Synonyms

Metacognition; Metacognitive monitoring and control; Multi-representational learning environments; Scaffolding hypermedia; Self-regulated learning

Definition

Metacognition, often refered to as “thinking about thinking,” is defined as “one’s knowledge concerning one’s own cognitive processes and products or anything related to them, e.g., the learning-relevant properties of information or data. […] Metacognition refers, among other things, to the active monitoring and consequent regulation and orchestration of these processes in relation to the cognitive objects or data on which they bear, usually in the service of some concrete goal or objective” (Flavell 1976, p. 232).

Contemporary definitions of metacognition characterize it as an individual’s cognition about his or her own cognitions, knowledge of and control over one’s own cognition, cognition that reflects on, monitors or regulates first-order cognition or knowing about knowing (for an author overview, see Opfermann 2008).

Metacognition and self-regulatory processes are especially important in open-ended learning, e.g., in nonlinear, multi-representational hypermedia learning environments. Hypermedia environments are also characterized by a high level of interactivity and network-like information structures (Scheiter and Gerjets 2007) which require learners to use metacognitive skills in order to make decisions that are enduringly required due to the high level of learner control. This chapter focuses on the relationship between metacognition and hypermedia learning and emphasizes the importance of metacognitive skills and self-regulatory processes for learners to benefit from these environments.

Theoretical Background

The importance and popularity of interactive learning environments has grown rapidly in the last few decades. Specifically with regard to hypermedia environments, their network-like structure allows learners to retrieve information flexibly, thus offering a high amount of learner control with regard to representational and navigational choices. A first major impetus for its use lies in the belief that this increased flexibility is associated with increased interest and motivation since learners are involved in decision-making processes (Alexander and Jetton 2003), although research in the cognitive and learning sciences reveals that that this is not always the case since metacognitive and self-regulatory skills are required in order for learners to successfully benefit from hypermedia. Second, hypermedia environments are expected to enhance the opportunity to adapt learning to one’s personal preferences and cognitive needs. Third, hypermedia’s high level of learner control includes affordances for active and constructive information processing; and, finally, hypermedia environments, due to their high level of learner control, are assumed to foster the acquisition of self-regulatory skills in that learners are continuously forced to decide between different information sources (e.g., which hyperlink they should follow next or whether to retrieve pictures or animations in addition to which texts to read) and to evaluate whether the information that they just retrieved helps them to achieve their learning goals (see Astleitner and Leutner 1996, on learning strategies to reach different goals of learning with unstructured hypermedia). Scheiter and Gerjets (2007, p. 288) state that enabling the acquisition of such meta-skills during learning is “one important criterion that learning environments for self-regulated learning may have to meet.”

Despite the panacea of hypermedia learning environments, there are many potential pitfalls that must be considered in terms of the role of metacognition and self-regulation with hypermedia learning. While, on the one hand, these environments may come along with all the benefits proposed before, they only do so once certain issues are addressed. With regard to metacognition, Azevedo (2005) states that hypermedia environments, despite their educational potential, have failed to enhance students’ learning because students often lack key self-regulatory and metacognitive skills and thus struggle with the open and in itself complex nature of hypermedia learning environments. More specifically, learners do not spontaneously deploy monitoring processes like feeling of knowing (FOK; linking the current content found in the hypermedia environment with their prior knowledge) or judgment of learning (JOL; assessing their emerging understanding of the content). They do not always plan their learning by creating relevant subgoals or activating their prior knowledge. Another important issue is that they tend not to use effective strategies such as making inferences, coordinating informational sources, or engage in knowledge elaboration. These activities, however, are seen as central in hypermedia learning. Following his own criticism with regard to existing hypermedia research which, according to Azevedo and Witherspoon (2009), has not yet addressed how exactly a learner regulates his or her learning with hypermedia, the authors introduce a model which is adapted from self-regulated learning research and allows a more direct view on the interplay between learner characteristics, cognitive processes, and system structure during hypermedia learning.

In line with other SRL researchers (Winne and Hadwin 2008), Azevedo (2005) sees self-regulated learning with hypermedia as a constructive process highlighted by several learning phases and cycles of metacognitive monitoring and control. Azevedo also proposes SRL as being a multiphase process where learners need to:

  • Analyze the learning situation

  • Set meaningful learning goals

  • Determine which strategies to use and assess whether these strategies are effective to meet the learning goals

  • Monitor and evaluate their understanding and, if necessary

  • Modify plans, goals, strategies, and effort in relation to contextual conditions (which include cognitive, motivational, and task conditions)

The model of Azevedo and colleagues (Azevedo 2005; Azevedo and Witherspoon 2009) includes over three dozen self-regulatory processes such as:

  • Planning, e.g., setting relevant goals, activating prior knowledge

  • Monitoring, e.g., feeling of knowing, judgment of learning, monitoring progress toward goals

  • Applying learning strategies, e.g., hypothesizing, coordinating information sources, drawing inferences, summarizing

  • Handling task difficulties, e.g., help-seeking behavior

In their model, Azevedo and colleagues do not explicitly label any of these variables as “good” or “bad” aspects of self-regulatory learning with hypermedia; however, they report that successful learners regulated their learning by using effective strategies, planning their learning by creating subgoals, activating prior knowledge, monitoring emerging understanding, and by planning their time and effort. On the other hand, less successful learners tended to use effective as well as ineffective strategies equally often, planned their learning by using subgoals and recycling goals in working memory, and handled task difficulties and demands through engaging in help-seeking behavior. In line with this, several researchers (for an overview, see also Opfermann 2008) have found that learners who possess sophisticated self-regulatory skills are better able to cope with the demands imposed by the complex and multifaceted structure of hypermedia environments. A main difference between successful and unsuccessful learners seems to be that the latter do not seem to deploy the key metacognitive and self-regulatory processes on their own.

Important Scientific Research and Open Questions

Based on the above-mentioned considerations, several attempts have been made to support hypermedia learning by various means of instructional support. For instance, Stadtler (2006) and Opfermann (2008) let students watch a metacognitive modeling video prior to learning with their respective hypermedia environments. In these videos, an exemplary good learner showed how to optimally navigate through an environment in a systematic fashion, how to compare pieces of information from different sources and how to evaluate one’s own learning progress and, if necessary, how to adapt one’s own way of learning. Other researchers in the field of hypermedia learning (e.g., Bannert 2006) or other types of open learning environments (e.g., Thillmann et al. 2009) worked with several kinds of reflection prompts that aimed at scaffolding students’ self-regulated learning process either before or during learning. Finally, authors such as Schmidt and Ford (2003) tried to induce metacognitive activities by presenting direct metacognitive instruction prior to the learning phase (e.g., how important it is for one’s own learning to monitor the own learning progress, to reflect upon what one is doing, etc.).

So far, efforts to provide metacognitive support have produced mixed results. On the one hand, students who make use of these support features obviously outperform those who do not with regard to learning performance and learning transfer. But the ability to use such support effectively, in turn, was primarily found for learners with high prior knowledge or expertise, respectively. In line with this, Schnotz et al. (2005) assume that to benefit from instructional support during computer-based learning, learners should possess certain prerequisites; otherwise such support might lead to cognitive overload which may interfere with one’s ability to self-regulate.

At first sight, it seems somehow counterintuitive that metacognitive support is only useful for learners who already possess knowledge and metacognitive abilities. Isn’t it more logical that learners who lack such abilities receive support to optimize their learning? And if so, how can it be assured that learners with little prior knowledge and little metacognitive abilities benefit from support features such as prompting? According to Schnotz et al. (2005), an important aspect for this group of learners is the optional use of support, i.e., giving students the freedom to decide if and when they retrieve instructional support. In order to enable students with low prior knowledge to benefit from metacognitive support, Azevedo and Witherspoon (2009) and Bannert (2006) emphasize the need of extensive metacognitive training for such learners to help them acquire, practice, retain, and learn to apply self-regulatory processes and therefore become more sophisticated learners who demonstrate gains in conceptual understanding and transfer their SRL skills and knowledge. In line with this, Azevedo and Witherspoon (2009) found that training students how to regulate their learning according to models of self-regulated learning (e.g., planning, monitoring, and strategic proceeding) led to greater shifts in mental models, higher posttest performance, and higher metacognitive activities such as prior knowledge activation, planning, or monitoring progress toward goals.

Taken together, these results (cf. Bannert 2005) seem to indicate that the issue of metacognitive support should be addressed from two perspectives. In particular, extensive, long-term metacognitive training as a form of direct metacognitive support should be distinguished from indirect support such as metacognitive prompts. While the first form of support is deemed adequate and necessary for students lacking metacognitive competence, the second support form might be rather suitable for students already possessing metacognitive skills but not being able to display them spontaneously. Most research investigating the impact of metacognitive support for web-based and hypermedia learning has made use of the latter option – mainly because time restrictions of the short-termed studies do not allow for extensive training, but also because participants in these studies are often university or high school students for whom a certain degree of metacognitive skills is presumed. On the other hand, research in the field of self-regulated learning from expository texts in high schools shows that even short-term trainings of metacognitive skills, aligned with cognitive learning strategies, can be very effective (e.g., Leutner et al. 2007).

In addition, and to conclude this chapter, it may also be assumed that sophisticated self-regulatory skills are necessary but not sufficient for hypermedia learning (Scheiter and Gerjets 2007). More specifically, motivation and interest might also be prerequisites that strongly influence how much effort someone invests in the resource-demanding activation of sophisticated self-regulatory and metacognitive learning strategies. Future research should therefore include motivational variables as well.

Cross-References

Interactive Learning

Interactive Learning Environments

Metacognition and Learning

Multimedia Learning

Self-regulated Learning

Situated Prompts in Authentic Learning Environments