. For analysis of individual metacognition, we identified metacognitive utterances in the think-aloud portion of the interview. By focusing our analysis of social and individual metacognition on in-the-moment methods for similar problems, we are able to make a more direct comparison between these two forms of metacognition.

First-cycle Coding of Metacognition.

We used deductive and inductive coding to characterize social metacognition. We began with open coding the groups’ transcript data to identify any new ideas in the data and capture our initial impressions of the groups’ metacognition through analytic memos. Following open coding, we applied a previously developed codebook of social metacognition to characterize students’ use of metacognition during group work (Halmo et al., 2022), and made changes to these pre-existing codes based on our open coding of the data (Supplemental Table 3). One code we used, “corrections of another student,” captures when a student directly corrects a group member. Corrections show verbal evidence of a monitoring event because a student needs to identify and then compare their own understanding with their group member's understanding prior to offering a verbal correction of their peer's thinking. For example, “corrections of another student” was applied in the data when a student said, “The C-terminal domain tail is on the ribosome” and their groupmate said, “No, it's on the RNA polymerase.” We also coded for “evaluations of others” when students assessed the thinking, approach, or solution shared by a group member and whether or not it was effective or relevant for problem solving. For example, the “evaluations of others” code was used when a student asked “Yeah, but does that explain why mitochondria and chloroplasts remain intact?” after their peer shared a potential solution that did not fully answer the question being asked. Two researchers (E.K.B. and O.K.M.) then individually coded the data using this codebook, met to discuss the applied codes, made changes to the codebook based on these conversations, and iteratively recoded the transcripts until the data were coded to complete consensus. We chose to code this data to consensus because it allowed us to consider multiple, diverse viewpoints and to refine our codebook definitions throughout the entirety of the data analysis process (Tracy, 2010). Another researcher (J.D.S.) provided insight on metacognition and the course content to aid in this iterative coding process.

Our coding approach was similar for individual metacognition. We also started with open coding the think-aloud data to document our initial impressions. We then used a previously developed codebook for individual metacognition during think-aloud problem solving of biology questions (Halmo et al., 2024). The individual metacognition codebook was similar to the codebook used for capturing social metacognition with some modifications. Codes that captured when students were monitoring or evaluating other students were not included in the individual metacognition codebook. For instance, the code “corrections of another student” was not applicable to the individual think-aloud data because they were solving problems on their own. However, the code “self-corrections” was used in both codebooks, which captures when a student corrects themselves without prompting while talking out loud (e.g., “…that causes more YFP to be produced. No, no, no, more CFP to be expressed”). Two researchers (E.K.B. and O.K.M.) iteratively coded and discussed the coding of the data until complete consensus was reached for all individual transcripts.

Second-cycle Coding of Metacognition.

To synthesize the coded data from the group transcripts and think-aloud interviews, we developed profiles for each participant. Creating a profile allowed us to focus on specific metacognitive skills (i.e., monitoring and evaluating), so we could then systematically compare a student's individual and social metacognition. For profile creation, we developed a profile template that aligned with our research question, which focuses on understanding how a student's use of metacognitive regulation was similar and different during individual and group problem solving. We used the profile template to capture when students individually and socially used the metacognitive regulation skill of planning, monitoring, and evaluating. More specifically, the profiles identified when participants 1) verbalized plans for problem solving, 2) used statements and questions to monitor their understanding, 3) acknowledged their confusion or verbalized their understanding, 4) offered corrections, and 5) used evaluations of their own and others’ thinking during problem solving (Supplemental Table 4). We considered the frequency of codes from MaxQDA and the most salient themes when recording the key forms of individual and social metacognition for each participant in the profile template.

To help identify salient themes regarding use of metacognition, we used analytic memos generated throughout data collection and data analysis (Saldaña, 2021). For instance, in an analytic memo following Hazel's interview, interviewer E.K.B. wrote: She described herself as the quiet person during group work, and that she prefers to think about the problem by herself first. She likes group work and thinks it affects her learning because she is able to “get different perspectives” on how to think about a problem. This analytic memo and others we wrote led us to attend to the reasons why Hazel might be quiet during group work, even though she likes group work. By following this idea, we uncovered possible themes regarding Hazel's self-efficacy and group dynamics in the data. This and other possible themes identified from analytic memos were recorded in all the profiles along with frequently used codes to help us determine the most salient themes. After creating profiles for each participant, we compared the data across all 12 profiles, which allowed us to cluster participants based on their use of individual and social metacognition.

Frequency Analysis of Metacognitive Approaches.

In addition to the profiles, we wanted to further analyze the relationship between students’ individual and social metacognitive approaches by considering metacognitive utterance frequency in both problem-solving contexts. We compared the frequency of metacognitive utterances during problem solving for both individual and social metacognition and identified the frequency of overlap between their individual and social metacognitive approaches. For quantifying social metacognition frequencies, we restricted our analysis to areas of the group transcripts that corresponded with the isomorphic questions used in the individual interviews. The size and overlap of the circles in Figure 1 represent the number of coded metacognitive utterances from MaxQDA that participants used during the individual isomorphic questions and their corresponding breakout session problems. Circle size and overlap represent frequency of coded metacognitive utterances (Figure 1; Supplemental Figure 1). While this analysis is able to offer quantitative snapshots of each participant, we caution the reader from making broad generalizations as this analysis represents the quantification of a complex, contextual construct. Instead, as researchers we used the quantification of metacognitive utterances as a tool to aid in the interpretation of the relationship between individual and social metacognition while considering our participants’ rich qualitative data.

Multiple Case Selection.

To answer our research question, we selected multiple cases to help explain the complex relationship we identified between individual and social metacognition. We decided to present our data in a series of cases that allow us to show contextualized quotes from students during their problem-solving process because previous work has shown that learning context affects students’ use of metacognition (Kelemen et al., 2000). To select cases, we compared profiles and then made final selection decisions based on data from the semi-structured interviews, the coded data, and our analytic memos (Supplemental Table 5). We chose individual cases that 1) illustrate the maximum variation in metacognitive approaches used during problem solving (Flyvbjerg, 2006) and 2) represent each recorded group in this study. Our cases meet the criteria outlined by Yin (2009) and Stake (2006), where cases have defined units of analysis, study a phenomenon in a bound context, and utilize multiple data sources. Our cases’ units of analyses are the individual students, and we are studying their individual and social metacognition in a bound context (i.e., cell biology problem solving). These cases are also informed by multiple sources of data such as recordings and data analysis from individual and group data sources (Supplemental Table 5). Throughout this case selection process, a researcher (E.K.B.) would identify potential themes by comparing within and then across profiles and would receive comments and feedback on these ideas (J.D.S.) until cases were agreed upon.

By creating and then comparing profiles, we found that seven of the 12 students in our sample used similar metacognitive approaches for individual and group problem solving (Supplemental Figure 1). Of these students, we identified that a few students were often able to accurately pinpoint when they were incorrect during individual problem solving and regulate their thinking to improve their problem solving (Isabel and Sophia). We ultimately selected Sophia because she displayed instances of high learning self-efficacy during individual problem solving that helped her proceed in the problem-solving process (Table 1). Sophia's case reflects the high end of what we observed for over half of the students in our sample who used similar metacognitive approaches individually and socially.

While most students in this upper-level sample displayed evidence of a reciprocal relationship between individual and social metacognition, we identified a few students who did not display a reciprocal relationship between these two forms of metacognition. For example, we observed that two students in our sample showed limited evidence of metacognition during individual and group problem solving (Yasmin and Davina). We then used data from the semi-structured interviews and the analytic memos we created throughout data collection and analysis to make a selection between Davina and Yasmin. We selected Yasmin as a case because she displayed less metacognition than Davina. Additionally, her approaches for metacognition often did not lead to improved problem solving due to her monitoring inaccuracy, meaning that she often did not accurately identify what she did and did not know (Tobias and Everson, 2009).

We also found that some students in our sample showed strong evidence of metacognition in one problem-solving setting but displayed limited evidence of approaches for metacognition in the other problem-solving setting. We identified that one individual in our sample showed strong evidence of metacognition during the individual think-aloud, but she did not always use these same metacognitive approaches during group work (Hazel). We also identified two individuals in our sample that displayed stronger evidence of metacognitive approaches during group work, but they did not employ these strategies during the think-aloud interview (Dana and Cal). We ultimately selected Dana because she represented a group that had not yet been represented through this case selection process (Table 1).

RESULTS AND DISCUSSION

Previous research has mainly focused on individual metacognition (e.g., Stanton et al., 2015, 2019; Dye and Stanton, 2017; Heidbrink and Weinrich, 2021) and a few studies have characterized metacognition socially (Goos et al., 2002; Lippmann Kung and Linder, 2007; Kim et al., 2013; Kim and Lim, 2018). However, the relationship between a student's use of individual and social metacognition has been understudied (Volet et al., 2009). We aimed to qualitatively explore a potential relationship between students’ approaches for metacognition when problem solving on their own and in groups. We asked, how does a student's approach for monitoring and evaluating while problem solving individually relate to their metacognitive approaches when working in groups? To answer our research question, we collected in-the-moment metacognition data while life science students were solving problems on their own and in groups.

We found that seven of the 12 students in our sample displayed a reciprocal relationship between their metacognitive approaches individually and socially (Supplemental Figure 1). This finding aligns with a previous quantitative study of Belgian masters of education students that identified use of individual metacognition was significantly related to use of metacognition during group work (De Backer et al., 2021). However, we also found that some students primarily showed evidence of metacognition either individually or socially, highlighting that the relationship between these two forms of metacognition is more complex. To help illustrate the variation in metacognition we observed, we present our results in a series of individual cases. This approach allows us to represent complex data that contextualizes students’ approaches for monitoring and evaluating during individual and group problem solving. We purposefully chose four cases from our sample that capture the greatest variation in students’ use for monitoring and evaluating during individual and group problem solving (Flyvbjerg, 2006), and we explore the extent to which their use of metacognition is similar in these settings. We are presenting in-depth cases from four students: Sophia, Yasmin, Hazel, and Dana (Figure 1). Following the cases, we share hypotheses that emerged from our data that may help explain the variation we observed for use of metacognition during individual and group problem solving in these cases. These hypotheses can be tested in future studies with additional populations and samples.

Sophia: Strong Evidence of Metacognitive Skills during Individual and Group Problem solving

Sophia showed evidence of multiple approaches for monitoring and used similar monitoring skills individually and during group work. To monitor her understanding, Sophia used corrections, questions, and acknowledgments of confusion during individual and group problem solving. Importantly, once Sophia monitored on her own, she used strategies that helped her persist in the problem-solving process. She also used evaluations, or appraisals, of her thinking and reasoning when problem solving individually and socially. Sophia's evaluations socially allowed her to rethink her solutions and focus her group members’ reasoning.

Monitoring Understanding.

Students monitor their understanding when they assess their knowledge while completing a learning task. During individual and group problem solving, we found that Sophia monitored her understanding by correcting knowledge, asking questions, and acknowledging confusion. For example, while writing down her solution during the individual interview, Sophia realized she made an error and verbally offered a correction of her idea: “I'm just now realizing that I explained it wrong earlier because if those are mitotic spindles then that should span the entirety of the cell and not just the nucleus.” While crafting her response, Sophia checked her understanding of the data and identified that her initial interpretation was incorrect, which is in contrast to other students in this study. Nearly half of the students interviewed solved the problem as if it was the same breakout session question they recently solved in groups. Sophia displayed that her monitoring is well calibrated because she was able to accurately identify her initial interpretation of the data was incorrect.

We found that Sophia also corrected herself during group work. When working through the nuclear import breakout session, Sophia corrected her own idea, and then asked her group members for confirmation on the idea she corrected: “Ooh. Another similarity is that both of them aren't cleaved or actually no. The mitochondrial one is cleaved, but not the internal ones. Right?” In addition to correcting herself, Sophia also corrected her group members’ knowledge in order to help monitor the group's understanding:

Raj:	And then in the nucleus, the GAP would make it from RanGDP into RanGTP.
Sophia:	You mean the GEF?

In general, use of corrections during group work can allow students to reach a shared understanding as a group and help their group members monitor their understanding (Halmo et al., 2022).

Sophia also checked her understanding during problem solving by asking questions. When asked to describe the steps of her thinking process after problem solving individually, Sophia mentioned that she used questions to help scaffold her problem solving. She described that she used questions to breakdown the data, saying, “I read the question and then I looked at the data sequentially, like ‘what is each one showing?’” Sophia also used questions during individual problem solving to identify and work through areas of confusion. For example, Sophia acknowledges “I'm a little bit confused here because I'm trying to rationalize it. If [RanGTP] is found there, what function would it have? [Rereading question] Explain where RanGTP is found during mitosis.” In this example, Sophia acknowledged her own confusion and then asked herself a question. Although she did not answer the question she posed to herself, she used a strategy to help her continue in the problem solving process (i.e., rereading the question).

Sophia also used questions to help monitor her understanding when problem solving in groups. In general, we found that students rarely asked themselves questions during group work. Instead, students often asked their group members questions to help clarify their understanding. We found that Sophia frequently used questions to monitor her own and her group members’ understanding, asking “So what is endo H? Why did they even put it there?” or “Wait, why would you see cyan?” These questions either helped clarify Sophia's own understanding or encouraged her group members to expand upon their reasoning.

Last, we found that Sophia monitored her knowledge by verbalizing areas of confusion individually and during group work. For instance, Sophia acknowledged her confusion when she said, “So I'm confused why RanGTP, which is involved in nuclear import and export, would be kind of like everywhere in the cell” individually and “I don't know PK and TX” during group work. We observed that when students like Sophia acknowledged their lack of understanding in front of their peers, they often received clarification from their group members or used class resources such as the instructor to help them move forward in problem solving.

Evaluating Thinking.

Students evaluate when they appraise how future implementation of their strategies could be improved (Stanton et al., 2021). During problem solving, evaluations could look like appraisals of one's reasoning or knowledge, or their experience when problem solving (Halmo et al., 2024). We found that Sophia evaluated her ideas by rethinking her answers and determining how relevant they were for problem solving when working on her own and in groups. For example, in her think-aloud, Sophia evaluated how accurate her response was saying, “I don't think that's the right answer because I'm kind of confused as to why there would be RanGTP on like the mitotic spindle and around it as well.” During individual problem solving, we also identified multiple instances where Sophia rethought her solution after problem solving. In one instance, Sophia evaluated her solution and then wrote a different response to describe the data more accurately. During the reflection portion of her interview, Sophia also realized, “Oh and now I'm just seeing this, but it would've been helpful to read microsomes 'cause then I could actually say that it's with the ER and not just the cell in general. So that would've helped.” In this instance, Sophia evaluated how her solution could be more accurate and specific for answering the question.

During the individual interviews, we observed that following use of metacognition, some students struggled to amend or change their answer when they appraised that their solution was inaccurate or their ideas were incorrect. However, we found that Sophia used strategies such as rereading the prompt or self-coaching that helped her persist during individual problem solving after she monitored or evaluated her thinking. Self-coaching, or words of encouragement, are statements that act as reassurance when students are faced with uncomfortable feelings after monitoring incorrect ideas or after evaluating that their solution may contain errors (Halmo et al., 2024). Sophia used self-coaching when individually problem solving, saying “And now I'm realizing I wrote plasma membrane instead of ER membrane while explaining this, but that's okay.” Here, Sophia rethought her solution and acknowledged that it was okay that her initial idea was inaccurate. In contrast, a few other students, such as Hazel, would give up or stop problem solving when they detected an error in their thinking or solution.

Sophia, who rethought her solutions individually, also used this form of evaluation during group work. After suggesting an idea to her group, Sophia reconsidered her own idea and evaluated, “But then again, that doesn't make sense either. Because it would have to denature the primary sequence too for it to not show up at all.” Sophia's evaluation of her own thinking allowed the group to rethink their interpretation of the ER import assay data. Sophia also evaluated her group members’ ideas during breakout sessions. For example, Sophia told her group members, “I thought this question was only referring to import,” which helped focus her group's reasoning on what was relevant for answering the question. Sophia also evaluated whether her group members’ responses were sufficient to answer the question, asking “Why would you say that was used as a control? Because it says it's used as a control.” In this evaluation, Sophia is appraising how the group's solution addresses the question being asked.

Summary of Sophia's Metacognition.

Sophia used multiple similar approaches for monitoring and evaluating when problem solving individually and in groups. For example, Sophia monitored understanding by offering corrections for her and her group members’ ideas. Previously, we have found that requesting and receiving corrections are related to students’ use of higher quality reasoning during group work (Halmo et al., 2022). We also found that Sophia used similar monitoring approaches during individual and group problem solving by asking questions and acknowledging her confusion. While monitoring a lack of understanding can be uncomfortable for students (e.g., Dye and Stanton, 2017), we found that after Sophia monitored her understanding she was able to continue in the problem solving process by using problem-solving strategies or self-coaching. Self-coaching can help students persist in problem solving individually and in groups by allowing them to move past the discomfort of being incorrect or monitoring areas of confusion so they can continue problem solving (Halmo et al., 2024). We also found that Sophia could reach a more accurate answer by revising her content knowledge, rethinking her solutions, or appraising what is relevant for solving the problem. We hypothesize that when students have strong metacognitive skills, they will show similar evidence for individual and social metacognition and use multiple approaches to monitor and evaluate when problem solving on their own and in groups.

Hypothesis 1: When Students have Strong Metacognitive Skills, They can Transfer their Approaches to Monitor and Evaluate between Similar Individual and Group Problem-solving Tasks

By holistically analyzing students’ approaches for individual and social metacognition, we identified that seven of the 12 students in our sample displayed a reciprocal relationship between their use of metacognition individually and socially. These students used similar metacognitive approaches during individual and group problem solving. Of the students that used similar metacognitive approaches during problem solving, we present results from Sophia. In both settings we observed that Sophia 1) accurately monitored her understanding through the use of questions and corrections and 2) evaluated her own ideas and her group member's ideas during individual and group problem solving. For example, Sophia corrected herself when problem solving individually (“I'm just now realizing that I explained it wrong earlier…”) and socially (“Another similarity is that both of them aren't cleaved or actually no”), and she also corrected her group members. Successful transfer of metacognitive skills tends to be domain-specific (Scott and Berman, 2013), and occurs more readily for similar tasks (Gentner et al., 1993). We predict that students with developed metacognitive skills are able to transfer their approaches for monitoring and evaluating when completing similar problem solving tasks on their own and in groups. Assessing the transfer of metacognitive skills for students at different points in their metacognitive development could help test this hypothesis.

Yasmin: Limited Evidence of Effective Metacognition during Individual and Group Problem solving

We found that Yasmin showed fewer examples of monitoring and evaluating individually and socially (Figure 1). To monitor her understanding, we observed that Yasmin would often ask questions to her group members, and individually would acknowledge what is familiar to her. During the think-aloud interview, Yasmin also showed some evidence of correcting her understanding and asking herself questions; however, these metacognitive approaches often were ineffective for her problem solving because she would then focus on information that was irrelevant for problem solving or these approaches illustrated that her monitoring skills might still be developing. For example, Yasmin sometimes monitored by reconsidering a statement she made and determining it was accurate, when her statement was actually inaccurate. Yasmin only evaluated her thinking when prompted during the individual think-aloud interview. During group work, Yasmin did not evaluate herself nor her peers. In our data, we found that Yasmin's use of monitoring and evaluating did not always move her problem solving forward.

Monitoring Understanding.

Yasmin predominantly acknowledged her familiarity with the problems to check her understanding during individual problem solving, and occasionally would correct or use questions as approaches for monitoring. We observed that when Yasmin would show evidence of metacognitive approaches, they could be inaccurate or ineffective for problem solving. When describing her process for solving the problem during her individual interview, Yasmin mentioned what she remembered while solving: “For B, I kind of looked at the image and then I was remembering that when you're looking at the like coloring of the panels, if it's towards blue, you know that it's higher in RanGTP and like remembering that RanGTP is also high in concentration in the nucleus and RanGDP is higher in concentration in the cytoplasm.” In this example, Yasmin describes that she is solving the problem as if it is the previous question she solved with her group. Yasmin describes that she relied on her memory when problem solving, and these feelings of familiarity may have contributed to her inaccurate monitoring during her think-aloud interview (Thiede et al., 2003). While Yasmin was activating prior knowledge, using her previous group work experience of interpreting YRC probe data misled her when solving a different yet related isomorphic question individually.

In addition to acknowledging what is familiar to her, Yasmin occasionally used corrections and questions to monitor her understanding. For example, during the interview Yasmin corrected her understanding of the reagent protease K: “Well in [lane] two you're seeing like the presence of protease K which eats proteins essentially. Yeah, yeah it eats proteins, a little bit. It degrades. No, it degrades membrane. No, it degrades proteins. Degrades proteins.” Yasmin initially described that protease K “eats” proteins, which is somewhat correct, and then incorrectly stated that this reagent might actually degrade membranes. Following this, Yasmin correctly described that protease K degrades proteins. While Yasmin was able to correctly monitor her understanding, this example highlights that Yasmin's monitoring was not always accurate. Monitoring accuracy refers to a learner's ability to accurately identify when they do or do not know something (Tobias and Everson, 2009). Metacognitive regulation that results in effective or optimal changes to learning strategies requires the ability to accurately monitor understanding. We found that Yasmin showed some evidence of corrections during individual problem solving, although she did not correct herself or her group members during the recorded breakout sessions. Yasmin's lack of corrections during group work was atypical because we observed that nearly every other student in our sample either corrected themselves or a group member during the recorded breakout sessions.

Last, Yasmin described that she also used questions to check her understanding while problem solving individually. We observed a few examples where Yasmin used questions to monitor her understanding. The questions Yasmin used individually were similar to Sophia, in that they helped scaffold her problem solving. Yasmin described that she asked herself questions to help her interpret the data for the isomorphic ER import assay data: “…going through this question, I would look at the data then I'd think about like, ‘okay, why would just detergent be able to do this?’ And I'd be like, ‘so how did we get the results from three, we know we can't have any protein being expressed?’ All right, ‘well how does protein not get expressed?’” Yasmin mentioned that she used questions to help structure her problem solving; however, the last few lines of questioning (i.e., questions about proteins being expressed) might not fully help scaffold the problem because it does not accurately address the data that she is interpreting. In the case of ER import assays, the data do not represent whether a protein is being expressed (like a Western blot would), but instead if a protein is being degraded by protease K. In this example, we see that Yasmin uses monitoring skills, but they may not be fully refined yet or they may potentially reflect a lack of content knowledge. During group work, Yasmin predominantly used questions to check her understanding. For example, she asked, “…so when the ribosome reaches a stop codon, translation essentially stops?” Yasmin also used questions as a way to receive feedback on her ideas: “So, this proposed signal sequence encodes for a protein, right?” We observed that Yasmin mainly used close-ended yes or no questions when checking her understanding during group work. In comparison to other students, we found that Yasmin did not frequently use a variety of approaches to monitor understanding during group work.

Evaluating Thinking.

We found that Yasmin only evaluated her solution when asked to rate her confidence in her solutions to the problems after her think-aloud whereas half of the students in our sample, including Sophia, evaluated their thinking without prompting. Yasmin evaluated based on perceived accuracy of her statements and her feelings of confidence. For example, when prompted to rate her confidence on her response to a question, Yasmin describes: “Well, I am pretty confident in my answer. I feel like my answer for A is supported by the image on the right and then on top of that, I think that the information I presented in A is very correct. So, I would give that like maybe a 95. I feel like I'd get a good grade on a question like that.” Here, Yasmin evaluates based on her perceived accuracy of her response and also predicts that she would receive “a good grade” on her solution. Yasmin also evaluated a solution by describing her feeling of a lack of confidence when asked to rate her response: “For B because I can't for the life of me distinguish between metaphase and prophase. I just feel like I'm not very confident in that answer. Like on a scale of 1 to 10, I'm feeling like six [out of] 10 on confidence on that one.” Yasmin describes that she is not very confident in her solution because she cannot distinguish between the phases of mitosis. This appraisal of her solution is related to what is relevant for answering the question, which asked her to describe the data that is occurring during metaphase of mitosis. While Yasmin evaluated herself individually when prompted to reflect on her problem solving, there is no evidence of her evaluating herself or her peers during group work. Yasmin was one of the few students in our sample that did not offer evaluations of themselves or others during group work.

Summary of Yasmin's Metacognition.

The data suggest that Yasmin is still developing her metacognitive skills. We found that Yasmin may use corrections, questions, and acknowledgments of familiarity as approaches for monitoring her understanding when solving problems on her own. While we observed that Yasmin showed evidence of monitoring, her use of monitoring did not always accurately characterize her own understanding (or her lack of understanding). We observed that Yasmin's individual monitoring approaches were 1) often ineffective either because they could be inaccurate due to her reliance on the problem's familiarity or 2) irrelevant for problem solving. Making changes to improve problem solving relies on a student's ability to accurately monitor the bounds of their knowledge or when they are incorrect. However, being able to accurately monitor one's understanding is challenging (Pressley et al., 1990) and even prior knowledge of the task and receiving performance feedback may not improve students’ monitoring accuracy over time (Foster et al., 2017; Morphew, 2021). During group problem solving, we observed that Yasmin predominantly asked her peers close-ended questions (e.g., yes or no) to monitor her understanding. Previously, we found that when students asked open-ended questions it could prompt their group members to provide explanations, which has been associated with higher-quality scientific reasoning (Halmo et al., 2022). Yasmin's use of close-ended questions may affect the types of social metacognition that are elicited. Asking yes or no questions may also prevent her from receiving explanations from her peers that could help her better monitor her understanding during group work. For the metacognitive skill of evaluating, Yasmin used few evaluations when working on her own and none during group work. During individual problem solving, she appraised her solutions based on their perceived accuracy, her feelings, and what grades she would receive when asked to rate her confidence in her responses—and she only evaluated when prompted by the interviewer. During group work, Yasmin did not use any approaches to evaluate her group's solution. We observed that Yasmin's use of metacognitive approaches often did not lead to more successful problem solving due to a lack of content knowledge and monitoring accuracy. Yasmin's case suggests that metacognition alone is not sufficient to improve problem solving.

Hazel: Primarily Shows Evidence of Metacognitive Skills When Solving Problems Individually

Hazel primarily showed evidence of monitoring and evaluating individually. We found that Hazel used multiple strategies to monitor her understanding and evaluate her ideas during the individual think-aloud interview. These monitoring and evaluating approaches are similar in type to Sophia. However, we found that Hazel did not use these monitoring and evaluating strategies as often during group work (Figure 1).

Monitoring Understanding.

Hazel checked her understanding during her think-aloud by correcting her wording, asking herself questions, and describing what she knows. While problem solving individually, Hazel superficially corrected herself by amending her wording. For example, Hazel corrected her wording: “which helps the protein dissociate from GDP to bind to GD, GTP.” However, during group work, Hazel did not correct herself or her group members.

We found that Hazel also used questions to check her understanding when problem solving on her own and in groups. Hazel asked herself questions to aid in her monitoring, and these questions highlighted areas of confusion or uncertainty. In one instance, Hazel acknowledged what she thought she knew and then asked herself a question during her think-aloud interview. Hazel said, “The higher ratio, the more YFP you have. Does that mean you have more GTP?” Here, the question is identifying an area of uncertainty. Ultimately, Hazel did not answer her own question and instead said, “I think that's all I have” and then proceeded to type her final answer to the problem. While Hazel used questions as an approach for monitoring her knowledge, in this example she is unable to clarify her confusion and so she ends her problem-solving process. This contrasts with Sophia, who used a strategy to help her continue problem solving after asking herself a question.

During group work, Hazel asked questions to her group members to check her understanding. For both analyzed breakout sessions, Hazel asked questions to monitor understanding, such as, “What do the next lines represent? If there's an absence of the rough ER?” While Hazel's questions were always acknowledged by her group members, sometimes they were not answered because her group members admitted they had similar areas of confusion. We also observed that Hazel's questions would sometimes be met with frustration. For example, one of Hazel's group members answered another question, but in a terse manner:

Hazel:	The NLS is not cleaved?
Samira:	Mm-hmm (affirmative). Because it's not a specific sequence, it's literally in the next question. It's fine. I think it's not cleaved because this would alter the coding sequence of the protein itself, possibly. So that's the advantage of having one that's not cleaved.

While Hazel's question was answered, it was met with frustration based on Samira's tone from the audio recording. Dealing with frustrated or negative group members could potentially discourage monitoring when working in groups. This group member's response may help explain why Hazel did not use the same approaches for monitoring and evaluating during breakout sessions.

Last, Hazel monitored her knowledge during problem solving by verbalizing her understanding and confusion. Individually, Hazel frequently acknowledged what she knows during the problem solving process. For example, Hazel stated, “I know that the detergent would destroy the membrane” and “I know with nuclear transport… I know that there's a specific sequence.” This strategy contrasts with Hazel's monitoring during group work. During breakout sessions, Hazel acknowledged her confusion. For example, Hazel acknowledged, “What am I doing wrong? It's not clicking.” Hazel's monitoring differed between these two problem-solving settings; she was more likely to acknowledge confusion in front of her group members instead of verbalizing what she knows, which is an approach she frequently used when problem solving individually.

Evaluating Thinking.

We found that Hazel used similar approaches to evaluate as Sophia when solving problems individually, albeit less frequently (Figure 1). Hazel evaluated her reasoning by considering her response's accuracy and rethinking her solution. For instance, when evaluating her response's accuracy, Hazel said “I'm not sure if I'm remembering everything correctly,” when problem solving individually. Additionally, in one instance, Hazel rethought her solution after evaluating her response during individual problem solving:

Here, Hazel rethought whether she initially described the ratios of the false color accurately and provided additional background that could alter her solution. We found that Hazel also evaluated her response based on her feelings of confidence and the grade she would receive during the reflection portion of the think-aloud: “I guess if I got the chance to refresh my memory on ER import, then I would feel more confident, but right now. I mean, I won't say I'm not confident at all…I don't think I would get every point, but I feel like I would get some credit for this problem.” After the think-aloud, Hazel evaluated that she would receive partial credit for her response to one of the isomorphic questions. While Hazel used multiple approaches to appraise her solutions during individual problem solving, she did not verbally evaluate her own or her group members’ ideas during breakout sessions.

Summary of Hazel's Metacognition.

Hazel used similar approaches for monitoring and evaluating as Sophia when she worked through cellular biology problems individually. Both used corrections and questions to monitor their understanding individually. Additionally, Hazel frequently acknowledged what she knows while problem solving on her own, which could help her continue in problem solving because she is making her prior knowledge explicit. Sophia and Hazel also used similar approaches for evaluating during individual problem solving. However, during group work, Hazel did not show much evidence of the metacognitive skills she used individually. We found that Hazel did not use corrections or evaluations during group work, despite being one of the students that evaluated her thinking without prompting during the think-aloud interview.

Hypothesis 2: Self-efficacy and Group Dynamics can Influence Use of Social Metacognition during Problem solving

Previous work has shown that learning self-efficacy can influence a student's use of individual metacognition (Bouffard-Bouchard et al., 1991; Coutinho and Neuman, 2008). We observed that learning self-efficacy can also affect use of social metacognition during group problem solving. In our sample, some students described having lower self-efficacy when problem solving in groups. For example, when asked about her perceptions of group work, Hazel mentioned feeling less confident when working with peers:

Hazel described that she prefers to solve problems on her own during group work because she feels less confident when working with peers. Her lower self-efficacy for group problem solving might explain why she primarily showed evidence of metacognition individually.

We also observed that Hazel's group had some negative dynamics. One of her group members was often frustrated by Hazel's questions, even though Hazel's questions helped her group monitor their understanding. In one breakout session, Hazel asked whether excitation and emission spectrums are inversely related:

Hazel:	So they're indirectly proportional? Sorry, irreverse[ly]. I cannot speak.
Samira:	There's going to be a loss of energy is basically what it is. I don't think it matters for this question. Let's just…

While Samira answered Hazel's question, Samira remarked that the question was not important for problem solving and encouraged the group to move on (despite the question asking them to explain the difference between excitation and emission). These types of negative interactions with peers may affect a student's willingness to continue asking questions, correcting group members, or evaluating others during group work (Chiu and Khoo, 2003). Previous work has shown that student behaviors during group work can affect learning outcomes (Theobald et al., 2017; Paine and Knight, 2020). For example, Paine and Knight observed that a student who dominated the discussion during group work negatively impacted the group's quality of scientific reasoning (2020).

Group interactions can also be affected by social identity. Students experience group work differently, which can impact their comfortability or participation in groups (Eddy et al., 2015; Theobald et al., 2017). Previous research has identified that women verbally participated less frequently than men in an introductory biology course (Eddy et al., 2014), and Black students have described being excluded from group work due to their race (e.g., Stanton et al., 2022). Additionally, STEM students have reported that their depression or learning disabilities can impact interactions with their peers or their ability to focus in active learning classrooms (Araghi et al., 2023; Pfeifer et al., 2023), which could influence their use of metacognition during group work. We did not find clear evidence that sexism, racism, or abelism was at play for Hazel's group, however, similar interactions could be the result of a student being from a marginalized group.

Our findings suggest that students with lower learning self-efficacy during group work may be reluctant to correct and evaluate their group members, and thus show reduced social metacognition. We also predict that negative group dynamics, which may contribute to an individual's lower group learning self-efficacy, potentially affects use of social metacognition.

High Self-efficacy and Metacognition.

Alternatively, we hypothesize that students with higher learning self-efficacy may be better positioned to persist during the problem-solving process. We observed that some students in our sample would use a form of self-efficacy called self-coaching that helped them persist when they monitored or evaluated an incorrect or insufficient idea. Our lab previously found that self-coaching helped introductory biology students continue in the problem-solving process when they realized they did not understand something (Halmo et al., 2024). In the present study, we observed that Sophia and a few other students used self-coaching during individual problem solving, which helped them persist when they monitored an incorrect idea or evaluated that their response was insufficient. Another student in our study, Isabel, monitored that she could not remember the names of proteins involved in ER import and described her self-coaching: “I still couldn't remember the names [of the proteins] anyway, so I thought, ‘okay well I might get the question wrong but might as well move on.’” In this example, Isabel acknowledged it was okay she didn't know something and continued problem solving without knowing the specific proteins involved. In contrast, a few students would give up or stop problem solving when they monitored something they did not know during the individual interview. We predict that students like Sophia and Isabel may be able to model productive approaches such as self-coaching when working in groups (Andersen, 2003), which may benefit students that give up when confronted with the discomfort of monitoring or evaluating. Additional research on social metacognition is necessary to better understand how group dynamics and learning self-efficacy affect the use of metacognition during group work.

Dana: Primarily Shows Evidence of Metacognitive Skills during Group Work

We observed that Dana primarily shows evidence of metacognitive skills when working with others but not individually. We found that when problem solving individually, Dana would primarily acknowledge what is familiar to her when monitoring her understanding. This contrasts with Dana's approaches for monitoring in groups; she used corrections, questions, and acknowledgments of confusion to check her and her group's understanding when problem solving with peers. Dana only evaluated her reasoning when prompted to reflect after the think-aloud interview. Alternatively, Dana evaluated her group member's reasoning during breakout sessions.

Monitoring Understanding.

We found that Dana did not use corrections or questions to check her understanding when problem solving individually. Instead, Dana verbalized what she knows during the think-aloud interview. For example, while problem solving, Dana described the mental notes she made to herself: “And then making a note that it's more blue, like within the nucleus which is the little blob. And I'm inferring that because that's where the DNA is present.” In this example, Dana made an incorrect inference about the data that would be correct for the breakout session problem she previously solved with her group, but not for the isomorphic question. Here, Dana might be relying on her memory of the breakout session question rather than accurately interpreting the new isomorphic question. In other parts of the think-aloud, Dana also described what she was familiar with when problem solving. For example, Dana mentions, “Glancing at the YRC probe. But not super closely because I'm really familiar with it at this point or at least relatively familiar with it at this point.” Dana's feelings of familiarity may have contributed to her inaccurate monitoring and evaluating during her think-aloud interview (Thiede et al., 2003).

These individual approaches for monitoring contrast with the approaches Dana used to check understanding during group work. She offered verbal corrections of her own ideas and her group member's ideas during group work. For example, Dana corrected her wording when sharing a new idea saying, “You could probably also move it into sucrose solutions of different densities… concentrations, not densities.” Additionally, Dana corrected her group members’ thinking during breakout sessions:

Alannah:	–and so if there's separation, that means that there are things of different density and if there's not, then it's just like that one.
Dana:	Wait, no. You–
Alannah:	Oh, it's after centrifuging.
Dana:	Yeah. That's after centrifuging…

This correction allowed Dana and her group to reach a shared understanding. While Dana displayed evidence of this monitoring skill during group work, she did not correct herself during individual problem solving.

Dana also monitored her understanding by asking questions and acknowledging areas of confusion during group work. Dana was one of the few students that answered her own questions during group work, asking “What's the thing called? FRET?” and then explaining what FRET is. She also asked her group members questions to help her clarify her understanding, for example asking, “SDS–PAGE is charge, not size, right?” Additionally, Dana verbalized her confusion during group work, saying for example, “Now it's my turn for this to not process at all.” When students like Dana asked questions and acknowledged confusion during group work, this allowed their group members to offer clarification. Dana showed evidence of monitoring approaches similar to Sophia during group work but did not always use these approaches when solving problems on her own.

Evaluating Thinking.

When problem solving individually, Dana only evaluated when prompted to rate her confidence in her response to the isomorphic questions. When prompted to evaluate her responses during the interview, Dana evaluated based on accuracy and how she felt about her response. Dana described that she does not think she “said anything objectively wrong,” but that she has “a fear that I forgot to mention something important and just didn't realize that I wasn't saying it.” Dana evaluated her response's accuracy and also expresses her concern about the possibility of excluding important information. While she appraised her response's correctness when prompted, she did not accurately evaluate that she did not fully answer the question.

In contrast, Dana frequently evaluated her group members during the recorded breakout sessions. When working with others, Dana evaluated her group members’ ideas based on relevance and sufficiency for answering the question, similar to Sophia. For example, Dana appraised an irrelevant idea a group member had, saying “it says later on so ignore it for now.” In another example, Dana accurately evaluated that her group member's idea is not sufficient for answering the question. In this example, her group member Alannah is proposing an experiment for determining whether a signal sequence is necessary and sufficient for import into the ER. Dana evaluated Alannah's reasoning and appraised that her proposed experiment does not adequately answer the question:

Alannah:	If you just removed the first signal sequence and it was not taken up, that would prove the necessity of the first.
Dana:	Yes. Yes.
Alannah:	Then putting the viral sequence and seeing if now it gets taken up again, proves sufficiency of the second.
Dana:	I don't think it does because there might be something else that like is part of that protein. I know like, theoretically, that wouldn't be the case, but it could be that some part of that protein that is normally taken up interacts with that signal sequence in some way.

Dana's evaluation later resulted in the group asking the instructor for clarification on their ideas and then making changes to their solution. Dana provided evidence that she appraises solutions based on relevance and sufficiency for answering the question during group work.

Summary of Dana's Metacognition.

We found that Dana primarily showed evidence of social metacognition. Dana used similar approaches as Sophia to monitor and evaluate during group work. However, she did not show evidence of these forms of monitoring and evaluating when problem solving on her own. Dana seemed to rely on her memory of the answer to the breakout session questions while problem solving on her own, which may have led to overconfidence in her answer and consequently, a lack of monitoring. We hypothesize that group work may make monitoring and evaluating more fruitful because there is the potential for other students to correct or ask questions about your own ideas.

Hypothesis 3: Group Work can Provide Students with Opportunities to Practice Their Metacognition

Based on our analysis of individual and social metacognitive approaches, we predict that group work can help students practice their metacognitive skills, which can then be transferred to individual problem solving. Throughout the data, we found that students more readily corrected and evaluated other students’ thinking compared with correcting or evaluating their own ideas. Additionally, we observed that a few students in our sample primarily showed evidence of metacognition when working in groups. For example, Dana showed strong evidence of corrections and evaluations of others but did not use these approaches individually.

Dana's use of social metacognition may be related to her exposure to thinking that conflicted with her own ideas. Students can experience sociocognitive conflict when hearing viewpoints from their peers that conflict or contrast with their own thinking (Mugny and Doise, 1978; King, 2002). Conflicting ideas can promote students to further explain or reconsider their thinking (Mugny and Doise, 1978; King, 2002). For example, Goos et al. identified that groups who successfully solved mathematics problems experienced disagreements during group work (Goos et al., 2002). These researchers postulate that disagreements or conflict during group work can stimulate monitoring at the group-level, thus allowing them to identify errors in their solution or thinking (Goos et al., 2002). We hypothesize that students may be able to more readily correct, challenge, or evaluate a group member's idea, especially when it does not align with their own thinking (sociocognitive conflict). In contrast, detecting problems with your own thinking may be more challenging than being able to monitor someone else's idea (Rickey and Stacy, 2000).

Students may not use monitoring when they are presented with concepts that they are familiar with, even though they may not fully understand those concepts. For example, Dana relied heavily on her memory of problems solved in class when she solved problems individually for this study. Dana may not have felt the need to check her understanding because the isomorphic problems seemed familiar (although they were different). Previous research has identified that familiarity can affect monitoring accuracy (Thiede et al., 2003; Kornell et al., 2011; Fitzsimmons et al., 2020). We found that other students in our sample also relied heavily on their memory of the problem solving with their group while they were solving problems on their own. Half of the students in our sample solved an isomorphic question as if it was the same problem from their recent breakout session, which means they did not correctly assess the task, monitor their incorrect idea, nor evaluate their solution. These data suggest that it may be more difficult for students to identify when their ideas are incorrect, and that familiarity may negatively impact their metacognition. Students may benefit from peers monitoring and evaluating their ideas during group work when they are not positioned to do this on their own (Rickey and Stacy, 2000; Chiu and Kuo, 2009). Additionally, we predict that because students can more readily correct and evaluate their group members, they can practice these metacognitive skills during group work and potentially transfer these skills to individual problem solving. Future work could assess the transfer of metacognitive skills from collaborative learning to individual problem solving in order to further explore this hypothesis.

POSSIBLE IMPLICATIONS FOR INSTRUCTORS

In this study, we identified that seven of the 12 students we studied from an upper-level biology course used similar metacognitive approaches during individual and group problem solving (Supplemental Figure 1), however, some students displayed a more nuanced relationship between their individual and social metacognition. We expect that other college students in the life sciences will vary in their use of metacognition, and that students may use different metacognitive approaches during individual and collaborative learning. To help promote metacognition for students at different stages in their metacognitive development, we suggest implementing multiple strategies to encourage students to practice using metacognition while learning. Practice is important because we know life sciences students are still developing their metacognition when they enter college (Stanton et al., 2019). We offer potential implications for instructors to address this variation in students’ use of metacognition (Table 2).

Our data suggest that students like Sophia can transfer their metacognitive skills across contexts and explicitly teaching metacognitive skills can make transfer possible (Adey and Shayer, 1993; Schuster et al., 2020). Instructors can incorporate individual metacognitive activities such as self-evaluations after exams, or model metacognitive strategies when reviewing problems in class. For example, instructors can use practice questions that align with the exam, and then model metacognitive strategies such as self-questioning when reviewing these practice questions in class. CBE-Life Sciences Education has published resources with guidance for using and promoting student metacognition (e.g., Tanner, 2012; Stanton et al., 2021), including an evidence-based teaching guide (https://lse.ascb.org/evidence-based-teaching-guides/student-metacognition/).

In addition to incorporating individual metacognitive activities into a course, implementing group work may help students practice their metacognition while problem solving. Our data indicate that students benefit from group work due to the reciprocal nature of metacognition. During group work, students can identify areas of confusion, get their questions answered, and receive feedback on ideas; students are also able to use these metacognitive approaches for their peers’ ideas (Goos et al., 2002; Volet et al., 2009; De Backer et al., 2020). In this study we found that even upper-level students more readily correct or evaluate their group members’ ideas compared with their own ideas. Utilizing group work may help students practice and develop these metacognitive skills, so they can then apply them when problem solving individually. We also hypothesize that students with stronger metacognition skills may be able to model metacognitive approaches to their peers (Andersen, 2003), highlighting another benefit of group work. Our lab has identified that first-year students are open to learning new study strategies and metacognitive approaches from their peers (Stanton et al., 2024). so it is possible that they may also be receptive to learning metacognitive approaches for problem solving from their peers during in-class group work. For additional information on incorporating group work into life science classrooms, see CBE-Life Sciences Education's evidence-based teaching guide on group work (https://lse.ascb.org/evidence-based-teaching-guides/group-work/).

While group work has the potential to promote use of metacognition, students often need guidance during collaborative learning to fully reap its benefits (e.g., Kim and Lim, 2018). Also, students are likely to need additional support to help improve their self-efficacy or increase their sense of belonging. Instructors can use noncontent–related talk, or Instructor Talk, to help promote students’ buy-in for group work and increase sense of belonging by explaining pedagogical choices or establishing classroom culture (Seidel et al., 2015; Harrison et al., 2019). For example, explaining that the course uses group work in order to help students identify areas of confusion could improve student buy-in and may also reduce frustration when students monitor a lack of understanding with peers. Instructors can also provide guidance during group work through written metacognitive prompts that promote collaboration. Previous work has identified that use of metacognitive scripts and prompts in educational psychology and online courses can help promote collaboration (Miller and Hadwin, 2015; Kim and Lim, 2018), which can encourage participation. We are currently testing metacognitive prompts in a college biology context. Use of prompts that promote collaboration may also help build students’ group learning self-efficacy (Aikens and Kulacki, 2023) and further signal to students that working together is valued in the course, which may discourage negative group dynamics.

LIMITATIONS

This foundational study begins to reveal the nuanced relationship between a student's individual and social metacognition. While this work provides insight about these two forms of metacognition in the same students, this study has a few limitations. First, we are studying a specific learning environment and set of problem-solving tasks for a population that likely has developed their metacognition since they are senior-level honors students taking a Cell Biology course (Stanton et al., 2019). Additionally, our sample size was small due to the time intensive nature of analyzing conversational data. For these reasons, we cannot make generalizations from these cases about the relationship between individual and social metacognition for all life science students across learning contexts and identities. Other factors such as learning environment and task can affect use of metacognition (Kelemen et al., 2000; Kuhn, 2000). The group work for this study occurred in a SCALE-UP classroom that is meant to promote collaboration, which may have impacted students’ use of social metacognition. In the future, we can extend our findings with larger sample sizes and additional populations in the life sciences.

Another limitation of this study is that it is unclear whether timing of the think-aloud interviews following the breakout sessions affected students’ use of metacognition during individual problem solving. A student's framing, or their own expectations and understanding of the think-aloud, may have impacted their participation in the interview (Hammer et al., 2004; Russ et al., 2012). For instance, students may have interpreted that the purpose of the think-aloud was to test their knowledge since the individual interviews occurred after the recorded group sessions. To circumvent this issue, we told students prior to the think-aloud that we were more interested in how they approached these cell biology problems versus the grade they would potentially receive (see Interview Protocol in Supplemental Materials). While students may have framed the purpose of these interviews differently from intended, think-aloud interviews are considered the best method for measuring individual metacognition because this method captures metacognition during learning (e.g., Hofer and Sinatra, 2010; Schellings et al., 2013). Future work can consider this limitation and make changes to the timing of individual interviews and group work.

Self-regulated learning is complex and additional factors may impact students’ use of individual metacognition. For example, the think-aloud interviews remove students from a social learning context, which may impact their motivation. Motivation has been shown to be influential in self-regulated learning (e.g., Pintrich et al., 1993; Pintrich, 1999; Zimmerman and Schunk, 2009), and our participants may have felt less motivated to get the answer correct or completely answer the question in an interview compared with their motivation when working with their peers on the group problem set. Many of the students in our sample were in the last semester of their college career, so their motivation to answer cell biology questions during an interview may have been limited, representing a potential limitation of this work. Additionally, use of isomorphic questions in the individual interview may have contributed to students’ reliance on familiarity for solving the problem. Isomorphic questions allowed us to focus our analysis of metacognition on a specific cellular biology task. However, similarity of the problems may have misled students during the individual interview.

This study provides insight into upper-level biology students’ use of individual and social metacognition, which generated hypotheses related to metacognition. Additional research can test these hypotheses and extend on these findings for other life science students. Longitudinal studies that assess the development of individual and social metacognition over time could help address additional questions in the field, such as which form of metacognition develops first.

CONCLUSION

We found that seven of the 12 students in our study used similar forms of metacognition in social and individual settings. To explore the range of our data, we examined cases from four students in an upper-level cellular biology course that represented the greatest variation we observed between a student's use of individual and social metacognition. Sophia, like most of the students in our sample, showed a positive relationship between her approaches for individual and social metacognition during problem solving. We hypothesize that students like Sophia can transfer their metacognitive skills between similar individual and group problem solving tasks. In contrast, Yasmin displayed limited evidence of metacognition while problem solving, and Hazel and Dana demonstrated stronger evidence of metacognition either individually or socially.

These data suggest a more complex relationship between individual and social metacognition than originally conceived. Across our sample, we found that students more readily corrected and evaluated their group members’ ideas compared with correcting and evaluating their own ideas. We hypothesize that group work can help students practice their metacognition that can then be transferred to solitary learning contexts. Additionally, we observed that Hazel did not display much evidence of metacognition socially. We hypothesize that Hazel's learning self-efficacy and her group's dynamics may be influencing her use of social metacognition.

By qualitatively comparing individual and social metacognition for the same upper-level biology students, we identified nuances in the relationship between these two forms of metacognition such as the impact group dynamics have on social metacognition. Comparing individual and social metacognition in upper-level biology students can begin to reveal how these two forms of metacognition develop. Understanding the development of both individual and social metacognition can inform the development of metacognitive interventions that are reflective of how students learn and problem solve in the life sciences.

REFERENCES

• Adey, P., & Shayer, M. (1993). An exploration of long-term far-transfer effects following an extended intervention program in the high school science curriculum. Cognition and Instruction, 11(1), 1–29. Google Scholar
• Afflerbach, P. (2001). Verbal reports and protocol analysis. In Methods of Literacy Research (pp. 97–114). London, UK: Routledge. Google Scholar
• Aikens, M. L., & Kulacki, A. R. (2023). Identifying group work experiences that increase students’ self-efficacy for quantitative biology tasks. CBE—Life Sciences Education, 22(2), ar19. https://doi.org/10.1187/cbe.22-04-0076 Medline, Google Scholar
• Andersen, C. (2003). Distributed metacognition during peer collaboration. In Proceedings of the Annual Meeting of the Cognitive Science Society (Vol. 25, No. 25). Google Scholar
• Araghi, T., Busch, C. A., & Cooper, K. M. (2023). The aspects of active-learning science courses that exacerbate and alleviate depression in undergraduates. CBE—Life Sciences Education, 22(2), ar26. https://doi.org/10.1187/cbe.22-10-0199 Medline, Google Scholar
• Artz, A. F., & Armour-Thomas, E. (1992). Development of a cognitive-metacognitive framework for protocol analysis of mathematical problem solving in small groups. Cognition and Instruction, 9(2), 137–175. https://doi.org/10.1207/s1532690xci0902_3 Google Scholar
• Asterhan, C. S. C., & Schwarz, B. B. (2009). Argumentation and explanation in conceptual change: Indications from protocol analyses of peer-to-peer dialog. Cognitive Science, 33(3), 374–400. https://doi.org/10.1111/j.1551-6709.2009.01017.x Medline, Google Scholar
• Beichner, R. J., Saul, J. M., Abbott, D. S., Morse, J. J., Deardof, D., Allain, R., ...& Risely, J. (2007). Student-Center Activities for Large Enrollment Undergraduate Programs (SCALE-UP) Project. College Park, MD: American Association of Physics Teachers. Google Scholar
• Billing, D. (2007). Teaching for transfer of core/key skills in higher education: Cognitive skills. Higher Education, 53(4), 483–516. Google Scholar
• Bouffard-Bouchard, T., Parent, S., & Larivee, S. (1991). Influence of self-efficacy on self-regulation and performance among junior and senior high-school age students. International Journal of Behavioral Development, 14(2), 153–164. https://doi.org/10.1177/016502549101400203 Google Scholar
• Brown, A. L., & Palincsar, A. S. (1989). Guided cooperative learning and individual knowledge acquisition. In L. B. Resnick (Ed.), Cognition and Instruction: Issues and Agendas (pp. 393–451). Mahwah, NJ: Erlbaum. Google Scholar
• Brownell, S. E., Kloser, M. J., Fukami, T., & Shavelson, R. J. (2013). Context matters: Volunteer bias, small sample size, and the value of comparison groups in the assessment of research-based undergraduate introductory biology lab courses. Journal of Microbiology & Biology Education, 14(2), 176–182. https://doi.org/10.1128/jmbe.v14i2.609 Medline, Google Scholar
• Chiu, M. M., & Khoo, L. (2003). Rudeness and status effects during group problem solving: Do they bias evaluations and reduce the likelihood of correct solutions? Journal of Educational Psychology, 95(3), 506–523. https://doi.org/10.1037/0022-0663.95.3.506 Google Scholar
• Chiu, M. M., & Kuo, S. W. (2009). From metacognition to social metacognition: Similarities, differences, and learning. Journal of Education Research, 3(4), 321–338. Google Scholar
• Coutinho, S. A., & Neuman, G. (2008). A model of metacognition, achievement goal orientation, learning style and self-efficacy. Learning Environments Research, 11(2), 131–151. https://doi.org/10.1007/s10984-008-9042-7 Google Scholar
• Cross, D. R., & Paris, S. G. (1988). Developmental and instructional analyses of children's metacognition and reading comprehension. Journal of Educational Psychology, 80(2), 131–142. https://doi.org/10.1037/0022-0663.80.2.131 Google Scholar
• De Backer, L., Van Keer, H., & Valcke, M. (2020). Variations in socially shared metacognitive regulation and their relation with university students’ performance. Metacognition and Learning, 15(2), 233–259. https://doi.org/10.1007/s11409-020-09229-5 Google Scholar
• De Backer, L., Van Keer, H., & Valcke, M. (2021). Examining the relation between students’ active engagement in shared metacognitive regulation and individual learner characteristics. International Journal of Educational Research, 110, 101892. https://doi.org/10.1016/j.ijer.2021.101892 Google Scholar
• Dye, K. M., & Stanton, J. D. (2017). Metacognition in upper-division biology students: Awareness does not always lead to control. CBE—Life Sciences Education, 16(2), ar31. https://doi.org/10.1187/cbe.16-09-0286 Link, Google Scholar
• Eddy, S. L., Brownell, S. E., Thummaphan, P., Lan, M.-C., & Wenderoth, M. P. (2015). Caution, student experience may vary: Social identities impact a student's experience in peer discussions. CBE—Life Sciences Education, 14(4), ar45. https://doi.org/10.1187/cbe.15-05-0108 Link, Google Scholar
• Eddy, S. L., Brownell, S. E., & Wenderoth, M. P. (2014). Gender gaps in achievement and participation in multiple introductory biology classrooms. CBE—Life Sciences Education, 13(3), 478–492. https://doi.org/10.1187/cbe.13-10-0204 Link, Google Scholar
• Ericsson, K. A. (1988). Concurrent verbal reports on text comprehension: A review. Text – Interdisciplinary Journal for the Study of Discourse, 8(4), 295–326. https://doi.org/10.1515/text.1.1988.8.4.295 Google Scholar
• Fitzsimmons, C. J., Thompson, C. A., & Sidney, P. G. (2020). Confident or familiar? The role of familiarity ratings in adults’ confidence judgments when estimating fraction magnitudes. Metacognition and Learning, 15(2), 215–231. https://doi.org/10.1007/s11409-020-09225-9 Google Scholar
• Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. American Psychologist, 34(10), 906–911. https://doi.org/10.1037/0003-066X.34.10.906 Google Scholar
• Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry, 12(2), 219–245. https://doi.org/10.1177/1077800405284363 Google Scholar
• Foster, N. L., Was, C. A., Dunlosky, J., & Isaacson, R. M. (2017). Even after thirteen class exams, students are still overconfident: The role of memory for past exam performance in student predictions. Metacognition and Learning, 12(1), 1–19. https://doi.org/10.1007/s11409-016-9158-6 Google Scholar
• Gentner, D., Rattermann, M. J., & Forbus, K. D. (1993). The roles of similarity in transfer: Separating retrievability from inferential soundness. Cognitive Psychology, 25(4), 524–575. https://doi.org/10.1006/cogp.1993.1013 Medline, Google Scholar
• Goos, M., Galbraith, P., & Renshaw, P. (2002). Socially mediated metacognition: Creating collaborative zones of proximal development in small group problem solving. Educational Studies in Mathematics, 49(2), 193–223. https://doi.org/10.1023/A:1016209010120 Google Scholar
• Halmo, S. M., Bremers, E., Fuller, S., & Stanton, J. D. (2022). Oh, that makes sense: Social metacognition in small-group problem solving. CBE—Life Sciences Education, 21(3), ar58. Medline, Google Scholar
• Halmo, S. M., Yamini, K. A., & Stanton, J. D. (2023). Metacognition and problem solving: How self-coaching helps first-year students move past the discomfort of monitoring [Preprint]. Scientific Communication and Education. https://doi.org/10.1101/2023.08.16.553589 Google Scholar
• Halmo, S. M., Yamini, K. A., & Stanton, J. D. (2024). Metacognition and self-efficacy in action: How first-year students monitor and use self-coaching to move past metacognitive discomfort during problem solving. CBE—Life Sciences Education, 23(2), ar13. https://doi.org/10.1187/cbe.23-08-0158 Medline, Google Scholar
• Hammer, D. M., Elby, A., Scherr, R. E., & Redish, E. F. (2004). Resources, framing, and transfer. Retrieved August 7, 2024 from, https://api.semanticscholar.org/CorpusID:53048300 Google Scholar
• Harrison, C. D., Nguyen, T. A., Seidel, S. B., Escobedo, A. M., Hartman, C., Lam, K., …& Tanner, K. D. (2019). Investigating instructor talk in novel contexts: Widespread use, unexpected categories, and an emergent sampling strategy. CBE—Life Sciences Education, 18(3), ar47. https://doi.org/10.1187/cbe.18-10-0215 Link, Google Scholar
• Heidbrink, A., & Weinrich, M. (2021). Encouraging biochemistry students’ metacognition: Reflecting on how another student might not carefully reflect. Journal of Chemical Education, 98(9), 2765–2774. https://doi.org/10.1021/acs.jchemed.1c00311 Google Scholar
• Hofer, B. K., & Sinatra, G. M. (2010). Epistemology, metacognition, and self-regulation: Musings on an emerging field. Metacognition and Learning, 5(1), 113–120. https://doi.org/10.1007/s11409-009-9051-7 Google Scholar
• Jost, J. T., Kruglanski, A. W., & Nelson, T. O. (1998). Social metacognition: An expansionist review. Personality and Social Psychology Review, 2(2), 137–154. https://doi.org/10.1207/s15327957pspr0202_6 Medline, Google Scholar
• Kelemen, W. L., Frost, P. J., & Weaver, C. A. (2000). Individual differences in metacognition: Evidence against a general metacognitive ability. Memory & Cognition, 28(1), 92–107. https://doi.org/10.3758/BF03211579 Medline, Google Scholar
• Kim, D., & Lim, C. (2018). Promoting socially shared metacognitive regulation in collaborative project-based learning: A framework for the design of structured guidance. Teaching in Higher Education, 23(2), 194–211. https://doi.org/10.1080/13562517.2017.1379484 Google Scholar
• Kim, Y. R., Park, M. S., Moore, T. J., & Varma, S. (2013). Multiple levels of metacognition and their elicitation through complex problem-solving tasks. The Journal of Mathematical Behavior, 32(3), 377–396. https://doi.org/10.1016/j.jmathb.2013.04.002 Google Scholar
• King, A. (2002). Structuring peer interaction to promote high-level cognitive processing. Theory Into Practice, 41(1), 33–39. https://doi.org/10.1207/s15430421tip4101_6 Google Scholar
• Kornell, N., Rhodes, M. G., Castel, A. D., & Tauber, S. K. (2011). The ease-of-processing heuristic and the stability bias: Dissociating memory, memory beliefs, and memory judgments. Psychological Science, 22(6), 787–794. https://doi.org/10.1177/0956797611407929 Medline, Google Scholar
• Kotovsky, K., Hayes, J. R., & Simon, H. A. (1985). Why are some problems hard? Evidence from Tower of Hanoi. Cognitive Psychology, 17(2), 248–294. https://doi.org/10.1016/0010-0285(85)90009-X Google Scholar
• Kruger, A. C. (1993). Peer collaboration: Conflict, cooperation, or both? Social Development, 2(3), 165–182. https://doi.org/10.1111/j.1467-9507.1993.tb00012.x Google Scholar
• Kuhn, D. (2000). Metacognitive development. Current Directions in Psychological Science, 9(5), 178–181. https://doi.org/10.1111/1467-8721.00088 Google Scholar
• Lippmann Kung, R., & Linder, C. (2007). Metacognitive activity in the physics student laboratory: Is increased metacognition necessarily better? Metacognition and Learning, 2(1), 41–56. https://doi.org/10.1007/s11409-007-9006-9 Google Scholar
• Miller, M., & Hadwin, A. (2015). Scripting and awareness tools for regulating collaborative learning: Changing the landscape of support in CSCL. Computers in Human Behavior, 52, 573–588. https://doi.org/10.1016/j.chb.2015.01.050 Google Scholar
• Moog, R. S.Spencer, J. N. (eds.). (2008). Process Oriented Guided Inquiry Learning (POGIL), 994. Washington, DC: American Chemical Society. https://doi.org/10.1021/bk-2008-0994 Google Scholar
• Morphew, J. W. (2021). Changes in metacognitive monitoring accuracy in an introductory physics course. Metacognition and Learning, 16(1), 89–111. https://doi.org/10.1007/s11409-020-09239-3 Google Scholar
• Mugny, G., & Doise, W. (1978). Socio-cognitive conflict and structure of individual and collective performances. European Journal of Social Psychology, 8(2), 181–192. https://doi.org/10.1002/ejsp.2420080204 Google Scholar
• Ohtani, K., & Hisasaka, T. (2018). Beyond intelligence: A meta-analytic review of the relationship among metacognition, intelligence, and academic performance. Metacognition and Learning, 13(2), 179–212. https://doi.org/10.1007/s11409-018-9183-8 Google Scholar
• Paine, A. R., & Knight, J. K. (2020). Student behaviors and interactions influence group discussions in an introductory biology lab setting. CBE—Life Sciences Education, 19(4), ar58. https://doi.org/10.1187/cbe.20-03-0054 Link, Google Scholar
• Pfeifer, M. A., Cordero, J. J., & Stanton, J. D. (2023). What I wish my instructor knew: How active learning influences the classroom experiences and self-advocacy of STEM majors with ADHD and specific learning disabilities. CBE—Life Sciences Education, 22(1), ar2. https://doi.org/10.1187/cbe.21-12-0329 Medline, Google Scholar
• Pfeifer, M. A., & Stanton, J. D. (2021). Necessary and sufficient? Solving the mystery of the mitochondrial pyruvate transporter. CourseSource. https://qubeshub.org/community/groups/coursesource/publications?id=2667&v=1 Google Scholar
• Pintrich, P. R. (1999). The role of motivation in promoting and sustaining self-regulated learning. International Journal of Educational Research, 31(6), 459–470. https://doi.org/10.1016/S0883-0355(99)00015-4 Google Scholar
• Pintrich, P. R., Smith, D. A. F., Garcia, T., & Mckeachie, W. J. (1993). Reliability and predictive validity of the motivated strategies for learning questionnaire (Mslq). Educational and Psychological Measurement, 53(3), 801–813. https://doi.org/10.1177/0013164493053003024 Google Scholar
• Pressley, M., Ghatala, E. S., Woloshyn, V., & Pirie, J. (1990). Sometimes adults miss the main ideas and do not realize it: Confidence in responses to short-answer and multiple-choice comprehension questions. Reading Research Quarterly, 25(3), 232. https://doi.org/10.2307/748004 Google Scholar
• Rickey, D., & Stacy, A. M. (2000). The role of metacognition in learning chemistry. Journal of Chemical Education, 77(7), 915. https://doi.org/10.1021/ed077p915 Google Scholar
• Rosenthal, R. (1965). The volunteer subject. Human Relations, 18(4), 389–406. https://doi.org/10.1177/001872676501800407 Google Scholar
• Russ, R. S., Lee, V. R., & Sherin, B. L. (2012). Framing in cognitive clinical interviews about intuitive science knowledge: Dynamic student understandings of the discourse interaction. Science Education, 96(4), 573–599. https://doi.org/10.1002/sce.21014 Google Scholar
• Saldaña, J. (2021). The Coding Manual for Qualitative Researchers (4th ed). Thousand Oaks, CA: SAGE Publishing Ltd. Google Scholar
• Schellings, G. L. M., van Hout-Wolters, B. H. A. M., Veenman, M. V. J., & Meijer, J. (2013). Assessing metacognitive activities: The in-depth comparison of a task-specific questionnaire with think-aloud protocols. European Journal of Psychology of Education, 28(3), 963–990. https://doi.org/10.1007/s10212-012-0149-y Google Scholar
• Schraw, G., & Moshman, D. (1995). Metacognitive theories. Educational Psychology Review, 7(4), 351–371. https://doi.org/10.1007/BF02212307 Google Scholar
• Schuster, C., Stebner, F., Leutner, D., & Wirth, J. (2020). Transfer of metacognitive skills in self-regulated learning: An experimental training study. Metacognition and Learning, 15(3), 455–477. https://doi.org/10.1007/s11409-020-09237-5 Google Scholar
• Scott, B., & Berman, A. (2013). Examining the domain-specificity of metacognition using academic domains and task-specific individual differences. Australian Journal of Educational & Developmental Psychology, 13, 28–43. Google Scholar
• Seidel, S. B., Reggi, A. L., Schinske, J. N., Burrus, L. W., & Tanner, K. D. (2015). Beyond the biology: A systematic investigation of noncontent instructor talk in an introductory biology course. CBE—Life Sciences Education, 14(4), ar43. https://doi.org/10.1187/cbe.15-03-0049 Link, Google Scholar
• Stake, R. E. (2006). Multiple Case Study Analysis. New York, NY: The Guilford Press. Google Scholar
• Stanton, J. D., & Dye, K. M. (2021). Investigating the function of a transport protein: Where is ABCB6 located in human cells? CourseSource. https://qubeshub.org/community/groups/coursesource/publications?id=2597&tab_active=about&v=1 Google Scholar
• Stanton, J. D., Dye, K. M., & Johnson, M. (2019). Knowledge of learning makes a difference: A comparison of metacognition in introductory and senior-level biology students. CBE—Life Sciences Education, 18(2), ar24. https://doi.org/10.1187/cbe.18-12-0239 Link, Google Scholar
• Stanton, J. D., Halmo, S. M., Carter, R. J., Yamini, K. A., & Ososanya, D. (2024). Opportunities for guiding development: Insights from first-year life science majors’ use of metacognition. Journal of Microbiology & Biology Education, e0005324. https://doi.org/10.1128/jmbe.00053-24 Medline, Google Scholar
• Stanton, J. D., Means, D. R., Babatola, O., Osondu, C., Oni, O., & Mekonnen, B. (2022). Drawing on internal strengths and creating spaces for growth: How Black science majors navigate the racial climate at a predominantly White institution to succeed. CBE—Life Sciences Education, 21(1), ar3. https://doi.org/10.1187/cbe.21-02-0049 Medline, Google Scholar
• Stanton, J. D., Neider, X. N., Gallegos, I. J., & Clark, N. C. (2015). Differences in metacognitive regulation in introductory biology students: When prompts are not enough. CBE—Life Sciences Education, 14(2), ar15. https://doi.org/10.1187/cbe.14-08-0135 Link, Google Scholar
• Stanton, J. D., Sebesta, A. J., & Dunlosky, J. (2021). Fostering metacognition to support student learning and performance. CBE—Life Sciences Education, 20(2), fe3. https://doi.org/10.1187/cbe.20-12-0289 Link, Google Scholar
• Tanner, K. D. (2012). Promoting student metacognition. CBE—Life Sciences Education, 11(2), 113–120. https://doi.org/10.1187/cbe.12-03-0033 Link, Google Scholar
• Theobald, E. J., Eddy, S. L., Grunspan, D. Z., Wiggins, B. L., & Crowe, A. J. (2017). Student perception of group dynamics predicts individual performance: Comfort and equity matter. PLoS One, 12(7), e0181336. https://doi.org/10.1371/journal.pone.0181336 Medline, Google Scholar
• Thiede, K. W., Anderson, M. C. M., & Therriault, D. (2003). Accuracy of metacognitive monitoring affects learning of texts. Journal of Educational Psychology, 95(1), 66–73. https://doi.org/10.1037/0022-0663.95.1.66 Google Scholar
• Tobias, S., & Everson, H. T. (2009). The importance of knowing what you know. Handbook of Metacognition in Education, 107–128. New York, NY: Routledge, Taylor & Francis Group. Google Scholar
• Tracy, S. J. (2010). Qualitative quality: Eight “Big-Tent” criteria for excellent qualitative research. Qualitative Inquiry, 16(10), 837–851. https://doi.org/10.1177/1077800410383121 Google Scholar
• Van Boxtel, C., Van Der Linden, J., & Kanselaar, G. (2000). Collaborative learning tasks and the elaboration of conceptual knowledge. Learning and Instruction, 10(4), 311–330. https://doi.org/10.1016/S0959-4752(00)00002-5 Google Scholar
• Volet, S., Vauras, M., & Salonen, P. (2009). Self- and social regulation in learning contexts: An integrative perspective. Educational Psychologist, 44(4), 215–226. https://doi.org/10.1080/00461520903213584 Google Scholar
• Wang, M. C., Haertel, G. D., & Walberg, H. J. (1990). What influences learning? A content analysis of review literature. The Journal of Educational Research, 84(1), 30–43. https://doi.org/10.1080/00220671.1990.10885988 Google Scholar
• Wilson, K. J., Brickman, P., & Brame, C. J. (2018). Group work. CBE—Life Sciences Education, 17(1), fe1. https://doi.org/10.1187/cbe.17-12-0258 Link, Google Scholar
• Yin, R. K. (2009). Case Study Research: Design and Methods (4th ed.). Thousand Oaks, CA: Sage Publications. Google Scholar
• Zimmerman, B. J.Schunk, D. H. (eds.). (2009). Motivation and Self-regulated Learning: Theory, Research, and Applications (Reprint). London, UK: Routledge. Google Scholar