Search

One of the major purposes of this study is to investigate the potential impact of gender and information and computer technology (ICT) resources on students' computational thinking (CT) competencies. To this end, the Computational Thinking Test for Junior High Students (CTT-JH) was developed and validated. Research participants included 437 junior high school students in Taiwan. The surveyed schools were categorized into "more" or "fewer" ICT resources. Then, discrimination analyses and Rasch modeling for item analyses and two-way ANOVA were conducted. Results showed that the final version of CTT-JH is of good item quality. Students in schools with more ICT resources had higher CT test mean scores regardless of gender. Nevertheless, at schools with limited resources, male students had significantly lower CT test mean scores than female students did. This study provides new insights into how gender and ICT resources can interact with and impact on students' CT competencies. It also provides a valid and reliable tool for assessing young adolescents' CT abilities.

This study aimed to develop the Computational Thinking Test for Elementary School Students (CTT-ES) to assess young children's CT competencies in non-programming contexts and also examine the relationship between CT competencies and CT dispositions. A survey including a pool of CTT-ES candidate items and the Computational Thinking Scale (CTS) was administered to 631 elementary school students. Rasch model of the Item Response Theory and the discrimination analysis of the Classical Testing Theory were conducted for item analyses. Pearson's correlation analyses and hierarchical multiple regression analyses were used to examine the relationships between CTT-ES and CTS scores. The results showed that the final CTT-ES including 16 items had a good fitness, discrimination, and reliability to evaluate elementary students' domain-general CT competencies. The convergent validity of CTT-ES was confirmed by its significant correlations with the CTS scores. The significant regression model not only showed students' CT competencies can be predicted by their CT dispositions but also supported The Developmental Model of CT. This study provided a valid and reliable tool for assessing young children's CT abilities. It also furthered our understanding about the developmental orders of CT abilities and contributed to the theoretical construction of CT.

A prior study developed the Computational Thinking Scale (CTS) for assessing individuals' computational thinking dispositions in five dimensions: decomposition, abstraction, algorithmic thinking, evaluation, and generalization. This study proposed the Developmental Model of Computational Thinking through validating the structural relationships among the five factors of the CTS. To examine the model, a questionnaire including the CTS was administered to 472 middle school students. A confirmatory factor analysis was used to confirm the construct of the measurements, and a PLS-SEM analysis was used to validate the structural relationships among the factors. The results confirmed that the 19-item CTS has good item reliability, internal consistency, and construct reliability for measuring computational thinking (CT). In the Developmental Model of CT, decomposition and abstraction significantly predict all other three CT dispositions, suggesting that they are the two fundamental factors required for CT development. Moreover, a significant linear prediction path was shown starting from algorithmic thinking, evaluation, until generalization. Thus, a multi-level model was confirmed for the conceptual framework of CT. This model suggests a possible sequence for CT development which may provide a guideline for the teaching objectives of CT for different learning stages in different school levels. Decomposition and abstraction are especially suggested to be emphasized in school curricula before teaching algorithmic thinking or algorithm designs.

The purpose of this study was to explore how students interacted with guidance to conduct a scientific inquiry in a physics simulation by using the eye-tracking techniques. The participants were 51 7th graders, and an eye-tracking system was used to record their visual behaviors and log data while they were using the simulation. As for data analysis, we first checked each participant's log data to examine whether they completed the requirement of the guidance, and then checked the correctness of her/his answer to the inquiry task. The participants were thus divided into two groups (correct vs. wrong), and the patterns of their visual behaviors were examined by a set of eye-movement indices, normalized heat maps and lag sequential analyses. The results indicate that both spatial distributions and temporal sequences of the participants' visual attention were associated with their performances on the inquiry task. Regarding the spatial distribution, the correct group tended to allocate more visual attention to the regions presenting the target phenomenon than the wrong group. Concerning the temporal sequence, the correct group tended to make more visual transitions among the content of the guidance, the relevant control panels and the target phenomena than the wrong group.

Computational thinking has received tremendous attention from computer science educators and educational researchers in the last decade. However, most prior literature defines computational thinking as thinking outcomes rather than thinking processes. Based on Selby and Woodland's framework, this study developed and validated the Computational Thinking Scale (CTS) to assess all students' thought processes of computational thinking for both general and specific problem-solving contexts in five dimensions: abstraction, decomposition, algorithmic thinking, evaluation and generalization. A survey including 25 candidate items for CTS as well as demographic variables was administered to 388 junior high school students in Taiwan. An explorative factor analysis using the principal axis method with the oblimin rotation was used to validate the scale. Finally, 19 items were extracted successfully under the designed five dimensions, with a total explained variance of 64.03% and an overall reliability of 0.91. Results of the demographic comparisons showed that boys had a greater disposition than girls in decomposition thinking when solving problems using computer programming. In addition, programming learning experience, especially self-directed learning and after-school learning, had significant positive effects on all dimensions of CTS. Several future studies are suggested using this tool.

Although a modality effect indicating better learning from an animation with spoken learning guidance than from an animation with written learning guidance has often been found in research on learning from animation, the effect has hardly been investigated for learning from computer simulation. This study examined the interaction effects between the multimedia types of animation and computer simulation, and the presentation modes of written and spoken learning guidance. Based on cognitive load theory and the specific characteristics of the multimedia types it was hypothesized that a modality effect would be replicated in the animation condition and a reverse modality effect would be found in the simulation condition. Results partially supported the hypotheses, indicating that participants learned more from animation when it was accompanied with spoken text, and from a simulation when it was accompanied with written text. However, no effect was found for cognitive load. Based on the results, spoken text is recommended to be used to guide learners in animations assisted learning environments, and written text is recommended to be used to guide learners in simulations assisted learning environments.

This study aimed to investigate differences in junior and senior (categorised by their years of teaching experience) elementary school teachers' perceptions of the Technological Pedagogical Content Knowledge -- Games (TPACK-G), attitudes towards game, and actual teaching usage. It also aimed to probe how each of the two groups' TPACK-G perceptions predicted their attitudes towards games and actual teaching usage. 376 in-service teachers in Taiwan filled out the questionnaire consisting of game knowledge (GK), game pedagogical knowledge (GPK), game content knowledge (GCK), game pedagogical content knowledge (GPCK), attitudes towards games (AT), and actual teaching usage (AU). The results showed that junior teachers tended to have higher GK, GCK and GPCK than the seniors. GK alone might not be sufficient to predict the teachers' actual teaching usage. To predict teachers' actual teaching usage, the junior teachers might rely on their GPCK, whereas the senior teachers might rely on their GPK.

Introduction: The clinical characteristics of patients positive for anti-small ubiquitin-like modifier 1-activating enzyme subunit 1 (SAE1) antibodies and diagnosed with idiopathic inflammatory myopathies (IIMs) vary across different cohorts and ethnicities, particularly concerning interstitial lung disease (ILD). We aimed to assess the clinical utility of the line immunoblot assay (LIA) in detecting anti-SAE1 autoantibodies and evaluate the clinical relevance and chronology of ILD development in relation to SAE1 autoantibody positivity among Taiwanese patients. Methods: We retrospectively conducted a population-based cohort analysis involving 6,496 patients who visited Chang Gung Memorial Health System across Taiwan from May 2018 to December 2021. Patients were assayed for myositis-specific autoantibodies (MSAs) and myositis-associated autoantibodies (MAAs) using the LIA method, and the antinuclear antibody (ANA) indirect immunofluorescence (IIF) method was used to evaluate ANA patterns. Of these, 70 SAE1-positive patients (1.08%) were included and followed up until December 2023. Associations with clinical characteristics and final diagnosis, particularly ILD, were assessed. Results: Among the 70 SAE1-positive patients, 10 (14.3%) were strongly positive and 60 (85.7%) were weakly positive. In the strong positive group, 70% (7/10) were diagnosed with IIM, with most (5/7) showing a concordant ANA IIF pattern (speckled type). Six patients presented ILD either before (1/6) or after (5/6) IIM diagnosis; the majority (4/6) were classified as organizing pneumonia. The remaining 30.0% (3/10) had connective tissue disease (CTD) other than IIM without detectable ILD during follow-up, and none demonstrated a concordant ANA IIF pattern. In the weakly positive group, only 5.0% (3/60) had IIM and 3.3% (2/60) had ILD. The positive predictive value for strong positive SAE1 autoantibodies in diagnosing IIM was significantly higher than for weak positives (70.0% vs. 5.0%; p < 0.001). Conclusions: The study suggests that strong positive SAE1 autoantibodies detected via LIA are more closely associated with IIM compared to weak positive results. A high prevalence of ILD was observed among strong positive Taiwanese patients, indicating the need for prompt screening. Patients with weak positive or discordant ANA IIF results may represent false positives with a lower ILD risk. [ABSTRACT FROM AUTHOR]

While research has shown that students' epistemic beliefs can be a strong predictor of their academic performance, cognitive abilities, or self-efficacy, studies of this topic in computer education are rare. The purpose of this study was twofold. First, it aimed to validate a newly developed questionnaire for measuring students' epistemic beliefs about computer programming at middle school level. Second, the study aimed to understand how students' epistemic beliefs predict their computational thinking disposition and computer programming self-efficacy. A total of 406 middle school students completed a survey consisting of three research instruments: the Computer Programming Epistemic Belief Inventory (CPEBI), the Computational Thinking Scale (CTS), and the Computer Programming Self-Efficacy Scale (CPSES). The results indicated that the epistemic belief of "uncertainty" was a significant and positive predictor of the "abstraction" and "generalization" dimensions of computational thinking dispositions. In addition, the belief of "complexity" was an antecedent to three dimensions of computational thinking, namely, "algorithmic thinking," "evaluation," and "generalization." Only the "evaluation" dimension of CT disposition had significant and positive relationships with "programming self-efficacy." The roles of "uncertainty" and "complexity" of computer programming epistemic beliefs are discussed. Implications and suggestions for future computer programming education, especially for young students, are provided. [ABSTRACT FROM AUTHOR]

In the field of science education, laboratory learning environment has gained renewed interest in the recent decade. This study aimed to investigate the relationships among students' conceptions of science laboratory learning, perceptions of the science laboratory learning environment, and their academic self-efficacy in science learning by adopting the structural equation modeling (SEM) technique. A total of 513 senior high school students (262 females) in Taiwan were invited to participate in this survey study. Three instruments were adapted and implemented to investigate the aim of the study (i.e. the conceptions of science laboratory learning questionnaire, the science laboratory environment inventory, and the academic self-efficacy in science learning questionnaire). The results indicated that the students' conceptions of science laboratory learning made a significant contribution to their perceptions of the science laboratory environment, which consequently fostered their science learning self-efficacy. More specifically, students with conceptions of science laboratory learning as reviewing their prior learning profiles tended to highlight the "student cohesiveness," "integration," and "material environment" aspects of the laboratory environment. Moreover, students who held personal ideas of science laboratory learning as acquiring manipulative skills tended to perceive actual science laboratory environments as much more open-ended and to attain advanced academic science learning self-efficacy. In addition, those students who viewed laboratory learning as achieving in-depth understanding, and who perceived that laboratory activities are guided by clear rules, were prone to express a stronger sense of academic self-efficacy. Based on the results, practical implications and suggestions for future research are discussed.

This cross-sectional research was conducted to probe how 626 Taiwanese children's conceptions of learning differ by age and gender via drawing. Participants were from the kindergarten, the first grade, and the third grade. A coding scheme categorizing the people, learning domain, and place in the drawings was developed. Chi-square tests were then performed to examine the associations among the children's three grade levels, gender, and three categories: people, learning domain, and place. The findings showed that the kindergarteners tended to draw more family members as well as family and outdoor scenes, while the third graders tended to draw more school figures as well as classroom and school outdoor scenes. The first graders drew more peers than the third graders did. The kindergarteners drew significantly more social, informal, and playful activities than the other two groups did. The place for literacy learning switched from school (kindergarteners and first graders) to home (third graders). No gender difference was identified. This study provides insight into developing curricula that smooth children's transition from kindergarten to elementary school.

This study, first of all, aimed to develop a new survey to assess Taiwanese teachers' perceived self-efficacy in STEM knowledge. Second, it aimed to probe any differences in teachers' perceived self-efficacy in STEM knowledge regarding their gender and teaching subjects. Last, we examined the structural relations among teachers' perceived self-efficacy in STEM knowledge and their attitudes toward STEM education. The participants were 220 high school teachers in Taiwan. The 30-item instrument consisted of six factors: scientific inquiry, technology use, engineering design, mathematical thinking, and synthesized knowledge of STEM, as well as attitudes toward STEM education. The results showed that the proposed instrument was valid and reliable. In addition, male teachers outperformed female teachers in each dimension of the survey. Last, teachers' self-efficacy in synthesized knowledge of STEM had two mediating effects. One was in the relationship between self-efficacy in engineering design and attitudes toward STEM education. The other was in the relationship between self-efficacy in Mathematical Thinking and Attitudes toward STEM education.

This study was to investigate the role of students' game self-efficacy in their game performance as well as in their visual behavior distributions and transfer patterns during their gameplay. A total of 48 university students were recruited for participating in a gameplaying task while wearing with an ASL Mobile Eye eye-tracker. After the gameplay, each participant was asked to report their self-efficacy of the gameplay. Based on the self-reported game self-efficacy (GSE) score, the participants were divided into the higher and lower GSE groups. Independent t tests were used to compare their game performances and eye-tracking indices for each area of interest. A lag sequential analysis was used to profile the visual transfer pattern for each group. The results showed that the higher GSE group had better performances than the lower GSE group. During the gameplay, the higher GSE students tended to pay more attention to the critical information for a successful play. In addition, the lag sequential analyses revealed different visual transfer patterns for the two GSE groups. The higher GSE group tended to show a critical-factor-centralized visual-trasfer pattern which demonstrated more planning and monitoring gameplaying strategies; while the lower GSE students, probably lacking the self-confidence of gameplaying, tended to pay more attention to the remaining time for the game.This study may provide suggestions for future game-based leraning system designs.

Objective: To investigate the psoriatic disease risk among patients with previous appendicitis. Methods: This study was a nationwide population‐based case–control study about the association between the psoriatic disease risk among patients with a history of appendicitis in Taiwan. The study population consisted of newly diagnosed psoriatic disease with at least two outpatient visits, and the control group included those patients without psoriatic disease at the same index date. Patients with a previous diagnosis of appendicitis or who underwent appendectomy surgery prior to their psoriatic disease diagnosis were recorded. The odds ratio of psoriatic disease diagnosis in the two groups with and without a history of appendicitis were analyzed and compared. Results: A total of 48 894 individuals diagnosed with psoriatic disease were matched with 292 656 controls by age and gender. Notably, the proportion of patients with a history of appendicitis or primary appendectomy was significantly elevated among those with psoriatic disease compared with the control cohort (both p <.05). On average, the occurrence of appendicitis preceded the index date by 3.3 ± 2.3 years. Multivariate analysis revealed a heightened incidence rate of psoriatic disease in patients previously diagnosed with appendicitis, periodontal disease, Charlson comorbidity index score (CCIS) ≧1, and ill‐defined intestinal infections. This association persisted after adjusting for confounding factors, such as periodontal disease, CCIS, Salmonella, and ill‐defined intestinal infections. The odds ratios for psoriatic disease in individuals with a history of appendicitis, periodontal disease, CCIS ≧1, and ill‐defined intestinal infections were 1.16, 1.008, 1.69, and 1.23, respectively, with corresponding 95% confidence intervals of 1.03–1.31, 1.05–1.11, 1.65–1.74, and 1.20–1.26. These findings underscore the independent association between appendicitis and subsequent development of psoriatic disease, even after adjusting for relevant comorbidities and potential confounders. Conclusion: Our research illustrates that appendicitis is associated with an increased likelihood of developing a psoriatic disease, despite several limitations. These limitations encompass variables such as dietary and smoking habits, alongside other potential confounding factors that were not fully considered. Moreover, inherent biases in utilizing national health insurance data, such as the absence of laboratory information, as well as the constraints inherent in a retrospective study design, should be acknowledged. [ABSTRACT FROM AUTHOR]

This research aimed to investigate the structural relationships among teachers' computational thinking (CT), design thinking (DT), robotics teaching beliefs, and robotics pedagogical content knowledge (RPCK). A total of 98 in-service and pre-service teachers who participated in a robotics teaching professional development workshop served as the sample of the study. A survey including the Computational Thinking Scale, the Design Thinking Disposition Scale, the Robotics Teaching Beliefs Scale and the Technological Pedagogical Content Knowledge–Robotics Scale was conducted after the workshop. A confirmatory factor analysis was employed to validate the measurement constructs, and Partial Least Squares - Structural Equation Modeling (PLS-SEM) analysis was utilized to examine the relationships among the factors. The results revealed that both CT and DT dispositions could positively predict teachers' robotics teaching beliefs, which subsequently predicted their RPCK. Moreover, a direct positive relationship between CT and RPCK was identified, while such a relationship was not evident for DT. The model demonstrates the critical role of CT in shaping teachers' beliefs and pedagogical strategies of robotics teaching, and provides insights into the indirect influence of DT. Finally, the Model of Robotics Teaching Professional Development (MRTPD) was proposed to profile how to promote teachers' pedagogical content knowledge of robotics teaching from their CT and DT dispositions. [ABSTRACT FROM AUTHOR]

The purpose of this study was to develop and validate an online contextualized test for assessing students' understanding of epistemic knowledge of science. In addition, how students' understanding of epistemic knowledge of science interacts with learner factors, including time spent on science learning, interest, self-efficacy, and gender, was also explored. The participants were 489 senior high school students (244 males and 245 females) from eight different schools in Taiwan. Based on the result of an extensive literature review, we first identified six factors of epistemic knowledge of science, such as status of scientific knowledge, the nature of scientific enterprise, measurement in science, and so on. An online test was then created for assessing students' understanding of the epistemic knowledge of science. Also, a learner-factor survey was developed by adopting previous PISA survey items to measure the abovementioned learner factors. The results of this study show that: (1) by factor analysis, the six factors of epistemic knowledge of science could be grouped into two dimensions which reflect the nature of scientific knowledge and knowing in science, respectively; (2) there was a gender difference in the participants' understanding of the epistemic knowledge of science; and (3) students' interest in science learning and the time spent on science learning were positively correlated to their understanding of the epistemic knowledge of science.

Using the Game-based-learning Teaching Belief Scale (GTBS) and the Technological Pedagogical Content Knowledge--Games questionnaire (TPACK-G), this study investigated 316 Taiwanese in-service teachers' teaching beliefs about game-based learning and their perceptions of game-based pedagogical content knowledge (GPCK). Both t-tests and ANOVA analyses were used to examine the differences due to their demographic backgrounds. Correlation and regression analyses were used to examine the predicting variables of the teachers' GPCK. The results showed that, among the factors of GTBS and TPACK-G, GPK plays the most critical role in predicting GPCK. In addition, the elementary school teachers had stronger Belief, Confidence and Motivation for using game-based teaching approaches, and had higher GPK and GPCK than middle school teachers. A gender difference only existed in GK. Younger teachers had better GK and GCK than older teachers. Compared with experienced teachers, novice teachers tended to believe that digital games can assist learning and instruction, and perceived higher self-efficacy in their TPACK-G.

Background: Rheumatoid arthritis (RA) is a major risk factor for osteoporosis/osteoporotic fractures. We aimed to elucidate the role of treatment choices among osteoporosis/osteoporotic fractures. Methodology: We utilized the Chang‐Gung Research Database to assess the risks of osteoporosis/osteoporotic fractures among independently treated RA patients, using retrospective time‐to‐event outcomes analysis. Results: A total of 3509 RA patients with a mean of 63.1 ± 8.6 years were analyzed. Among all, 1300 RA patients (37%) were diagnosed with newly diagnosed osteoporosis. The crude incidence of newly diagnosed osteoporosis was the highest among those treated with other conventional disease‐modifying anti‐rheumatic drugs (cDMARDs; 74.1 events/1000‐PYs, 95%CI 66.0–82.3), followed by those with a non‐treatment period (68 events/1000‐PYs, 95%CI 63.1–72.9), methotrxate (MTX) monotherapy (60.7 events/1000‐PYs, 95%CI 41.2–80.3), MTX plus other cDMARDs (51.9 events/1000‐PYs, 95%CI 43.4–60.3), and abatacept/rituximab (48.6 events/1000‐PYs, 95%CI 14.9–82.3). The lowest crude incidence was found in patients treated with anti‐TNFi biologics (40.4 events/1000‐PYs, 95%CI 28.6–52.2) and other biologic disease‐modifying anti‐rheumatic drugs (bDMARDs; 40.1 events/1000‐PYs, 95%CI 8.0–72.1). A total of 270 patients (20.8%) suffered from an incident fracture during follow‐ups. The crude incidence of fracture was the highest among those treated with abatacept/rituximab (49.0 events/1000‐PYs, 95%CI 6.0–91.9), followed by those with non‐treatment periods (24.3 events/1000‐PYs, 95%CI 19.3–29.4), other cDMARDs (24.2 events/1000‐PYs, 95%CI 18.1–30.2), anti‐TNFi biologics (20.2 events/1000‐PYs, 95%CI 8.8–31.6). Other bDMARDs (13.3 events/1000‐PYs, 95%CI 0–39.2), MTX mono (12.5 events/1000‐PYs, 95%CI 0.3–24.8), and MTX plus other cDMARDs (11.4 events/1000‐PYs, 95%CI 5.4–17.4) were low incidences. Conclusion: The treatment option has emerged as a critical determinant in the context of future osteoporosis and osteoporotic fracture risks among RA. These findings offer a valuable resource for clinicians, empowering them to tailor bespoke treatment strategies for RA patients, thereby mitigating the potential for future osteoporosis and fractures. [ABSTRACT FROM AUTHOR]

The purpose of this study was to examine the impacts of embedding collaboration into a game with a self-explanation design for supporting the acquisition of light and shadow concepts. The participants were 184 fourth graders who were randomly assigned to three conditions: a solitary mode of the game with self-explanation, a collaborative mode with self-explanation, or the control condition of a single-user game without integrating self-explanation. Students' conceptual understanding was measured through an immediate posttest and a retention test with a three-week delay. Further, students' engagement in answering the prompts was also investigated. The findings showed that having students collaboratively play science-based games with a self-explanation design embedded was not sufficient to help them learn the science concepts. Rather, it was the level of engagement in responding to the self-explanation prompts that mattered.

Previous studies have suggested that understanding the relationships and developmental paths among TPACK constructs plays a crucial role in teachers' TPACK development process. Researchers (Chai et al.," Edu Technol Soc" 13(4):63-73, 2010) have found that technological knowledge, pedagogical knowledge, and content knowledge are all significant predictors of pre-service teachers' TPACK. Thus, one of the major concerns that educators may have is, which type of knowledge should come first in an educational technology course in order to enhance the participants' TPACK. This study was conducted to examine the effects of the technology- and pedagogy-oriented course design on improving in-service preschool teachers' technological pedagogical content knowledge-Games (TPACK-G) as well as their acceptance of digital game-based learning. The participants were 49 in-service preschool teachers. They were assigned into a technology- and a pedagogy-oriented group. The results show that when integrating the TPACK-G framework into the preschool context, teachers who were taught with game knowledge first tended to have higher competencies in game knowledge and game pedagogical content knowledge than those who were first instructed with game pedagogical knowledge.

Online information searching tasks are usually implemented in a technology-enhanced science curriculum or merged in an inquiry-based science curriculum. The purpose of this study was to examine the role students' different levels of scientific epistemic beliefs (SEBs) play in their online information searching strategies and behaviors. Based on the measurement of an SEB survey, 42 undergraduate and graduate students in Taiwan were recruited from a pool of 240 students and were divided into sophisticated and naïve SEB groups. The students' self-perceived online searching strategies were evaluated by the Online Information Searching Strategies Inventory, and their search behaviors were recorded by screen-capture videos. A sequential analysis was further used to analyze the students' searching behavioral patterns. The results showed that those students with more sophisticated SEBs tended to employ more advanced online searching strategies and to demonstrate a more metacognitive searching pattern.

Current technological pedagogical content knowledge (TPACK) studies are inclined to treat technology in a general manner, an approach which may not be able to provide adequate guidelines to improve teacher preparation and professional development when teaching with games. This study developed two new questionnaires, namely the Technological Pedagogical Content Knowledge--Games (TPACK-G) survey to investigate 352 preschool teachers' confidence in their TPACK-G, and the Acceptance of Digital Game-Based Learning (ADGBL) survey to assess their acceptance of game-based learning. The results show that both instruments exhibit satisfactory validity and reliability. Through a path analysis, the framework of the TPACK-G proposed in this study was supported. In addition, preschool teachers' experience with games and their attitudes toward game-based learning contributed positively to their reported game knowledge. Perceived learning opportunities and attitudes toward game-based learning had direct relations with game pedagogical knowledge. Lastly, experience with games contributed to game pedagogical content knowledge.

Educational researchers have indicated that although computer games have the potential to promote students' motivation and engagement, the work on how to design effective games that fulfil educational purposes is still in its infancy. This study aimed to examine how integration of self-explanation into a computer game affected primary schoolers' acquisition of light and shadow concepts. The participants were randomly assigned to either an experimental group or a control group and played a computer game with or without self-explanation prompts individually as a treatment. Students' conceptual understanding was evaluated through a pretest and a posttest administered right after the treatment. The results revealed that by controlling the pretest scores, students who played the game with self-explanation features did not outperform those who played the game without any prompts in the posttest. Further analyses of the experimental group students' responses to the self-explanation prompts also indicated that the students with more correct responses to the prompts did not perform better than those with lower accuracy rates. The deficits in the use of self-explanation prompts are identified, and possible improvements to enhance the function of self-explanation in educational games are proposed. (Contains 3 tables and 3 figures.)