Optimizing cognitive load for learning from computer-based science simulations.

Online courses are becoming ubiquitous and increasingly tend to use authentic learning tasks as the driving force for teaching and learning. Nevertheless, designing online courses that incorporate real– world tasks is more challenging as these problems require more cognitive processes (van Merriënboer Sluijsmans, 2009). This phenomenon can be explained by Cognitive Load Theory (CLT) introduced by Sweller (1994). CLT distinguishes three types of cognitive load: intrinsic, extraneous and germane load. The level of intrinsic load is assumed to be determined by the level of element interactivity. An element can be a definition, concept, formula and procedure that needs to be or has been learned. Extraneous load is mainly imposed by instructional procedures that are suboptimal, whereas germane load refers to the learners’ working memory resources available to deal with the complexity of the task or learning material (Sweller, 2010). Accordingly, the experienced cognitive load is mainly dependent of students’ prior knowledge. Nevertheless, cognitive load can also be determined by students’ motivation (Feldon, Franco, Chao, Peugh, Maahs-Fladung, 2018; Verhoeven, Schnotz, Paas, 2009). As a consequence, when designing an online course for complex tasks, it is important to understand how the different types of cognitive load are affected by students’ cognitive and motivational characteristics. Therefore, in the current study, a high and low complex task was developed relating to the learning and teaching of geometry. The complexity of the task was manipulated by increasing the element interactivity for the high complex task (Sweller, 2010). In the low complex task one element was questioned each time, and consequently students had to apply a single rule, formula or procedure. By contrast, the high complex task was based on a real-life context (e.g., teaching geometry), and had higher element interactivity. Subsequently, the high complex task required learners to engage in a series of cognitive activities such as analysing, decision making, implementing and evaluating, while holding several procedures and rules in mind. Accordingly, we expected the high complex task to induce more cognitive load. The same amount of support containing the same content, was provided during both tasks. Consequently, in this context, students could take initiative in diagnosing their learning needs by identifying appropriate support. Since students could consult different amounts of support, this self-directed learning strategy could also influence the perceived cognitive load (van Merriënboer Sluijsmans, 2009). Accordingly, the amount of consulted support was also taken into account during the analyses. The aim of the study was twofold. First, as a manipulation check of task complexity, we investigated differences in the experienced cognitive load while solving a high and low complex task. Secondly, we examined whether students’ cognitive and motivational characteristics influence the different types of perceived cognitive load, when taking into account the amount of consulted support for both the high and low complex task. A multivariate approach was chosen to assess the degree of interplay that may exist among students’ cognitive, motivational characteristics, consultation of support and the different types of perceived cognitive load. By conducting this study, we wanted to gain insight into whether the cognitive, motivational characteristics and consultation of support influence the perceived cognitive load differently for a high and low complex task.

Providing prompts to induce focused processing of the central contents of instructional explanations is a promising instructional means to support novice learners in learning from instructional explanations. However, within research on the expertise reversal effect it has been shown that instructional means that are beneficial for novices can be detrimental for learners with more expertise if the instructional means provide guidance that overlaps with the internal guidance provided by the prior knowledge of learners with more expertise. Under such circumstances, prompts to induce focused processing might even be detrimental for learners with expertise whose prior knowledge already provides internal guidance to learn from explanations. On this basis, we aimed at experimentally varying expertise by developing prior knowledge. Specifically, we used a preparation intervention with contrasting cases to enhance learners’ prior knowledge (expertise). Against this background, we tested 71 university students in a 2 × 2 factorial experimental design: (a) Factor of expertise. Working with contrasting cases to develop prior knowledge and expertise to provide internal guidance to learn from instructional explanations (with vs. without), (b) Factor of prompts. Prompts to induce focused processing of the explanations (with vs. without). The results showed that prompts to induce focused processing fostered conceptual knowledge for novice learners whereas prompts hindered the acquisition of conceptual knowledge for learners with expertise that was developed by working with contrasting cases beforehand. Moreover, measures of subjective cognitive load and learning processes suggest that the instructional guidance provided by prompts compensated for the low internal guidance of novice learners and overlapped with the internal guidance of learners with expertise.

Concept maps (Cmaps) are graphical organisers with great potential to foster meaningful learning. However, just a few studies explore the use of teacher-prepared Cmaps as chemistry instructional material. In this thesis, chemical explanations for two macroscopic phenomena (the colour in fireworks and electric conductivity) are chosen to be represented and taught through Cmaps. The possibility of reading the Cmap in different ways, the need for establishing a reading sequence and the content difficulty can lead, for the learner who is interacting with the material, a sense of disorientation. Disorientation impairs the cognitive resources that should be used for learning (generative processes) due to either a high content complexity (i.e., intrinsic load) or an unsuited format of instruction (i.e., extraneous load). Based on the assumptions of Cognitive Load Theory and Multimedia Learning Theory, these materials should be elaborated in a way that the cognitive load imposed to the learners does not exceed the limited working memory capacity, otherwise, there are no left cognitive resources to process and acquire new knowledge (i.e., learn). The thesis' goal was to evaluate the impact of some strategies in minimizing the disorientation of Cmap that organizes chemistry contents and, consequently, to foster scientific conceptual learning. Three studies were conducted to evaluate the impact of: using navigational graphical cues (Study I), different types of conceptual organisation (Study II), using digital resources linked to the Cmap (i.e., hypertext) and the level of learners' prior knowledge (Study III) on the gain of factual and conceptual knowledge and the cognitive load (e.g., mental effort) related to understanding the chemical content. In total, 253 freshman university students to the School of Arts, Science and Humanities of the University of So Paulo participated in this research between 2014 and 2017. A quantitative methodology (by a pretest-to-post-test quasi-experimental design) were adopted for statistically comparing the mean of performance and mental effort from students in different experimental conditions. Considering the results, it was concluded that: (1) regardless the type of Cmap used as instructional material, all students were capable to improve their level of factual knowledge about chemical concepts and (2) the use of colour and conceptual hierarchy as navigational graphical cues decreased Cmap disorientation by managing intrinsic load and reducing extraneous load, respectively. Generative processing was fostered using digital resources linked to the Cmap by offering a scaffold for hypertext navigation. Some educational implications, research limitations and future studies are also presented.

Prior knowledge activation facilitates learning. Note taking during prior knowledge activation (i.e., note taking directed at retrieving information from memory) might facilitate the activation process by enabling learners to build an external representation of their prior knowledge. However, taking notes might be less effective in supporting prior knowledge activation if available prior knowledge is limited. This study investigates the effects of the retrieval-directed function of note taking depending on learners' level of prior knowledge. It is hypothesized that the effectiveness of note taking is influenced by the amount of prior knowledge learners already possess. Sixty-one high school students participated in this study. A prior knowledge test was used to ascertain differences in level of prior knowledge and assign participants to a low or a high prior knowledge group. A 2×2 factorial design was used to investigate the effects of note taking during prior knowledge activation (yes, no) depending on learners' level of prior knowledge (low, high) on mental effort, performance, and mental efficiency. Note taking during prior knowledge activation lowered mental effort and increased mental efficiency for high prior knowledge learners. For low prior knowledge learners, note taking had the opposite effect on mental effort and mental efficiency. The effects of the retrieval-directed function of note taking are influenced by learners' level of prior knowledge. Learners with high prior knowledge benefit from taking notes while activating prior knowledge, whereas note taking has no beneficial effects for learners with limited prior knowledge.

Research on learning engagement and cognitive load theory have proceeded in parallel with little cross-over of ideas. The aim of this research was to test an integrative model that examines how prior knowledge influences learning engagement via cognitive load and help-seeking strategies. A sample of 356 students from two middle schools in the north of China participated in the study. Analyses using structural equation modeling revealed that prior knowledge was positively associated with learning engagement, and that this relationship was mediated by cognitive load and instrumental help-seeking. Cognitive load also mediated the impact of prior knowledge on instrumental help-seeking, executive help-seeking and avoidance of help-seeking. The study shows that students with more prior knowledge and lower cognitive load are able to exercise higher levels of instrumental help-seeking, leading to good quality learning engagement. On the other hand, students with less prior knowledge and higher cognitive load are less able to engage in instrumental help-seeking, leading to lower engagement. Based on the research findings, recommendations for how teachers can improve learning engagement through decreasing cognitive load are discussed.

Expertise reversal effect: Cost of generating new schemas

Background:Advancing patient safety during handoffs remains a public health priority. The application of cognitive load theory offers promise, but is currently limited by the inability to measure cognitive load types.Objective:To develop and collect validity evidence for a revised self-report inventory that measures cognitive load types during a handoff.Methods:Based on prior published work, input from experts in cognitive load theory and handoffs, and a think-aloud exercise with residents, a revised Cognitive Load Inventory for Handoffs was developed. The Cognitive Load Inventory for Handoffs has items for intrinsic, extraneous, and germane load. Students who were second- and sixth-year students recruited from a Dutch medical school participated in four simulated handoffs (two simple and two complex cases). At the end of each handoff, study participants completed the Cognitive Load Inventory for Handoffs, Paas’ Cognitive Load Scale, and one global rating item for intrinsic load, extraneous load, and germane load, respectively. Factor and correlational analyses were performed to collect evidence for validity.Results:Confirmatory factor analysis yielded a single factor that combined intrinsic and germane loads. The extraneous load items performed poorly and were removed from the model. The score from the combined intrinsic and germane load items associated, as predicted by cognitive load theory, with a commonly used measure of overall cognitive load (Pearson’s r = 0.83, p < 0.001), case complexity (beta = 0.74, p < 0.001), level of experience (beta = −0.96, p < 0.001), and handoff accuracy (r = −0.34, p < 0.001).Conclusion:These results offer encouragement that intrinsic load during handoffs may be measured via a self-report measure. Additional work is required to develop an adequate measure of extraneous load.

Introduction: Patient handoffs remain a significant patient safety challenge. Researchers have used cognitive load theory (CLT) to unpack the complexity of health professions workplace learning tasks including handoffs. 1 When the cognitive load of a handoff exceeds the working memory capacity of the trainee, learning and performance suffer. CLT focuses on the development of instructional strategies that match intrinsic load (IL) to the developmental stage of the learner, decrease extraneous load (EL), and optimize germane load (GL). CLT strategies have traditionally focused on modifying the complexity of the task, the knowledge of the learner, and/or the design of the instruction. CLT researchers have given little attention to how the learner’s emotions influence cognitive load. 2 Yet, we know that emotion modulates learning through numerous cognitive processes including motivation, attention, working memory, and long-term memory. 3 To address this gap, this study explores how emotion influenced the IL, EL, and GL imposed upon the working memory of trainees performing patient handoffs. Methods: From January to March 2019, we administered a cross-sectional survey to 1,807 residents and fellows from a large 24-hospital health system in the United States. Participants completed the survey after performing a handoff. The survey included questions about features of the learner, task, and instructional environment. IL, EL, and GL were measured with the 16-item Cognitive Load Inventory for Handoffs. 4 Emotions were assessed with an 8-item scale based on the circumplex model of affect. 5 The authors used factor analysis to identify the core dimensions of emotion and then explored the relationship between emotion and cognitive load with 3 types of analyses. First, bivariate correlational analyses examined the relationship between the emotion factors and learner, task, and environment features known to predict cognitive load. Second, we conducted univariate regression analyses to investigate whether emotion factor scores predicted IL, EL, and GL. Third, multivariate regression determined whether the addition of the emotion factors to models including these previously established predictors of cognitive load types increased the variance explained (Δ R2). Results: Six hundred and ninety-three (38.7%) of 1,807 residents and fellows completed a survey. Exploratory and confirmatory factor analysis identified 2 dimensions of emotion representing positive activation (labeled invigoration) and positive deactivation (labeled tranquility). The correlational analyses revealed that higher invigoration was associated with higher levels of training and use of a standardized verbal protocol, while lower levels of invigoration were associated with interruptions, difficulty obtaining essential information, and fatigue. Higher tranquility was associated with higher levels of training and lower levels of tranquility with interruptions, difficulty obtaining information, fatigue, longer handoffs, and greater number of patients. In univariate regression analyses, higher levels of invigoration, tranquility, and their interaction were independently associated with lower intrinsic load and extraneous load. The interaction of invigoration and tranquility predicted lower germane load. Finally, in the multivariate models of predictors for each type of cognitive load, the addition of the emotion term increased the amount of variance explained from 0.10 to 0.16 for IL, 0.15 to 0.27 for EL, and 0.11 to 0.14 for GL. Discussion: Emotion significantly influenced the cognitive load experienced by trainees while performing in vivo handoffs. Like studies in the simulation context, the structure of emotion during patient handoffs consisted of 2 factors. Higher invigoration/tranquility predicted lower IL, El, and GL. Traditionally, CLT researchers assumed that emotion primarily affected EL. Our results suggest a much broader impact of emotion. Significance: The study models how emotion and cognitive load can be measured during a workplace learning activity. These results suggest that enhancing invigoration and tranquility will reduce cognitive load and improve learning and performance. This highlights the importance of developing instructional strategies that help trainees regulate their emotions.

Research has shown that interest in knowledge facilitates students' academic achievement in learning. Because individual interest is often based on how much one knows, in other words existing or prior knowledge, studying adolescents' interest in health-enhancing physical activity and its benefits should address the relation between the interest and their existing or prior physical activity knowledge. Understanding this relationship may help us facilitate students to not only develop interest in knowing more about but also actual adopt a healthy, active lifestyle. This study used a large-sample structural equation design to identify the relationship between middle school students' interest in physical activity knowledge and their prior physical activity knowledge, and to assess the change of this relationship over time. Guided by the declarative-procedural knowledge framework, latent growth models were developed and tested on data collected from a random sample of 3882 students from ten middle schools. The latent growth curve model indicated that, 1) on average, students experienced a significant interest decline in both procedural and declarative knowledge; 2) prior knowledge helped slow the decline and facilitated interest growth in knowledgeable students. The results suggest that existing knowledge determined the interest change.

IntroductionClinical reasoning has been fostered with varying case formats including the use of virtual patients. Existing literature points to different conclusions regarding which format is most beneficial for learners with diverse levels of prior knowledge. We designed our study to better understand which case format affects clinical reasoning outcomes and cognitive load, dependent on medical students’ prior knowledge.MethodsOverall, 142 medical students (3 rd to 6 th year) were randomly assigned to either a whole case or serial cue case format. Participants worked on eight virtual patients in their respective case format. Outcomes included diagnostic accuracy, knowledge, and cognitive load.ResultsWe found no effect of case format on strategic knowledge scores pre- vs post-test (whole case learning gain = 3, 95% CI. -.01 to .01, serial cue learning gain = 3, 95% CI. -.06 to .00 p = .50). In both case formats, students with high baseline knowledge (determined by median split on the pre-test in conceptual knowledge) benefitted from learning with virtual patients (learning gain in strategic knowledge = 5, 95% CI .03 to .09, p = .01) while students with low prior knowledge did not (learning gain = 0, 95%CI −.02 to .02). We found no difference in diagnostic accuracy between experimental conditions (difference = .44, 95% CI −.96 to .08, p = .22), but diagnostic accuracy was higher for students with high prior knowledge compared to those with low prior knowledge (difference = .8, 95% CI 0.31 to 1.35, p < .01). Students with low prior knowledge experienced higher extraneous cognitive load than students with high prior knowledge (multiple measurements, p < .01).ConclusionsThe whole case and serial cue case formats alone did not affect students’ knowledge gain or diagnostic accuracy. Students with lower knowledge experienced increased cognitive load and appear to have learned less from their interaction with virtual patients. Cognitive load should be taken into account when attempting to help students learn clinical reasoning with virtual patients, especially for students with lower knowledge.

Transaction analysis, the process of discerning which accounts are impacted (either increased or decreased) by a given transaction, is a fundamental accounting principle. While proficiency in this concept is vital for introductory accounting students, it poses a significant challenge for many. We addressed this hurdle using Excel and Excel VBA to develop a fully automated Journal Entry Learning Object (JELO) spreadsheet designed to aid first-year accounting students in grasping the intricacies of analyzing business transactions. Grounded in cognitive load theory (CLT), our spreadsheet based JELO provides a user-friendly interface that helps students develop their skills without becoming overwhelmed. Cognitive load theory, pioneered by John Sweller, suggests that learning is optimized when instructional materials reduce unnecessary cognitive load, allowing students to allocate their cognitive resources efficiently toward understanding and mastering the task at hand. CLT distinguishes between three types of cognitive load: intrinsic, extraneous, and germane (Sweller, Ayres, &amp; Kalyuga, 2011). Intrinsic cognitive load refers to the inherent difficulty associated with a specific instructional topic (Sweller, J. 2010). In transaction analysis, the intrinsic load is naturally high due to the complexity of understanding the impact of transactions on various accounts. To manage this, JELO has three tiers of transaction questions. Students are allowed to progress to more complex transactions only after demonstrating proficiency with the lower levels. Extraneous cognitive load is generated by inefficient presentation of information to learners (Paas &amp; Sweller, 2014). JELO minimizes extraneous load through its intuitive and interactive design. By providing clear instructions, a logical interface, and eliminating unnecessary information, JELO ensures that students can focus on the essential aspects of transaction analysis without being distracted by irrelevant information. Germane cognitive load refers to the mental effort required to create schemas and integrate new information into existing knowledge structures (Sweller et al., 2011). JELO enhances germane load by offering scaffolded and immediate feedback. This feedback is designed to reinforce correct responses and guide students through their errors with hints and explanations, facilitating deeper cognitive processing and better schema construction. Our goal was to make skill acquisition enjoyable and, more realistically, to reduce the pain of learning transaction analysis. The JELO offers an alternative to traditional “drill and practice” methods by providing an easy-to-use, interactive learning tool that dynamically adjusts the difficulty of its questions to match the student's skill level. This adaptive learning approach ensures that students are continually challenged just enough to promote learning without becoming overwhelmed. Additionally, JELO's feedback mechanism aligns with CLT principles by providing immediate positive reinforcement and corrective guidance. By doing so, it helps students build confidence and competence in transaction analysis, leading to a more effective and engaging learning experience. By incorporating principles of cognitive load theory, JELO ensures that students are neither under-challenged nor overwhelmed. This careful balancing of students’ overall cognitive load facilitates a more effective learning environment, fostering both proficiency and satisfaction in mastering transaction analysis.

Background: The use of virtual patients (VPs), due to their high complexity and/or inappropriate sequencing with other instructional methods, might cause a high cognitive load, which hampers learning.Aim: To investigate the efficiency of instructional methods that involved three different applications of VPs combined with lectures.Method: From two consecutive batches, 171 out of 183 students have participated in lecture and VPs sessions. One group received a lecture session followed by a collaborative VPs learning activity (collaborative deductive). The other two groups received a lecture session and an independent VP learning activity, which either followed the lecture session (independent deductive) or preceded it (independent inductive). All groups were administrated written knowledge acquisition and retention tests as well as transfer tests using two new VPs. All participants completed a cognitive load questionnaire, which measured intrinsic, extraneous and germane load. Mixed effect analysis of cognitive load and efficiency using the R statistical program was performed.Results: The highest intrinsic and extraneous load was found in the independent inductive group, while the lowest intrinsic and extraneous load was seen in the collaborative deductive group. Furthermore, comparisons showed a significantly higher efficiency, that is, higher performance in combination with lower cognitive load, for the collaborative deductive group than for the other two groups.Conclusion: Collaborative use of VPs after a lecture is the most efficient instructional method, of those tested, as it leads to better learning and transfer combined with lower cognitive load, when compared with independent use of VPs, either before or after the lecture.

The worked example and expertise reversal effect in less structured tasks: Learning to reason about legal cases

The purpose of this study is to investigate the effects of signaling and prior knowledge on the cognitive loads, motivations, and learning of college students in an immersive virtual reality (IVR) environment. This study applied a 2 (signaling vs. no signaling) by 2 (high vs. low prior knowledge levels) between-subjects factorial design. The results revealed that signaling directed the attention of students with low prior knowledge levels, effectively helped them select relevant information and reduced their cognitive loads, whereas signaling had no significant effect on the cognitive loads, intrinsic motivation, and learning performance of learners with high levels of prior knowledge. These results suggest that IVR environments for students with low prior knowledge levels should reduce cognitive load and improve learning, and signals in the form of text annotations and color changes are recommended for additional support. Students with high prior knowledge levels do not require additional signals to support learning; therefore, the IVR environment needs to be designed in such a way as to be tailored to the individual differences of students.

PD41-11 A DEEPER LOOK INTO SURGEON USE OF 3D VIRTUAL IMAGING PLATFORM (IRISTM): ANALYSIS OF GAZE PATTERNS AND PUPIL METRICS USING HEAD MOUNTED EYE TRACKERS (HMET)