Search

Adaptive learning systems aim to provide learning items tailored to the behavior and needs of individual learners. However, one of the outstanding challenges in adaptive item selection is that often the corresponding systems do not have information on initial ability levels of new learners entering a learning environment. Thus, the proficiency of those new learners is very difficult to be predicted. This heavily impairs the quality of personalized items' recommendation during the initial phase of the learning environment. In order to handle this issue, known as the cold-start problem, we propose a system that combines item response theory (IRT) with machine learning. Specifically, we perform ability estimation and item response prediction for new learners by integrating IRT with classification and regression trees built on learners’ side information. The goal of this work is to build a learning system that incorporates IRT and machine learning into a unified framework. We compare the proposed hybrid model to alternative approaches by conducting experiments on two educational data sets. The obtained results affirmed the potential of the proposed method. In particular, the obtained results indicate that IRT combined with Random Forests provides the lowest error for the ability estimation and the highest accuracy in terms of response prediction. This way, we deduce that the employment of machine learning in combination with IRT could indeed alleviate the effect of the cold start problem in an adaptive learning environment. ispartof: Computers and Education vol:137 pages:91-103 status: Published online

© 2018, Psychonomic Society, Inc. Electronic learning systems have received increasing attention because they are easily accessible to many students and are capable of personalizing the learning environment in response to students’ learning needs. To that end, using fast and flexible algorithms that keep track of the students’ ability change in real time is desirable. Recently, the Elo rating system (ERS) has been applied and studied in both research and practical settings (Brinkhuis & Maris, 2009; Klinkenberg, Straatemeier, & van der Maas in Computers & Education, 57, 1813–1824, 2011). However, such adaptive algorithms face the cold-start problem, defined as the problem that the system does not know a new student’s ability level at the beginning of the learning stage. The cold-start problem may also occur when a student leaves the e-learning system for a while and returns (i.e., a between-session period). Because external effects could influence the student’s ability level during the period, there is again much uncertainty about ability level. To address these practical concerns, in this study we propose alternative approaches to cold-start issues in the context of the e-learning environment. Particularly, we propose making the ERS more efficient by using an explanatory item response theory modeling to estimate students’ ability levels on the basis of their background information and past trajectories of learning. A simulation study was conducted under various conditions, and the results showed that the proposed approach substantially reduces ability estimation errors. We illustrate the approach using real data from a popular learning platform. ispartof: Behavior Research Methods vol:51 issue:2 pages:895-909 ispartof: location:United States status: published

This study aims to compare the effects of oral, paper and augmented reality (AR) instructions during assembly tasks. 44 assembly operators, the majority of which have cognitive or motor disabilities, participate in a within-subjects experiment in which they perform three different assembly tasks. In doing so, the different instructional media are alternated. The outcome variables under study are productivity, quality, stress, help-seeking behaviour, perceived complexity, physical effort and competence frustration. In addition, this study examines the effects of characteristics of operators by taking into account their cognitive skills and experience with each task type as predictor variables. Finally, as operators perform one of the assembly tasks four times, learning curves are studied as well. Results of a multilevel analysis show that operators using AR instructions outperform their colleagues using more traditional instructional media with regard to quality. In addition, AR instructions give rise to less help-seeking behavior as compared to oral instructions. Finally, there is some evidence for AR instructions to yield a lower perceived complexity than oral instructions. Furthermore, the results indicate that the operators’ cognition and experience play a significant role in terms of many outcome variables. Learning curves reveal that operators using AR instructions experience less stress from the first assembly attempt onwards, whereas for operators using oral or paper instructions, it takes several attempts before being equally relaxed. ispartof: International Journal Of Human-Computer Studies vol:143C status: Published online

Digital educational games play an increasingly important role in education. However, multiple questions about the effectiveness of educational games with respect to cognitive and non-cognitive effects remain unclear. The current study, a longitudinal, quasi-experiment with 336 first graders, examined the effects of two digital educational games, Number Sense Game (NSG) and Reading Game (RG). The NSG trained early numerical skills, the RG supported emergent reading. Children were pseudo-randomly assigned to either an experimental condition, comprising eight weeks of intensive game-based training, or a control condition in which they took part in regular education without game-based practice. A pretest-posttest design was used to examine the effects of the intervention on cognitive (digit comparison, number line estimation, letter knowledge, math and reading competence) and non-cognitive outcomes (math and reading anxiety). Delayed cognitive effects on math and reading competence were also investigated two months after the intervention. Furthermore, we examined variances of the impact of the training on cognitive outcomes as a consequence of differences in children's prior knowledge, prior affect and socio-economic status. For cognitive outcomes, results revealed that children who played a game performed better on number line estimation and reading competence, whereas no significant differences were observed for digit comparison, letter knowledge and math competence. Also, children who played a game showed better scores in the delayed reading posttest, but not in the delayed math posttest. For non-cognitive outcomes, game training did not affect math or reading anxiety. Regarding individual differences, children with less prior knowledge in the game play condition performed better on the number line estimation posttest compared to children in the control condition. Children with more prior knowledge in the game play condition still scored better on this test compared to the control condition, but the difference between the conditions was smaller.

The use of educational games is nowadays well-known, although its effectiveness in terms of learning and motivation has not been firmly corroborated yet. In this research, the focus is on the use of adaptive item sequencing in an educational math game (Monkey Tales) as a way to provide learning content that is adapted to the knowledge level of the learner or player. One way to provide this is adjusting the difficulty level of the content that is presented hence allowing for optimal learning during gameplay. In order to realize this, accurate estimates of the learners' skills are necessary, as well as accurate estimates of variables in the gaming environment that determine how much difficulty learners may experience while completing the items. The goal of this study is to compare difficulty ratings from users and experts in order to acquire reliable estimates of the difficulty of the math rules offered in Monkey Tales. The results suggest a double format effect: not only the gaming format adds difficulty to educational content, but also the format by which educational content is presented in a game is likely to add difficulty which affects learners' in-game performance.

This paper explores individual difference factors in relation with usage of optional CF in a written and task-based tutorial CALL environment for English grammar practice that contained gaming features. Previous research in CALL has highlighted the importance of prior knowledge for learners’ usage of CF options (Brandl, 1995; Heift, 2002), but the contribution of metacognitive and affective variables remains unexplored. Based on insights from the literature on tool use (e.g. Author 4 & Elen, 2009), this pilot study considered that learners’ usage of optional CF in CALL might additionally be determined by the perceived usefulness of CF and by learners’ achievement goal orientation. Quantitative analysis of questionnaire data in combination with tracking and logging data showed that usage of optional CF was associated with prior explicit knowledge, but no relation was found with perceived usefulness and achievement goal orientation. Future research on CF usage could benefit from qualitative in-depth analyses of learners’ perceptions and motives.

List of figures List of tables Acknowledgements Permissions List of acronyms List of contributors Introduction Fiona Farr and Liam Murray Part I: Historical and conceptual contexts * Language learning and technology: past, present, and future Deborah Healey * Theory in Computer-Assisted Language Learning research and practice Philip Hubbard and Mike Levy * Towards an 'ecological' CALL theory: theoretical perspectives and their instantiation in CALL research and practice Francoise Blin Part II: Core issues * Technology standards for language teacher preparation Greg Kessler * Researching participatory literacy and positioning in online learning communities Mirjam Hauck, Rebecca Galley and Sylvia Warnecke * Language materials development in a digital age Gary Motteram * Researching in language learning and technology Mike Levy * Literacies, technology and language teaching Gavin Dudeney and Nicky Hockly * Evaluation in CALL: tools, interactions, outcomes Catherine Caws and Trude Heift * Language testing and technology James Dean Brown * From age and gender to identity in technology-mediated language learning Elisabeth R. Hayes and Yoonhee N. Lee * Culture, language learning, and technology Robert Godwin-Jones * Language learning and technology in varied technology contexts HyunGyung Lee and Joy Egbert * Limitations and boundaries in language learning and technology Rick Kern and Dave Malinowski * Teacher education and technology Elizabeth Hanson-Smith * Sustainable CALL development Francoise Blin, Juha Jalkanen and Peppi Taalas Part III: Interactive and collaborative technologies for language learning * Telecollaboration and language learning Francesca Helm and Sarah Guth * Social networking and language learning Lara Lomicka and Gillian Lord * Computer supported collaborative writing and language learning Muriel Grosbois * Interactive whiteboards and language learning Euline Cutrim Schmid * Mobile language learning Glenn Stockwell * Virtual worlds and language learning: an analysis of research Mark Peterson * Online and blended language learning Pete Sharma and Kevin Westbrook Part IV: Corpora and data-driven learning * Introduction to data-driven learning Martin Warren * Spoken language corpora and pedagogic applications Andrew Caines, Michael McCarthy and Anne O'Keeffe * Written language corpora and pedagogic applications Angela Chambers * Learner corpora and pedagogic applications Fanny Meunier * Corpus types and uses Brona Murphy and Elaine Riordan * Designing and building corpora for language learning Randi Reppen Part V: Gaming and language learning 30. Metaphors for digital games and language learning Jonathon Reinhardt and Steven Thorne * Mini-games for language learning Frederik Cornillie and Piet Desmet * Gaming and young language learners Pia Sundqvist Part VI: Purpose designed language learning resources * CALL tools for lexico-grammatical acquisition Li Li * CALL tools for reading and writing Hsien-Chin Liou * CALL tools for listening and speaking Una Clancy and Liam Murray * Multimodality and CALL Nicolas Guichon and Cathy Cohen * Intelligent CALL and written language Cornelia Tschichold and Mathias Schulze * Translation and technology: the case of translation games for language learning Pierrette Bouillon, Cristiana Cervini and Manny Rayner

Frederik Cornillie, Steven L. Thorne and Piet Desmet ReCALL / Volume 24 / Issue 03 / September 2012, pp 243 ­ 256 DOI: 10.1017/S0958344012000134, Published online: Link to this article: http://journals.cambridge.org/abstract_S0958344012000134 How to cite this article: Frederik Cornillie, Steven L. Thorne and Piet Desmet (2012). ReCALL special issue: Digital games for language learning: challenges and opportunities. ReCALL, 24, pp 243­256 doi:10.1017/ S0958344012000134 Request Permissions : Click here

A key design issue for tutorial CALL is that controlled practice activities need to engage learners in meaningful L2 processing, so that any knowledge developed in such practice may transfer to meaningful L2 use in complex skills. Furthermore, activities for controlled practice ideally engender intrinsic motivation, so that learners are willing to practise and remediate problems outside of the classroom. This paper reports on an experimental and design-focused study that tackled these challenges by means of mini-games embedded in a mystery story read in class. Results show that intensive receptive practice helped learners to develop knowledge that was accurate and quickly retrievable on tests of near transfer, and that practice also transferred to more complex written productive tasks, as well as—to a smaller extent—to spoken productive tasks. We make suggestions for future design on the basis of the spoken productive language test used in this study.

In order to develop game-based learning environments (GBLEs) that accommodate to learners’ needs and individual differences, GBLEs can be enriched with learner models that describe learner profiles from which adaptive instruction can be offered during gameplay. Learner models can encompass several parameters or learner characteristics derived from measurements taken either prior to play (e.g., already available knowledge of the subject matter of which the GBLE is comprised) or during gameplay (i.e., learner behavior in the GBLE). This study makes a case for two skills which may be relevant from the perspective of adaptive gameplay, namely (1) the knowledge or skills with respect to the learning content and (2) the gaming skills. The current study investigates the joint inclusion of both gaming skills and domain knowledge creating learner profiles. In addition, this study sheds light on how performance during gameplay can be attributed to certain learner profiles. To investigate this, a commercially available 3D educational game for primary school children was offered to 53 children of the third grade. Learners’ behavior while playing in the GBLE was captured and logged. Prior to gameplay, math knowledge, and gaming skills were measured. Subsequently, learners’ in-game performance was measured. Results revealed that learners with high or low gaming skills can be distinguished into two learner profiles. More specific, learners with high gaming skills outperformed learners with low gaming skills in more complex mini-games. The findings of this study suggest that a learner’s gaming skills can be taken into account in developing learner profiles and hence in the design and development of GBLEs.

The use of educational games for teaching and training is nowadays well-known, although its effectiveness in terms of learning and motivation has not been firmly corroborated. A first reason for this is that research on instructional design research often does not reach the fields of game development and game design. Consequently, instructional design principles that have proven to be effective are often not incorporated in educational games. A second reason for the mixed results on the effectiveness of educational games can be found in the way instruction in such games is offered. To our knowledge, educational games rarely account for individual differences between players and research on adaptive educational games is rather sparse. This paper focuses on adaptive approaches in educational games and discusses various player and gameplay characteristics that can be integrated in a framework that conceptualizes player-centered adaptivity in educational games.

This paper reports on an ongoing project which aims to develop and evaluate web-based mini-games for language learning in an evidence-based and user-centred approach. In recent years, a shift is taking place towards more learner-centred learning environments, and designers of Computer Assisted Language Learning (CALL) software have stressed the need for a user-centred design approach (Colpaert, 2010; Hémard, 2003). In general, there is a growing interest in participatory design, in which users are involved in the design process (co-design), marking a move away from traditional design methods characterized by an expert mind-set (Sanders 2008). Many gaming elements, such as competition, interaction/interactivity, problem solving, scoring, and feedback, are also present in traditional learning and testing environments. Hence, it is conceivable that learners' perceptions of these "gaming elements" determine the reception of a language learning activity as a game to a significant extent, rather than the educational intentions of instructional designers alone (Hubbard, 1991). This warrants empirical research on language learners' perceptions of games for learning, as well as user evaluation studies with operational prototypes. © 2012 Springer-Verlag. ispartof: pages:79-81 ispartof: SERIOUS GAMES: THE CHALLENGE vol:280 pages:79-81 ispartof: ITEC/CIP and T2011 location:Ghent, Belgium date:Oct - Oct 2011 status: published

Although there is pedagogical support for using computer adventure and role-playing games in order to learn a second language (L2), commercial games often lack the instructional qualities for making their language comprehensible for learners. In an interdisciplinary approach, this paper proposes a technique for adapting in-game text in order to teach L2 vocabulary, grounded in research on second language acquisition and adaptive learning systems. ispartof: pages:131-146 ispartof: Interdisciplinary Approaches to Adaptive Learning: A Look at the Neighbours vol:126 pages:131-146 ispartof: International Conference on Interdisciplinary Research on Technology, Education and Communication location:Kortrijk date:25 May - 27 May 2010 status: published

One of the major challenges of mobile (language) learning consists in designing content that is based on sound pedagogical (theoretical) frameworks and empirical findings. At the same time, content should be adapted to the technological constraints of mobile devices (e.g. screen size, keyboard, etc.). Based on a literature review of existing mobile language learning applications and insights from an interactionist perspective in the SLA literature, we propose a design for mobile vocabulary learning. The design is centered around the idea of providing learners with rich input (multimedia material) and opportunities for receiving input enhancement through interaction. We justify the design choices that were made and illustrate the vocabulary activity by means of some screenshots of a prototype model.

In recent years, there has been a renewed interest in the use of digital games for purposes other than pure entertainment, including (formal) training and education. This paper examines the cognitive benefits of the genre of digital role-playing games for foreign language learning, with specific attention to the role of focus-on-form approaches and language-directed feedback. Suggestions for research are made, with particular reference to how formal properties and learner perceptions of feedback may mediate learning outcomes as well as gameplay experience.