In its recent resurgence, research on metacognition has claimed for it a number of educationally important characteristics that make its investigation a priority in science teaching and learning. Many studies have shown the potential benefits of well-facilitated metacognition for learning and problem solving in science (Schraw et al., 2006; Taasoobshirazi & Farley, 2013; Thomas, 2013; Adler et al., 2016; Zepeda et al., 2019). Recent studies have shown substantial gains in learning due to metacognition across primary, secondary and tertiary years, including: between primary mathematics classrooms where the metacognitive talk was stronger versus those where it was weaker (Smith & Mancy, 2018); in High School Physics, where student metacognition measures correlated with higher levels of performance (González et al., 2017); and in Chemistry classes for those in initial teacher education (Adadan, 2020). For example, after inquiry-based instruction, High School participants with high-level metacognitive knowledge, when compared to participants with low-level metacognitive knowledge, were more likely to change their conceptions to science-oriented ones (González et al., 2017). Additionally, the participants with high-level metacognitive knowledge developed a more coherent and consistent understanding of lunar phases and gas behaviour, and retained their scientific understandings months after instruction (González et al., 2017).
The importance and benefits of metacognition are mirrored in recent studies that are outside Science Education, but relevant to it: student metacognition is vital for design thinking, including experimental design (Kavousi et al., 2019); metacognition is beneficial across learning in schooling and university (Perry et al., 2019); and metacognition is vital for problem-solving involving the ‘extensive entanglement between metacognition and manipulation in working memory’ (Shea & Frith, 2019, p. 568). A synthesis of meta-analysis of metacognition studies asserted that some facilitated metacognitive strategies doubled the expected amount of learning of students when compared to standard instruction (Hattie, 2008; Hattie & Zierer, 2017). Moreover, earlier research claimed that metacognition promises a more substantial contribution to learning than general intelligence (Veenman et al., 2004, p. 92). The ubiquitous agreement about the nature and potential of facilitated student metacognition for learning gains, however, disguises disagreements about specific features of metacognition and its measurement.
Definition Of Metacognition
There has been broad agreement on what comprises core components of metacognition including across school and university study (Lai, 2011). In general terms, metacognition is the form of cognition whose subject is cognition (Acar, 2019). Ongoing agreement concurs with Flavell’s (1976) early characterization that metacognition consists of metacognitive knowledge and metacognitive experiences and that metacognition ‘regulates any aspect of any cognitive endeavor’ (p. 906). A study set in high school physics similarly operationalised metacognition as ‘two stages: First, an awareness of certain skills or strategies and resources to perform certain tasks effectively; second, the ability to use self-regulatory mechanisms to ensure the successful completion of that given task’ (Wade-Jaimes et al., 2018, p.715). Self-awareness of cognition, used synonymously with metacogntive knowledge, and self-regulation are repeated, non-contested, themes in the literature on metacognition.
Metacognition includes a learners' inner awareness about their content knowledge, learning processes, current cognitive state (Hennessey, 2003) and ‘management of one's own thought’ (Kuhn & Dean, 2004, p. 270). Metacognition comprises processes about person, task, and strategy, as well as declarative, procedural, and conditional knowledge (Flavell, 1979). Cognitive strategies are used for complex activities such as problem solving or planning tasks, whereas metacognitive strategies are employed in order to self-monitor, self-evaluate, control and understand the cognition used during those complex activities (Panahandeh & Asl, 2014).
Recent exploration showed the primacy of the affective domain to enable higher levels of metacognition, such as self-regulation (Author). There is a strong and integral affective element to cognition and metacognition, however affect is outside of the scope of this current article. Research has also explored group metacognition (Smith & Mancy, 2018) but this too sits outside of this current article’s focus on individual metacognition.
Problems With Metacognition
Despite the potential, the advantages of metacognition may be overstated due to problems associated with measurement (Akturk & Sahin, 2011), and these are, in large part, due to the lack of agreement on specific aspects of metacognitive models (Azevedo, 2020). Some of the problems with metacognitive models and measures lie with conflating non-metacognitive elements with metacognitive ones and there are different problems associated with the relationships articulated between metacognitive elements.
Metacognition’s role, processes and elements
Metacognition is frequently conflated with other forms of complex cognition. A recent commentary on the ongoing ‘discussion in the field has been the complex interaction between cognition and metacognition that continues to challenge researchers … This represents a dilemma of having a higher-order agent overlooking and governing the cognitive system while also simultaneously being part of it’ (Azevedo, 2020, p. 92). One controversial aspect of that dilemma concerns the status of metacognition as an order of thinking that is higher than other forms of cognition. When metacognition is activated, its role is to understand, diagnose and govern other forms of thinking. However, there are times when learners are not, and need not, be metacognitive, so at those times other forms of cognition have a priority, for example, due to the limited capacity of working memory (Martinez, 2006). In terms of governing thought, metacognition may be called higher order, but in terms of processes, it is not higher than other forms of cognition. For example, analytical processes are used during cognition and metacognition: for cognition the analysis is of external phenomenon, whereas for metacognition it is of the internal world of one’s own thinking and sentiments. Metacognition seen as higher-order-thinking is an emphasises on the role of metacognition to govern cognition, like a ‘top layer’ of cognition, an idea evident in a hierarchical model for cognition (Kayashima et al., 2011). The higher-order perspective is useful to understand what metacognition does, but not its mechanisms or how it may be developed. The framework presented below focuses on the cognitive processes that comprise metacognition, rather than its role.
Any framework for, or measure of, metacognition must consider aspects that are genuinely metacognitive only. Including non-metacognitive elements in metacognitive models compromises construct validity and therefore affects measurements in questionnaire design and observation studies that are based on those models. A commonly used, efficient method of evaluating metacognitive skills is self-report questionnaire, however there ‘is a wide range of questionnaires that measure a variety of components of metacognition’ (Craig et al., 2020, p156), and the variation flags uncertainty about the construct validity of such surveys. As an example of including non-metacogntive elements, one study’s Likert scale items that were used to measure metacognitive knowledge included ‘I really pay attention to important information’ (Acar, 2019, p. 658). Paying attention to important information may be a measure of metacognitive knowledge, but could just as easily be interpreted as learning for the test and lacking in metacognitive characteristics.
Observation-based research too seems to include non-metacognitive elements. In a well-cited study employing observation-based research that used quantitative measures for metacognition, the ‘frequency of scrolling back to earlier experiments’ was used as ‘a positive indicator of metacognitive skillfulness’ (Veenman et al., 2004, p. 98). Therefore, a student who did not scroll back to previous experimental data was measured on this item as having lower metacognition than a student who did. However, some students who did not scroll back may not need to because they deeply understood the experimental design or numerous other possibilities which may not relate at all to metacognition. The same study (Veenman et al., 2004) characterized hypothesizing and drawing conclusions as metacognition. However these sophisticated forms of thinking are forms of cognition not metacognition, for student awareness or control of thinking processes do not, by default, occur when hypothesising or drawing conclusions. Wade-Jaimes, et al., (2018, p.718) states that the teacher ‘would ask metacognitive questions that require the student to make predictions about a novel circuit configuration, such as “What will happen in the circuit when you add/remove this component?”’ Predicting the results of removing a component is a focus on the phenomena, not on students’ cognition and therefore does not necessarily prompt student metacognition.
Relationship between metacognitive elements
Another common feature in the literature is the unclear relationships between metacognitive elements. While ‘… the majority of researchers separate metacognitive knowledge from metacognitive skills’ (Perry et al., 2019, p. 485), the separation does not indicate the relationships between them. Moreover, metacognitive elements are sometimes represented as phases (e.g. Azevedo, 2005) implying a linear sequence that may be true for the role of metacognition but is not true of metacognitive processes.
When considering the processes, there is an unambiguous dependency of metacognitive skills on metacognitive self-Awareness: without metacognitive self-Awareness, cognitive skills necessarily function without the ‘meta’ component. To be metacognitive, these skills must operate within an awareness of cognition as a pre-condition. For example, high-level evaluation, is not dependent on metacognition, however, self-Evaluation of cognitive performance is a metacognitive skill because it is focused on cognition itself, enabled by self-Awareness. A comprehensive framework (Tarricone, 2011) clearly distinguishes between metacognitive knowledge and skills (Azevedo, 2020) by representing skills in parallel to, but not intersecting with, metacognitive knowledge. However, this clear distinction does not demonstrate the connections that exists between metacognitive knowledge and skills, leaving a need for clarification about the relationship between them (Bannister-Tyrrell et al., 2014), a clarification that may provide insight into the development and understanding of metacognitive processes.
Fundamental Construct problems
To deal effectively with the dilemma and problems above, theoretical progress in the field of metacognition has been called for before further data is generated (Sobocinski et al., 2020). Without a sound theoretical model, measures of, and outcomes attributed to, metacognition may be over-estimated due to the reliance on quantitative data generated with instruments that have debatable construct validity. In the rush to quantify and generate generalizable results and guidelines, there may have been and continue to be measures of metacognition that are not valid or reliable. The lack of clarity is amplified by the many different terms that are used to indicate metacognition, and include, but are not limited to, metacognitive awareness, metacognitive knowledge, higher-order skills, thinking strategies, learning strategies, self-control, self-regulatory strategies, and monitoring of comprehension (Veenman et al., 2004). The different terminology and measures add confusion to understanding of the nature of metacognition.
Some science education studies have recognized the limitations in terms of measuring interventions on metacognition. For example, a study set in junior high school physics classes that were using ‘metacognitive tools’ returned ambiguous results in short-to medium timeframes, finding that ‘students most likely failed to either deeply process the information provided during metacognitive training and were either unable or unwilling to apply this knowledge to their learning supported by the simulation.’ (Moser et al., 2017, p.959). The study concluded that ‘Further, training over an extended period of time might have been beneficial’ (Moser et al., 2017, 959). The study’s negative finding of the short-term use of metacognitive tools in science education either provides a cue about time for development of metacognitive skills, or that the training, measures or constructs it was based on are faulty.
The need for theoretical progress in the field of metacognition and the lack of clarity about the constructs for and the measurements of metacognition has led quantitative researchers in science education to call for deeper understanding of metacognition through qualitative studies that capture detailed and fine-grained data (Gonzalez et al., 2017; Adadan, 2020). Data comprising interviews with students who have rich learning experiences can provide some of that detail and nuance (Adadan, 2020).
To summarise, a clearer framework of metacognition is needed that must 1) deal with metacognition only, 2) show the relationship between elements and 3) provide insights into rich qualitative data before it may be further tested with quantitative studies. This paper addresses these gaps through its two aims. The primary aim is to present a synthesis of common elements in the literature into a viable framework of metacognition. The secondary aim is to conduct a preliminary test of the viability of the framework to articulate the nature of metacognition as determined by its capacity to generate analytical insights into rich qualitative data. A theoretically sound and empirically-viable framework of metacognition could better inform teachers about facilitating student metacognitive learning and enable researchers to probe metacognitive aspects of teaching and learning.
The empirical data to test the framework’s viability comes from interviews with Bachelor of Animal Science students in the final year of a degree that used the Research Skill Development (RSD: Author: see Appendix 1) framework to conceptually frame teaching, learning, and assessment in pertinent courses across the degree (Author). The RSD articulates six facets of research thinking, which are represented in this paper as verb-pairs linked by an ampersand: embark & clarify, find & generate, evaluate & reflect, organize & manage, anlalyse & synthesise and communicate & apply (see Author for details). The interview data was deemed to be appropriate for the testing of the viability of the metacognitive framework due to its similar status to data from a qualitative study of RSD use across a Medical Science degree (Author). In the latter study, evidence emerged that student metacognition was substantially developed by repeated exposure to the RSD framing of teaching, learning, and assessment in several, diverse units of study. However, no detailed analysis of metacognition was conducted in that study due to a lack of appropriate framework (Author). The analysis of the Animal Science student interview data in this current article, using the hierarchical metaphorical framework, is a step towards to determining the viability of the framework. The next section outlines the synthesis of this framework of metacognition, followed by the context and methodology of the empirical study that tested the viability of the framework, and then results, discussion and conclusions.
Amert Framework Of Metacognition
No one metacognitive framework provided the analytical insight into the graduate interview transcripts. Therefore the research team:
-
Identified non-controversial elements of metacognition that were prevalent in the literature
-
Excluded those elements that were not metacognitive by default, such as hypothesizing and planning.
-
Identified where pre-conditions are necessary for an elements of metacognition to be realized.
The metaphorical framework described below emerged from these processes to identify and relate metacognitive knowledge and skills.
Metacognitive Skill Hierarchy
Building on the foundation of metacognitive knowledge, commonly called self-Awareness, as a pre-condition for metacognitive skills, there is an articulated relationship among the metacognitive skills in the literature that is hierarchical in nature. While not all recent research agrees with a hierarchy among metacognitive skills (e.g. O'Leary & Sloutsky, 2019), it is broadly understood that ‘monitoring of cognition plays a causal role in self-regulation of cognitive processes’ (Rhodes, 2019, p.168). Monitoring of cognition is then a pre-condition for self-regulation of cognition, and such monitoring must be áppropriate for self-regulation to be effective (Mueller et al., 2016). In turn, for monitoring to be accurate it is based on the metacognitive knowledge that comprises student self-Awareness of cognition. A recent literature review of metacognition in primary and secondary school contexts delineated metacognitive skills commonly used by educators who:
… include strategies that help pupils to monitor, plan, evaluate and regulate … as well as strategies that consciously help pupils solve novel problems (Perry et al., 2019, p. 486).
This quote provides a consolidated list of practitioner-oriented metacognitive skills: monitor, plan, evaluate, regulate and solve novel problems. The list is similar to that from research finding from primary school science metacogntion (Georghiades, 2004). ‘Solve novel problems’ is here called cognitive Transfer as it is the standard term in the literature on metacognition (Ford et al., 1998; Smith et al., 2007; Heggen, 2008). Moreover, ‘plan’ in this paper is portrayed as an aspect of self-Regulation of cognition (as discussed below). A similar list at university level is monitor, evaluate, control and understand the cognition used during those complex activities (Panahandeh & Asl, 2014, p. 1409-10). Here control, as a parallel term to regulate, is for these authors an umbrella terms that covers transfer. Metacognition is not portrayed in the literature as qualitatively different across schooling and university (Perry et al., 2019).
The ‘causal’ connections, or more accurately, the preconditional processes, make for a hierarchical relationship between the elements of the framework introduced next. The elements of metacognition are, in order of hierarchical sequence, self-Aware of cognition, self-Monitor cognition, self-Evaluate cognition, self-Regulate cognition and Transfer of cognition (AMERT). In the AMERT hierarchy, higher levels occur on the basis of lower levels and depend on how effectively these lower metacognitive processes are functioning and remain functioning. This is the same sense as Maslow’s hierarchy of needs (Maslow & Lewis, 1987) where the lowest level of ‘physiological needs’ is not only a pre-condition for all higher levels to happen effectively but is also a co-condition that must remain in place. The AMERT hierarchy of process means that self-Regulating cognition is ineffective without self-Evaluating and self-Monitoring cognition which in turn are ineffective without being self-Aware of cognition. This hierarchy does not mean a sequence of steps up, such as first being self-Aware then self-Monitoring then self-Evaluating. Rather as a higher level on AMERT operates the lower levels also operate simultaneously, i.e. self-Evaluating cognition does not happen without self-Awareness of cognition and self-Monitoring cognition occurring at the same time. Regulation of cognition happens all the time at a level below conscious control, but self-Regulation here is intentional, metacognitive and reliant on the lower levels of the hierarchy.
A specific form of self-Regulation of cognition is the intentional Transfer of cognitive skills from one context to a less familiar context. Cognitive transfer is seen to be an outcome of active self-Monitoring (Perkins & Salomon, 1992) and self-Regulating (McKeachie, 1987). In this current article, Transfer of cognition is regarded as such a specific and difficult form of regulation that it is placed at the top of the AMERT framework, directly above but connected to self-Regulation of cognition. Moreover, the concept of planning cognition is treated in AMERT as a form of self-Regulation of cognition, in keeping with a literature review on metacognitive thinking (Lai, 2011). Planning cognition for tasks that are relatively close to other experiences is called near Transfer of cognition, and planning for unfamiliar tasks is far Transfer of cognition (McKeachie, 1987). While planning cognition is often placed as an early phase of metacognition, this is a focus on the role of metacognition to govern cognition; AMERT’s focus on process represents planning as part of the self-regulating process which relies on the lower levels of the framework to be effectively activated.
Amert Hierarchical Framework
Each level of AMERT is expounded in the following section.
Self-Aware of cognition
Self-Awareness of cognition involves students knowing what they already know and especially a realization of the processes and skills involved in cognition (Efklides, 2008). The conceptual power of all forms of metacognition is dependent on the quality of the metacognitive knowledge that students are self-Aware about. As students become aware of their cognitive processes, their knowledge may be trivial, minimal, incorrect or incoherent if it is not introduced and supported expertly. Self-Awareness implies self-articulation of the internal realms of thinking and benefits from appropriate and communicative labels for different kinds of cognition (Cellar & Barrett, 1987).
Teaching metacognitive self-Awareness can involve making explicit such cognitive labels which become a ‘metalanguage’ that facilitate student self-awareness (Cellar & Barrett, 1987). Cognitive labels name cognitive processes and so enable awareness of one’s own thinking, a requirement for metacognitive knowledge. Cognitive labels may be formed by the learner herself as her pre-verbal, core metacognitive knowledge emerges and becomes explicit (Goupil & Kouider, 2019), or may involve labels imparted by siblings, parents, or teachers. The student interviews below are from a program that, in effect, used the RSD facets as cognitive labels. Cognitive labels must be internalized to effectively become metacognitive self-Awareness of cognition, i.e. there must be a development from labels given, say by a teacher, to a deeper understanding by the students (Cellar & Barrett, 1987). Teacher impartation of cognitive labels may be a good beginning, but it may be a long way from student self-Awareness, especially if there is dissonance between the students’ existing knowledge of their cognition and the teacher’s labels. Moreover, what a teacher means by a label may be very different to the understanding of that label that each student constructs.
Self-Monitoring Cognition
Self-Monitoring draws on student self-Awareness of cognitive activities to explicitly gauge what thinking is currently taking place. Self-Monitoring will be off-track and impractical if an individual's own cognitive labels are not functional or students have a shallow understanding of appropriate labels. Self-monitoring activates thought processes that encourage critical thinking in the student (Facione & Facione, 1996), so without self-Monitoring, there is no internal data for effective self-evaluation of cognition. Therefore, self-Monitoring is foundational to, and precedes or occurs simultaneously with self-Evaluating in AMERT. Self-Monitoring indicates where a student is currently at cognitively, which includes the ongoing process of checking progress towards a set goal (Pintrich, 1999) and whether a specific reference point is reached during self-regulation towards some cognitive location (Zimmerman, 1995).
Self-Evaluation of Cognition
Self-Evaluation of cognition involves decisions about whether current cognition is good enough to get the job done based on each student’s thinking patterns (Elder & Paul, 2004). The information on which to make such judgements is provided by self-Monitoring of cognition and provides the data for self-Evaluative contrast and comparison. The contrast between where one is cognitively and where one needs to be provides the internal, self-Evaluative impetus for a change in cognition that involves self-Regulation. Comparing current cognition and perceived needed cognition provides a realization of insufficiency if cognition doesn’t or wouldn’t get the desired outcome, or of adequacy if it does.
Self-Regulation of Cognition
Self-Regulation of cognition is typically characterized as a change that works to improve or optimize student learning processes (McInerney et al., 1997). However, self-regulation involves not only intentional change but also intentional maintenance and consolidation of currently effective cognitive strategies, especially when managing one’s behavior and undertaking tasks to achieve an intended goal, process, or desire (Efklides, 2008; Samsonovich et al., 2008). Self-regulation involves ideation of, or planning, a desired cognitive place to be (Efklides et al., 2001) and volition to go or stay there. When current cognition is self-Evaluated and found to contrast with ideated cognition, this provides the impetus to self-Regulate cognition in that new direction.
The consolidation aspect of self-regulation is as crucial as the change aspect and involves a process whereby self-evaluation suggests present cognition is ideal or fit-for-purpose, resulting in an active decision to uphold cognition, to stay at present, optimum levels (Zimmerman & Schunk, 2011). Such an intentional stabilization of cognition, for example maintaining cognitive focus (Efklides, 2008; Samsonovich et al., 2008), may be a factor in maintaining immersion in learning, or ‘flow’, by the learner’s intentional removal or reduction in factors that could otherwise distract from the present fluid cognition (Landhäußer & Keller, 2012). Cognitive flow is by nature primarily unaware of all but its focus subject, not typically metacognitive, however, maintaining flow may sometimes require subtle self-monitoring of shifts away from the flow as well as self-regulation back into the flow.
Self-regulation of cognition relies on information from self-Evaluation of cognition in order to activate conscious control of cognition and to adapt cognitive functioning appropriately. Self-Monitoring and self-Evaluating cognition are precursors to, and co-conditions of, self-Regulating because intentional changes in thinking processes cannot be made effectively without knowing what is already happening cognitively. Self-regulation without self-evaluation during a cognitive task is like running in the dark; unaware and potentially leading to a fall.
As noted earlier, planning of cognition is not a separate component in the AMERT framework but is rather subsumed in self-Regulation of cognition. Planning is often characterized as a metacognitive first ‘phase’ but a phase orientation is a focus on the role of metacognition rather than its processes. Planning cognition is complex, requiring co-occurring self-Monitoring and self-Evaluating processes (O'Leary & Sloutsky, 2019) and is related to cognitive transfer, described next.
Self-regulation may also lead to cognitive overload, impairing students in the task at hand, in part because they are allocating finite cognitive resources to think about their thinking (Sweller, 2011). Self-regulation requires self-evaluating to determine when to discontinue metacognition and focus only on the task at hand rather than one’s own thinking about it.
Self-Transfer of cognition
Self-Transfer of cognition, the fifth level of AMERT, is the use of cognitive strategies that are developed in one context and employed intentionally as strategies in a different context (Tuomi-Gröhn & Engeström, 2003; Smith et al., 2007; Heggen, 2008). Transfer of cognitive skills to new contexts is not easy, because knowledge is acquired or developed in a language-and-culture-rich context, as are the metalanguages, or labels, used in metacognition (Cellar & Barrett, 1987). As cognitive skills like ‘analysis’ shift in meaning and nuance from context to context, the shift makes Transfer of cognition complex and for many students it requires facilitation by others.
For Transfer of cognition to occur, an individual must recognize prior cognitive knowledge and skills and how these need to be reconstructed or re-contextualized (Garraway et al., 2011) to meet the needs of the new context, by identifying what cognition is required and reflecting on any similarities to past experiences (Garraway et al., 2011). Self-Transfer of cognition is fully intentional and is an indicator of high-level metacognition in learning environments and situations that are different from where the cognitive skills were learned. The AMERT hierarchy suggests that if self-Transfer of cognition is happening effectively, self-Awareness, self-Monitoring, Self-Evaluating and self-Regulating are taking place simultaneously.
The hierarchy of the AMERT framework
The AMERT model (Table 1) shows that the lower levels, starting from self-Awareness of cognition, are pre-conditions for those above all the way to self-Regulation and its highest form, self-Transfer of cognition. Weak self-Awareness then, such as that due to the use of poor cognitive labels, drastically reduces the effectiveness of higher levels of metacognition. Moreover, students may intentionally change their cognition without self-Monitoring and self-Evaluating it, but that regulation would be uninformed, random, lack a feedback loop, and lack discernment into the current cognitive state and the effects of any changes. Change always shifts, whereas self-Regulation of cognition may involve maintenance of a steady state due to its reliance on self-Evaluation and the lower levels of AMERT.
Table 1: The hierarchical AMERT framework of metacognition
In the AMERT framework, self-Awareness is knowing about different forms of cognition e.g. ‘I know about the thinking involved in analysis’. Self-Monitoring is the comprehension of one’s current cognition: e.g. ‘I am analysing’. Self-Evaluation is to see if cognition is sufficient or should be different e.g. ‘I need to be more critical in my analysis’. Self-Regulation is to intentionally cognitively shift or to actively consolidate, whether the type, quality or intensity of cognition e.g. ‘I am going to critique more from a perspective that challenges the ideas in my current analysis’. Transfer is to intentionally move previous patterns of cognition to a new context or adapt cognition to fit a new context, e.g. ‘I will adapt and apply my analytical critique that I used in science investigation to my mathematics investigation.’ To use an analogy with heart rate to explain the hierarchy, self-Awareness is distinguishing the sound or feeling of one’s own heart and calling it a ‘heartbeat’; self-Monitoring is actively taking one’s pulse; self-Evaluating is comparing or contrasting the rate to where it should be, depending on rest, exercise or stress levels; self-Regulating is actively taking steps to increase, reduce or maintain heart-rate; Transfer is applying this regulating capacity to other contexts, such as a personal trainer might with a client. An ineffective label that would lead to poor self-Awareness, in this analogy, is exampled by the term ‘liverbeat’, which would render all the higher levels misleading.
An example applied to AMERT
Two quotes from a study that suggested metacognition due to RSD use across a program, but did not have available an appropriate metacognitive framework to probe this in detail, will provide an example of the use of AMERT to unpack metacognitive statements:
… you can see all the levels; you can see where you are. You compare yourself to the data. It takes a skill to be honest to yourself; that’s the first skill. When you look at the different levels, you can see where you are fitting and then you look at the levels ahead, at what are your areas of improvement so you can improve yourself (Author).
-
Self-Aware: … you can see all the levels
-
Self-Monitor: … you can see where you are fitting
-
Self-Evaluate: You compare yourself to the data. It takes a skill to be honest to yourself.… then you look at the levels ahead, at what are your areas of improvement.
-
Self-Regulate: … so you can improve yourself
So you can improve yourself has a strong planning sense and the potential for transfer to new contexts.
Self-Transfer is explicitly noted by another student in the same study: ‘… because they have been consistently applying this structure to all of our assignments, we have come to think that way for science (Author). This is a transfer from the range of specific assessment tasks to a scientific way of thinking more globally.
AMERT is an interpretation and synthesis of the findings on metacognition over the last four decades, however, how effectively does this hierarchical framework represent metacognition? Using López-Campos et al’s. (2008) framing, three questions need to be answered before AMERT framework should be used broadly:
1. Does the AMERT framework provide a viable interpretation of accounts where metacognition is evident? (Framework is functional and provides insight.)
2. Does the AMERT framework provide a valid understanding of metacognition? (Framework has construct validity.)
3. Do AMERT-based instruments provide reliable scores of metacognition within and between studies? (Framework informs instruments that demonstrate internal and external reliability.)
In terms of viability, is the framework functional, easy to interpret and readily able to help researchers distinguish different levels? Testing of viability may be done with rich qualitative data, so see if the different levels correspond to actual experiences. Researchers from outside of the team that formulated AMERT are need to also test viability, before further testing of validity is warranted. Testing the construct validity requires using AMERT a priori as the framework to generate data. However, it would be premature to devise a priori empirical tests given the calls that proritise deep qualitative understanding and the need to devise better theorisations of metacognition before further data is generated, as mentioned earlier. Therefore, the general question addressed by this study concerns viability only:
Does the AMERT hierarchy provide a viable understanding of student metacognition that is evidenced in interviews?