Desktop version

Home arrow Philosophy

  • Increase font
  • Decrease font


<<   CONTENTS   >>

The complexity of monitoring learning processes

One possible explanation of why monitoring one’s own learning seems to be difficult and prone to overestimation, is that it takes place at the same time as learning, or directly after learning. Moreover, learning tasks are often complex for students who are novices in a domain, leaving little room for monitoring and regulation processes. According to cognitive load theory (CLT; Sweller, Van Merrienboer, & Paas, 1998, 2019) it can be assumed that the competition for working memory (WM) resources between learning processes and self-regulation processes can have negative effects on either or both of these processes. Understanding the interplay between learning, monitoring, and the role of cognitive load is needed to provide insight into possible strategies to improve SRL processes when learning complex tasks.

According to CLT (Sweller, 2010; Sweller et al., 1998, 2019), complexity of learning tasks can be partially explained by the number of interacting information elements in a task. The higher the number of interacting information elements, the more complex a learning task is. Especially learning more complex materials can place a high demand on limited cognitive resources (Baddeley, 1986; Cowan, 2001). In addition, the expertise of the learner also plays a role in how complex a task is perceived by a learner. That is, with more expertise, information elements can be combined into schemata in long-term memory, and processed as one element in WM, lowering the number of interacting information elements. Therefore, the cognitive load a task imposes will be lower for learners with more expertise than for learners with less expertise (Kalyuga, 2007; Kalyuga & Sweller, 2004). Generally, it can be argued that monitoring one’s own learning in education, where typically new, complex tasks have to be learned, is difficult for learners. Moreover, as SRL involves monitoring the object-level and thereby informing the meta-level to control the learning process at the object-level (Nelson & Narens, 1990), SRL presumably causes high element interactivity in and of itself.

Monitoring one’s own learning can be seen as a secondary task next to the learning task itself (Griffin et al., 2008; Van Gog, Kester, & Paas, 2011). When tasks are complex and cognitive load is high, it can be hard to perform well on both the learning task and the monitoring task at the same time, because a learner will have to divide cognitive resources between the two tasks (Briinken, Plass, & Leutner, 2003). Due to

WM limitations, performance on one or both of the tasks may suffer when complexity is high and exceeds the learners processing capacity. Furthermore, the ability to cope with this dual task is dependent on the cognitive resources of the learner. A study by Griffin et al. (2008) showed that reading abilities and working memory capacity (WMC) affected monitoring accuracy. In two experiments, college students read explanatory texts, made monitoring judgments about their comprehension, and took a comprehension test about the texts they read. In the first experiment it was found that re-reading the text improved monitoring accuracy for low-ability readers, but not for high-ability readers. In the second experiment this was confirmed and results further showed that lower-WMC readers benefitted from re-reading in terms of monitoring accuracy whereas high-WMC readers did not. Griffin et al. (2008) concluded that contextual factors such as re-reading and individual differences such as reading abilities are possibly related to the ability of monitoring meta-level cues while reading. They pointed out that monitoring is a secondary process next to the primary task of understanding the text itself. Moreover, monitoring accuracy was assumed to be dependent on the cognitive resources of the reader.

A study by Van Gog et al. (2011) also confirmed the idea that concurrent monitoring can be seen as an additional task demanding resources. In their study, secondary school students had to solve Sudoku problems and rate their mental effort as a measure of cognitive load (see Paas, 1992). There were two conditions: a condition in which students had to keep track of what they were doing (i.e., monitoring) and a condition in which they did not monitor their performance (Van Gog et al., 2011). Using a within-subjects design, the effect of the complexity of the Sudoku problems was investigated. Results showed that the instruction to monitor led to higher cognitive load for the complex problems but not for the simple problems. Also, performance and efficiency of performance (see Paas & Van Merrienboer, 1993) on the complex problems were lower for students in the monitoring condition. Hence, the instruction to monitor performance when solving complex problems increased cognitive load and decreased performance and efficiency (Van Gog et al., 2011).

In sum, SRL, being the combination of monitoring and performing a learning task (e.g., Winne & Hadwin, 1998; Zimmerman, 2008), presumably imposes high cognitive load. Monitoring can take place at the same time as learning or directly after a learning task. In both scenarios, the additional task of monitoring demands cognitive resources. Hence, in the case of complex learning tasks, there might be too few resources to accurately monitor and regulate the learning process (Griffin et al., 2008; Van Gog et al., 2011). This could explain why monitoring judgments have been found to be accurate for relatively simple learning materials (e.g., Rhodes & Tauber, 2011) and inaccurate for relatively complex learning materials like expository texts (e.g., Thiede et al., 2009) and problem-solving tasks (e.g., Baars et al., 2018). Yet students are expected to monitor and regulate their own learning to a gradually increasing extent while tasks are getting more complex in (higher) education, especially when learning takes place in digital learning environments in which students operate independently (e.g., Wong et al., 2019). Therefore, it is important to consider strategies to decrease the load of monitoring during learning. One possibility is the use of collaborative learning, which is becoming increasingly popular in many educational settings (Johnson & Johnson, 2009). Collaborative learning could potentially be used as a strategy to reduce the demands on individual cognitive resources when monitoring learning because collaboration creates the opportunity to divide the load between the learners in the group.

 
<<   CONTENTS   >>

Related topics