Coherent Mental Model Research Articles

Enhancing knowledge construction in emerging technologies: the role of imagination training in immersive virtual reality environments

In the rapidly evolving landscape of educational technologies, Immersive Virtual Reality (iVR) stands out as a transformative tool for teaching and learning. The learner must actively engage, particularly when visual and auditory knowledge information is presented simultaneously. Generative learning strategies support learners in processing and integrating the information provided effectively and help them select, organize, and integrate information into a coherent mental model. Yet, learners often do not use such strategies spontaneously, so pre-training of the learning strategy is recommendable. IVR provides a spatial context for learning, where an essential strategy for promoting learning is to integrate auditory knowledge information into visually perceived space. We hypothesize that this can be done effectively through imagination, i.e., learners mentally visualize auditory information. This study investigated an imaginative pre-training strategy that targets integrating visual-spatial and auditory knowledge information. In a pre-post between-subjects design, we tested the effects of imaginative strategy pre-training. Students received either a pre-training (EG = 30) or no training (CG = 30) before exploring an iVR learning environment. Against our expectations, the EG did not outperform the CG on semantic and spatial knowledge. These findings may be explained by the demands on learners to process sensory input, and focus directed attention in a spatial iVR environment, suggesting that imagination strategies need to be anchored and complemented by other selection and organization strategies. This study contributes to a critical understanding of the possibilities and limitations of iVR in education, shedding light on the learners' processes and needs.

Document language moderates the text-belief consistency effect

ABSTRACT When researching socioscientific topics, particularly on the Internet, readers face multiple texts that they must integrate into a coherent mental model. Previous research in monolingual settings has found that comprehension is biased toward readers’ prior beliefs (text-belief consistency effect). Considering that the Internet is multilingual, this experiment focused on the influence of document language on the text-belief consistency effect. Assuming differences in epistemic prestige among languages, we hypothesized that document language would act as a credibility cue and moderate the text-belief consistency effect. Seventy-four German university students read two conflicting texts either in their L2 English (higher epistemic prestige) or in their L1 German (lower epistemic prestige). Participants constructed stronger situation models for the belief-consistent text than for the belief-inconsistent text only when they read the texts in German. These results indicate that document language can act as a source characteristic that may reduce belief biases in comprehension.

Coherence building while reading multiple complementary documents

Readers build a mental representation of text during reading. The coherence building processes readers use to build a mental representation during reading is key to comprehension. We examined the effects of self-explanation on coherence building processes as undergraduates (n = 51) read five complementary texts about natural selection and completed a post-reading measure of topic knowledge. Participants generated constructed responses (verbal protocols) while reading. We varied the use of constructed response prompt (self-explain vs. think-aloud) and constructed response format (typed vs. oral) to examine their impact on the quality of readers’ constructed responses and their coherence building processes (i.e., cohesion). Participants who received self-explanation instructions had higher quality constructed responses than participants who received think-aloud instructions, regardless of whether responses were typed or spoken aloud. Natural Language Processing (NLP) analyses indicated that participants who were prompted to self-explain generated more cohesive responses than those who were prompted to think-aloud. Participants who received self-explanation instructions had more coherent mental models during reading and typing was related to more cohesive responses when participants were asked to self-explain. Participants whose constructed responses were more lexically cohesive during reading had better performance on the post-reading test.

A Review of Seductive Details in Educational Activities Using Electronic Slide Presentations: Derailing the Learning Process.

This review article examines the use of seductive details in nursing education activities. Seductive details are irrelevant images that are placed into educational material. Although seductive details may be found in any type of educational material and at any academic level, this article focuses on the placement of seductive details in electronic slide presentations used in nursing education activities. Educators optimistically place seductive details in educational material to engage learners or provide a momentary break from the presentation. Although engaging the learner is important, educators must use methods that do not derail how the learner's brain remembers information. Added visuals should be relevant to the material needed to be understood. Using seductive details may adversely affect learning by overloading the working memory with extraneous cognitive information, affecting the ability of the learner to create a coherent mental model needed to store a long-term memory. Evidence-based information about seductive detail use is nonexistent in nursing education literature. Professional nurse educators must carefully consider the use of seductive details in electronic slide presentations. [J Contin Educ Nurs. 2023;54(6):261-267.].

Does active or passive signaling support integration of text and graphs?

AbstractIntegrating multiple representations into a coherent mental model is one of the challenges when learning with multimedia. In this experimental study (N = 173), we examined how highlighting corresponding information in text‐graph learning material can help higher education students to make the necessary connections and improve learning outcomes in two domains. We compared a control condition to a signaling condition with given highlights and an active signaling condition where students were asked to visually highlight corresponding text and graph information themselves. There was no overall benefit of given signals or active signaling. We discuss prior knowledge and the quality of learner‐generated signals as possible explanations. For biology learning material, prior knowledge moderated the effect of active signaling. In economics, the effect of prior knowledge was mediated by the quality of learner‐generated correspondences. Our findings suggest that different methods of supporting text‐graph integration might be useful for different students and learning material.

From “no clear winner” to an effective Explainable Artificial Intelligence process: An empirical journey

Abstract“In what circumstances would you want this AI to make decisions on your behalf?” We have been investigating how to enable a user of an Artificial Intelligence‐powered system to answer questions like this through a series of empirical studies, a group of which we summarize here. We began the series by (a) comparing four explanation configurations of saliency explanations and/or reward explanations. From this study we learned that, although some configurations had significant strengths, no one configuration was a clear “winner.” This result led us to hypothesize that one reason for the low success rates Explainable AI (XAI) research has in enabling users to create a coherent mental model is that the AI itself does not have a coherent model. This hypothesis led us to (b) build a model‐based agent, to compare explaining it with explaining a model‐free agent. Our results were encouraging, but we then realized that participants' cognitive energy was being sapped by having to create not only a mental model, but also a process by which to create that mental model. This realization led us to (c) create such a process (which we term After‐Action Review for AI or “AAR/AI”) for them, integrate it into the explanation environment, and compare participants' success with AAR/AI scaffolding vs without it. Our AAR/AI studies' results showed that AAR/AI participants were more effective assessing the AI than non‐AAR/AI participants, with significantly better precision and significantly better recall at finding the AI's reasoning flaws.

Immersive virtual reality or auditory text first? Effects of adequate sequencing and prompting on learning outcome

AbstractPresenting a pictorial representation followed by a textual representation supports learners to build a coherent mental model. Providing an elaboration prompt stimulates learners to process the learning content semantically. Hence, combining both approaches might result in synergetic effects as both foster mental model development, which could be particularly challenging in immersive virtual environments. In our 2 × 2 between‐subject design study, we investigated the beneficial effects of adequate sequencing of text and picture and prompting, as well as their potential synergic effect in an immersive virtual reality (VR) learning environment with N = 81 learners. We presented the learning content in form of one pictorial and one textual representation and varied their presentation sequence: pictorial followed by textual representation or textual followed by pictorial representation. As the pictorial representation, we used a VR animation displaying a service robot. As the textual representation, we chose an auditory text about robotics. Additionally, we varied whether or not learners received an elaboration prompt before the learning unit started. We measured learning outcome differentiated for knowledge, comprehension, and application levels. We found beneficial effects for VR animation followed by auditory text on the knowledge level, only. Prompting led to a beneficial effect only on the application level. We found no synergic effect of adequate sequencing and prompting on learning outcome. Hence, depending on the level of learning outcome targeted, different support approaches are advisable. Future research might consider the long‐term effects of adequate sequencing and prompting or different approaches to measuring mental models. Practitioner notesWhat is already known about this topic? Learning in Virtual Reality (VR) is a promising approach to present the learning material but does not per se guarantee deep learning and the development of a correct mental model. Prior findings imply that presenting pictorial representations first fosters mental model development. Beneficial effects of stimulating learners by an elaboration prompt before learning in virtual reality have been reported. Previous studies imply that instructional design and activation of learners might have synergetic, beneficial effects on learning outcome. What this paper adds We investigated the effects of sequencing order (VR animation or auditory text first) and prompting (with or without) on different levels of learning outcome. We found beneficial effects on the knowledge level when VR animation was presented first, while prompting fostered the application level. No synergetic effects of VR animation first and prompting were found as both support approaches seem to impact different levels of learning outcome. Implications for practice and/or policy Depending on the level of learning outcome that should be fostered while learning in VR, different support approaches are advisable. Presenting a VR animation first fosters learning of basic concepts and definitions at the knowledge level. Using an elaboration prompt before learning in VR enables the learners to process the content deeply. Presenting VR animations first can serve as a brief scaffold for the mental models and additional details can be added in the further learning process.

Learning in Virtual Reality: Bridging the Motivation Gap by Adding Annotations.

One challenge while learning scientific concepts is to select relevant information and to integrate different representations of the learning content into one coherent mental model. Virtual reality learning environments (VRLEs) offer new possibilities to support learners and foster learning processes. Whether learning in VR is successful, however, depends to a large extent on the design of the VRLE and the learners themselves. Hence, adding supportive elements in VRLEs, such as annotations, might facilitate the learning process by guiding attention and supporting the selection of relevant information. Additionally, the mapping of pictorial and verbal information is eased by these annotations. The beneficial effect of annotations is highly dependent on learners' intrinsic motivation as intrinsic motivation while learning also affects the information selection and visual search patterns. In our experimental study (N = 61), we compared two conditions: learning in a VRLE with or without annotations. We measured the learning outcome on three different levels (knowledge, comprehension, and application). Additionally, we investigated intrinsic motivation as a moderator for the effect of annotations on learning outcome. We found no significant main effect of annotations on learning outcome. The moderating effect of intrinsic motivation for annotations on the overall learning outcome was significant. Our results imply that learners are either intrinsically motivated or need additional support by annotations as these support the selection of relevant information in the VRLE and therefore enable them to learn successfully. Which type or quantity of annotations supports learning processes best needs to be explored in future research.

The role of vocabulary knowledge on inference generation: A meta-analysis

Abstract Successful text comprehension results in a coherent mental model of the situation being described. To achieve this, the reader has to infer certain information by connecting parts of the text to their prior knowledge. An important construct involved in this process is vocabulary knowledge, usually divided into breadth and depth. We conducted a meta-analysis on 23 studies, and explored the fit of five different models to establish an effect size of both dimensions of vocabulary on inference making, as well as its developmental trajectory in children aged 3-12. We found a significant and moderate effect of vocabulary knowledge of both modalities. Vocabulary type was not a significant moderator, but age was, meaning that there was a similar effect for both breadth and depth and that the strength of the correlations decreased with age. Heterogeneity was high overall, meaning that more moderators should be assessed in future studies.

Intelligence Comes from Within—Personality as a UI Paradigm for Smart Spaces

The buzzword “smart home” promises an intelligent, helpful environment in which technology makes life easier, simpler or safer for its inhabitants. On a technical level, this is currently achieved by many networked devices interacting with each other, working on shared protocols and standards. From a user experience (UX) perspective, however, the interaction with such a collection of devices has become so complex that it currently rather stands in the way of widespread adoption and use. So far, it does not seem likely that a common user interface (UI) concept will emerge as a quasi-standard, as the desktop interface did for graphical UIs. Therefore, our research follows a different approach. Instead of many singular intelligent devices, we envision a UI concept for smart environments that integrates diverse pieces of technology in a coherent mental model of an embodied “room intelligence” (RI). RI will combine smart machinery, mobile robotic arms and mundane physical objects, thereby blurring the line between the physical and the digital world. The present paper describes our vision and emerging research questions and presents the initial steps of technical realization.

Inference making in young children: The concurrent and longitudinal contributions of verbal working memory and vocabulary.

Inference making is fundamental to the construction of a coherent mental model of a text. We examined how vocabulary and verbal working memory relate to inference development concurrently and longitudinally in 4- to 9-year-olds. Four hundred and twenty pre-kindergartners completed oral assessments of inference making, vocabulary breadth, vocabulary depth, and verbal working memory each year until grade 3. Concurrently, hierarchical regressions revealed that a greater proportion of total variance in inference making was explained by vocabulary and verbal working memory for younger than older children. Vocabulary breadth was a stronger predictor of inference than verbal working memory but the opposite pattern was found for vocabulary depth and verbal working memory. The longitudinal relations between inference making, vocabulary and verbal working memory were investigated in two separate cross-lagged models: one with vocabulary breadth and a second with vocabulary depth. Both vocabulary breadth and depth explained subsequent inference making and verbal working memory throughout the early grades. Inference making also predicted subsequent vocabulary depth. The results highlight the critical role of vocabulary knowledge in the development of inference ability both within and across time, the importance of vocabulary in supporting the development of verbal working memory, and the changing dynamics between language and memory in early development.

Interactions between reasoning about complex systems and conceptual understanding in learning chemistry

Abstract“Complex systems” is a general‐purpose reasoning scheme, used in a wide range of disciplines to make sense of systems with many similar entities. In this paper, we examine the generality of this approach in learning chemistry. Students' reasoning in chemistry in terms of emergent complex systems is explored for two curricula: a normative and a complexity‐based one, so that the interaction could be examined under both the conditions. A quasi‐experimental pretest‐intervention‐posttest comparison group design was used to explore student's learning, complemented with interview data. The experimental group (n = 47) studied the topic of gases with a complexity‐based curriculum. A comparison group (n = 45) studied with a normative curriculum for the same duration. Students' answers to questionnaires were coded with a complexity‐based approach that included levels (distinguishing micro‐ and macro‐levels), stochastic particle behaviors, the emergence of macro‐level patterns from micro‐level behaviors, and the source of control in the system. It was found that students' reasoning about chemistry concepts in terms of complex systems falls into three distinct and coherent mental models. A sophisticated mental model included most of the above‐described complexity features, while the nonsophisticated model included none. The intermediate model is typified by distinguishing between levels, but not by stochastic and emergent behaviors. The nonsophisticated mental model was used mostly in the pretest. In the posttest, the experimental group used the intermediate and sophisticated models; while the comparison group used the nonsophisticated and intermediate models. Discussion approaches the topics of the generality of the complex systems approach; and the unique forms of reasoning that a complexity approach may contribute to learning science.

Examples and the facilitation of student learning: should instructors provide examples or should students generate their own?

ABSTRACTThis study examined learning differences for students who were given instructor-provided examples during a lesson compared with student-generated examples. In an experiment, 348 students were exposed to an online lesson about fear appeals and were randomly assigned to either a condition where (a) examples of key concepts were provided by the instructor, (b) students were asked to provide their own hypothetical examples of key concepts, or (c) students were asked to provide their own personal examples of key concepts. Data were analyzed to examine differences in students’ test scores while controlling for the impact of their familiarity with the material and grade point averages. Results revealed that, compared with students exposed to instructor-provided examples, students who generated their own scored lower on a test of recall and application. This outcome was modeled using path analysis specifying a moderated serial mediation effect. Specifically, instructor-provided examples created higher lesson clarity for students which helped them organize the lesson material into a more coherent mental model and, subsequently, score higher on the test. The serial indirect effect of example type on learning occurred most prominently for students who were disinterested in the lesson material.

How do people read the passages during a reading comprehension test? The effect of reading purpose on text processing behavior

ABSTRACTIn this study, we investigate the effect of reading purpose on students’ processing behavior during a reading comprehension test. In a repeated measures design, sixty undergraduates answered multiple-choice (MC) reading comprehension questions in a condition with no overarching goal for reading and in an alternate condition where the same students were first provided with the goal of summarizing the text before answering MC questions. Results from eye tracking analysis showed that when students read and answered questions without an overarching goal, they spent much less time reading the passages before answering the questions, more time re-reading the texts while answering the questions, and more time on parts of the text that were not necessary to answer the questions. We conclude that providing examinees without an explicit goal for reading may inadvertently encourage a “search for the answer” reading process, rather than on building a coherent mental model of text content.

The development of wrap-up processes in text reading: A study of children's eye movements.

Reading comprehension is the product of constructing a coherent mental model of a text. Although some of the processes that are necessary to construct such a mental model are executed incrementally, others are deferred to the end of the clause or sentence, where integration processing is wrapped up before the reader progresses further in the text. In this longitudinal study of 65 German-speaking children across Grades 2, 3, and 4, we investigated the development of wrap-up processes at clause and sentence boundaries by tracking the children's eye movements while they read age-appropriate texts. Our central finding was that children in Grade 2 showed strong wrap-up effects that then slowly decreased across school grades. Children in Grades 3 and 4 also increasingly used clause and sentence boundaries to initiate regressions and rereading. Finally, children in Grade 2 were shown to be significantly disrupted in their reading at line breaks, which are inherent in continuous text. This disruption decreased as the children progressed to Grades 3 and 4. Overall, our results show that children exhibit an adultlike pattern of wrap-up effects by the time they reach Grade 4. We discuss this developmental trajectory in relation to models of text processing and mechanisms of eye-movement control. (PsycINFO Database Record

Primary students’ conceptions of the Earth: Re-examining a fundamental research hypothesis on mental models

Research on pupils’ conceptions of the earth has proposed certain mental models within the theoretical perspective known as coherent or theory-like knowledge. Alternatively, the fragmented knowledge hypothesis refutes the existence of such models and proposes a different perspective. Although the relevant discussion has not been converged in to a definite answer, recently, the debate between the two theories has been brought up into consideration by the advances in methodology and statistical analysis. In this paper pupils’ conceptions of the earth were analyzed by latent class analyses. Children’s ideas (N=184, grades 1st to 3rd), were investigated using a closed-ended questionnaire, which includes illustrations corresponding to certain mental models. The results showed that pupils’ conceptions of the earth are not characterized by consistency and thus they do not support the existence of coherent mental models. Implications for theory and practice are discussed.

KEBERKESANAN PERMAINAN PENDIDIKAN TERHADAP PEMBELAJARAN HUKUM NEWTON <br> THE EFFECTIVENESS OF EDUCATIONAL GAMES ON NEWTON’S LAWS LEARNING

The purpose of this study is to examine the effectiveness of Newton’s Laws educational games in improving Newton’s Laws achievement and instructional efficiency. Two educational games were developed, namely educational games with self-explanation (DPK) and educational games without self-explanation (TPK). Based on Rapid Prototyping model, these educational games were developed in Malay Language and the contents were arranged according to Form Four Physics syllabus. Unity3D software was used as the main software in developing the educational games. A total of 196 students attended Physics play activities for two hours. Analyses of Newton’s Laws Test, mental load and interviews were conducted to examine the effectiveness of these Newton’s Laws educational games. The findings show that the students’ achievement and instructional efficiency in Newton’s Laws improved with the use of the educational games. This research concludes that the educational games mode plays an important role in increasing the students’ achievement and learning instructional efficiency with the DPK mode is the most effective and efficient. The educational games mode serves as a scaffold for the formation of a coherent mental model and automatic schema. Keywords: Educational games, Newton’s Laws, Physics, achievement, instructional efficiencyCite as: Alias, S.N., & Ibrahim, F. (2017). Keberkesanan permainan pendidikan terhadap pembelajaran hukum Newton [The effectiveness of educational games on Newton’s laws learning]. Journal of Nusantara Studies, 2(1), 71-85.

Echo questions as a means of building coherence in conversational discourse

Abstract The study focuses on the cognitive-communicative characteristics of echo questions in English conversational discourse. Drawing on van Dijk's sociocognitive (mental model) theory and cognitive discourse analysis, the paper suggests viewing echo questions as a means of building/updating a mental context model of a communicative situation. As discourse comprehension presupposes building its coherent mental model, echo questions resolving misunderstanding are regarded as an instrument for increasing coherence in conversational discourse. Based on the mental model theory, the study offers a typology of misunderstandings corrected by echo questions.

The Role of Task Understanding on Younger and Older Adults' Performance.

Age-related performance decrements have been linked to inferior strategic choices. Strategy selection models argue that accurate task representations are necessary for choosing appropriate strategies. But no studies to date have compared task representations in younger and older adults. Metacognition research suggests age-related deficits in updating and utilizing strategy knowledge, but other research suggests age-related sparing when information can be consolidated into a coherent mental model. Study 1 validated the use of concept mapping as a tool for measuring task representation accuracy. Study 2 measured task representations before and after a complex strategic task to test for age-related decrements in task representation formation and updating. Task representation accuracy and task performance were equivalent across age groups. Better task representations were related to better performance. However, task representation scores remained fairly stable over the task with minimal evidence of updating. Our findings mirror those in the mental model literature suggesting age-related sparing of strategy use when information can be integrated into a coherent mental model. Future research should manipulate the presence of a unifying context to better evaluate this hypothesis.

Systems learning with a conceptual representation: a quasi-experimental study

In this paper, we share results from a classroom intervention that used a conceptual representation to support reasoning about ecosystems. Engaging students in modeling allows them to make their ideas visible while being malleable and available for discussion, which enables students to make meaning out of systems. Further, the Components-Mechanisms-Phenomena (CMP) conceptual representation was designed to enable students to construct coherent mental models. Following our intervention, students deepened their understanding of ecosystem dynamics when compared to students who engaged in traditional instruction without use of the CMP conceptual representation. We discuss our results in terms of data that helped guide the design of the intervention and we describe a theoretical perspective that can be used to guide future instruction.