Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Developing Adaptive Serious Games for Children With Specific Learning Difficulties: A Two-phase Usability and Technology Acceptance Study

Developing Adaptive Serious Games for Children With Specific Learning Difficulties: A Two-phase... Background: Specific learning difficulties (SpLD) include several disorders such as dyslexia, dyscalculia, and dysgraphia, and the children with these SpLD receive special education. However, the studies and the educational material so far focus mainly on one specific disorder. Objective: This study’s primary goal is to develop comprehensive training material for different types of SpLD, with five serious games addressing different aspects of the SpLD. The second focus is measuring the impact of adaptive difficulty level adjustment in the children’s and their educators’ usability and technology acceptance perception. Receiving feedback from the children and their educators, and refining the games according to their suggestions have also been essential in this two-phase study. Methods: A total of 10 SpLD educators and 23 children with different types of SpLD tested the prototypes of the five serious games (ie, Word game, Memory game, Category game, Space game, and Math game), gave detailed feedback, answered the System Usability Scale and Technology Acceptance Model (TAM) questionnaires, and applied think-aloud protocols during game play. Results: The games’ standard and adaptive versions were analyzed in terms of average playtime and the number of false answers. Detailed analyses of the interviews, with word clouds and player performances, were also provided. The TAM questionnaires’ average and mean values and box plots of each data acquisition session for the children and the educators were also reported via System Usability Scale and TAM questionnaires. The TAM results of the educators had an average of 8.41 (SD 0.87) out of 10 in the first interview and an average of 8.71 (SD 0.64) out of 10 in the second interview. The children had an average of 9.07 (SD 0.56) out of 10 in the first interview. Conclusions: Both the educators and the children with SpLD enjoyed playing the games, gave positive feedback, and suggested new ways for improvement. The results showed that these games provide thorough training material for different types of SpLD with personalized and tailored difficulty systems. The final version of the proposed games will become a part of the special education centers’ supplementary curriculum and training materials, making new enhancements and improvements possible in the future. (JMIR Serious Games 2021;9(2):e25997) doi: 10.2196/25997 KEYWORDS serious games; adaptive games; specific learning difficulty; usability; system usability scale; technology acceptance model; training; development; adaptation; gaming; learning disability; children; education; teacher https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 1 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer included six mini-games, was created to diagnose dyslexia in Introduction children aged 6 to 7 years. Each of the games in the app tested a different ability such as word forming, syllabic memory, verbal Special education is a form of education that is customized to work memory, auditory memory, and word reading. Teachers the needs of children who have a disability, illness, or difficulty. found the app helpful in identifying children with potential In the United States, special education is offered in public dyslexia and correcting exercises. Nonetheless, the app was schools, and this right is secured by the Individuals with language-dependent, and children should have been able to read Disabilities Education Act (IDEA) [1]. IDEA categorizes and write. Another app [6] for screening dyslexia in children disabilities and conditions demanding special education into 13 aged 7 to 12 years was also developed to solve the language types, including autism, attention-deficit/hyperactivity disorder, dependency. There are also findings in the SpLD literature for and specific learning difficulty (SpLD). A specific learning dyscalculia. A serious virtual reality game was created to help impairment (also known as a specific learning disability or children with dyscalculia learn math [7]. Three separate SpLD) is a condition in which children learn specific concepts scenarios and game plays were included in the game, each with at a slower rate than would be expected for their age or three different difficulty levels. The participants in the study educational level. Learning disability (or learning disorder; LD) were 40 students aged 7 to 9 years. The suggested game was is not associated with SpLD, although it is a medical term used played by half of the children at random, while the other half, for diagnosis and is often referred to as LD. To avoid any the control group, was handled with the conventional Domino misunderstanding, the word SpLD will be used in this study to system. Children who played the proposed game took refer to specific learning difficulties. SpLD is not linked to substantially less time to fit simple mathematical operations to intelligence, which means that children with SpLD may have the correct answer when compared with the control group. average or superior intelligence but learn slower than their peers. SpLD is divided into five groups by the Learning Disabilities In a study performed in Germany [8], a mobile game–based Association of America as dyslexia (reading disability), intervention on syllable stress and literacy was developed for dyscalculia (math disability), dysgraphia (writing disability), German children with dyslexia. The study focused mainly on oral and written language disorder and specific reading the design criteria of the game-based intervention, while the comprehension deficit, and nonverbal learning disability. quantitative analysis was planned to be performed on children by using the mobile game 20 minutes a day, 5 days per week. Dyslexia is the most well-known form of SpLD, and it causes Similarly, Sood et al [9] also focused on developing games spelling mistakes, poor reading, and false reading in children. where the objectives were detecting, monitoring, and managing Children with dyscalculia have difficulties with numbers and dyslexia in young children (aged 4-18 years). The study abstract principles such as counting and basic numerical presented the details of the protocols and game design principles, operations. Misspelling, writing letters in wrong forms, and but the games were not tested with the control group yet. Flogie writing letters of various sizes in a word are all symptoms of et al [10] developed intelligent game interventions and used dysgraphia in children. Oral and written language disorder and them with 51 children with learning difficulties in the specific reading comprehension deficit affect children’s ability mainstream Slovenian education system. The results showed to articulate and comprehend sentences. Dyspraxia can show that intelligent game interventions provide personalized itself in various behaviors, from a lack of coordination between education and can be helpful while designing the specialized the lips and tongue to problems with handwriting. There is no curriculum. agreement on the types of SpLD, and the coverage of SpLD changes in every country. Since the proposed games in this This study is adapted from author OY’s Master’s thesis [11], study are intended for the Turkish training system and given supervised by author ES. This study concentrates on serious that these SpLD types are accepted as SpLD in Turkey [2] and games’ usability and technology acceptance results in the are suggested by educators in special education centers with training of children with SpLD. As previously mentioned, whom the authors collaborated, they are accepted as SpLD types serious games, or games with more than just entertainment value in this paper. In Turkey, 41,600 people have been diagnosed [12,13], have recently become an important research topic with SpLD, and 82% of dyslexic students drop out of school among researchers interested in children’s education with SpLD. before attending university. To date, the only verified therapy To the best of the authors’ knowledge, this is the first systematic for SpLD is special education. In addition to the regular school research covering all types of SpLD and measuring the impact education, children with SpLD attend extra classes taught by of using adaptive difficulty via quantitative and qualitative special education teachers in groups or individually. This special approaches. This research’s primary aim is to create a collection education can include reading, writing, algebra, developing of specially developed games for use in special education for motor coordination skills, and language education activities. children with any form of SpLD. Usability tests were used to investigate the potential of this series of games. Another aim is The impact of games on children’s training with SpLD has been to demonstrate how the adaptive difficulty system affects discovered recently, and the number of studies on this topic has children’s playing experience. An adaptive difficulty framework increased. Investigating the influence of daily video game was developed to assess its impact on children with SpLD. To playing on children with SpLD was the initial research focus that end, five serious games for children with SpLD were in SpLD research [3]. In a study [4], researchers collaborated designed and developed. Two of the five games were improved with 20 children with dyslexia to see if video games, Rayman with an adaptive difficulty system to increase the children’s Raving Rabbids, a promotional Wii video game, influenced experience when playing the games. Children with SpLD and their reading abilities. In another study [5], an app, which https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 2 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer their educators evaluated all five games, and their responses to multitasking such as picking up stars in the sky while solving questionnaires—System Usability Scale (SUS) [14] and multiplication problems on the ground. Technology Acceptance Model (TAM) [15]—and their Given that children with SpLD will play the games, the themes comments on the games were classified and analyzed. The of the games were decided cautiously. Games do not focus on results demonstrated the games’ usability as enhanced training a single type of SpLD but cover various types of SpLD. The material in special education. Both the children with SpLD and first game’s theme was the spelling for children with dyslexia their educators gave highly positive feedback regarding playing or dysgraphia. Simple mathematical operations were the second the established games, which was also repeated on the responses game’s theme for children with dyscalculia. The third game’s to the SUS and TAM questionnaires. This study was split into theme was writing examples of specific categories developed two parts to implement the children’s and educators’ for children with dysgraphia or dyslexia. Two more themes recommendations into the framework and thoroughly analyze were added for different purposes (ie, a memory game targeted the outcomes. at improving visual memory and a space game requiring time management and constant concentration). When choosing Methods themes, the educators showed what kind of practices they do in their special education and suggested game ideas, so the Overview opinions of educators were also taken into consideration during For this study, five different serious games were designed and the game design. The children had only 30 minutes per week, developed to train children with any type of SpLD. The details and they had to play all the games and answer the study of the games and their versions are explained in detail in the questions in 2 weeks, so the number of themes was limited due following subsections. The games were developed in Unity to the time limitations of the children. (version 2018.3.0f2), a cross-platform game engine that allows Developed Games its users to build games in 2D and 3D environments. For this study, an ASUS Zenpad 8 tablet is used during the data During this study, different versions of the five games were acquisition so that Unity’s Android device settings were used developed. The first version of the games were developed as during the game development procedure. After the game prototypes, but all the themes and rules were already development process, the games were tested twice by two well-defined. Educators examined the games’ prototype versions participant groups, including 10 educators and 23 students with so that the educators’ first interview could be considered an SpLD between the ages of 7 and 11 years. During the first initial quality check (Prototype I). Since the users requested no interview with the participant groups, the SUS and TAM modifications after Prototype III, only two prototypes were questionnaires were filled by the participants besides applying developed in the Word game. The Category game and Math the think-aloud methods during game play and answering game had three prototypes in total for the same reason (ie, no open-ended questions after playing the games. After the first updates were requested by the users). Four different prototypes interviews, a rule-based adaptive difficulty enhancement was were developed for the Memory game and Space game. added to two of the five games to measure the impact of Word Game: Prototype I adaptiveness in the developed serious games. During the second interview with the children, only open-ended questions and The word game (Figure 1) is focusing on spelling for children think-aloud methods were used. For the educators, the same with dyslexia or dysgraphia. In the game, a child character asks interview procedure was used as before. Multimedia Appendix 10 questions about a single predefined topic, and the answers 1 summarizes the prototyping and interview procedures of the are limited to 4 to 6 letters in Turkish. The player needs to drag study. and drop the letters to the appropriate yellow squares, representing a letter in the answer. If the letter is correct, it Design Criteria and Themes of the Games replaces the yellow square, and if it is a false letter, the letter At first, with special educators’ help, design criteria were does not fill the yellow square and returns to its initial position defined to develop games compatible with the children’s (bottom left corner). After entering all the correct letters, the educational background. The first criterion is about the games’ child avatar confirms “It is correct!” and asks the next question. target groups, including children with SpLD who know reading The game’s background image is chosen to keep the players and writing. Therefore, children’s age was limited to 11 years focused on the topic. For instance, “Kitchen” was the topic of since designing a decent game for both new readers and the first designed level, and the players were asked to answer secondary school students could be an unfeasible goal. The questions about cutlery or fruits, such as “What is the name of second criterion is that games should be easy-to-use given the the sweet fruit which has red, yellow, and green colors?” and children’s struggle with SpLD. Thus, games do not include “What do you use to eat your soup?” The player gets 10 points for each correct answer. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 3 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 1. Screenshot of the Word game from Prototype I. are two panels; on the left side of the screen, there are four Word Game: Prototype II colors (or four arrows each faced in another direction) and a A skip button was added to the screen so that unsolved questions counter. Nine boxes are aligned as a three-by-three grid on the could be passed. The number of correct answers, the number right side of the screen. When the game starts, initial colors of false letters, the number of skipped questions, and the total appear on three of the nine boxes randomly, and the counter time were recorded in the database. starts to go down. After the counter hits zero, from five, colors on the right side disappear. The player’s goal is to refill the Word Game: Prototype III three boxes with the correct colors as quickly as possible. The Given that the users did not request any new updates after player first needs to tap the correct color on the left panel and Prototype II, this game was not modified. tap the right panel’s accurate grid. To erase the grids, the player can tap on a filled grid. Whenever the player fills all three grids Word Game: Prototype IV correctly, “CORRECT!” feedback shows up at the counter’s Given that the users did not request any new updates after position. After 3 seconds, a new randomly generated memory Prototype II, this game was not modified. question is asked, where each correct answer is 10 points. Memory Game: Prototype I The Memory game (Figure 2) was created to improve the visual memory and attention of children with SpLD. In the game, there https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 4 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 2. Screenshot of the Memory game from Prototype I. lightning. Moreover, the appearing time can be set as 3, 4, or 5 Memory Game: Prototype II seconds before the game starts, increasing the difficulty. Thus, A hint button was added to the screen to show the colors or this game has four versions (colors, arrows, letters, and pictures), arrows once again to the players. Additionally, a timer was five difficulty options—one grid, two grids, three grids, four added to display the countdown. Both the difficulty and timer grids, and adaptive (two grids with variable appearing can be set just before starting the game by the educators or time)—with three different appearing time options (3, 4, and 5 parents. The difficulty setting affects the number of grids to be seconds), and an adjustable timer. remembered during the game. The number of correct answers, Category Game: Prototype I number of chosen false grids, number of grids filled with wrong color or direction, difficulty level, and session time were The Category game was developed to improve the thinking and recorded in the Memory game’s database. writing skills of children with dysgraphia or dyslexia—not handwriting but correctly ordering letters in a word. During the Memory Game: Prototype III game, a couple of categories are given, and the players are After analyzing the first two prototypes, two games—the expected to find words related to that category. For example, Memory game and Space game—were updated and enhanced in Figure 3, the category is “Things that we close,” and with the adaptive difficulty level. The games’ design or predefined words are “Window,” “Door,” and “Box.” In each mechanics were not modified. The purpose of the second category, there are 15 to 20 predefined answers. If the player interview with the students was to determine their behavior and writes down one of them, it appears on the screen with a random success in games with adaptive difficulty levels. color and position, and the player gains 10 points. If the answer is wrong or misspelled, it does not appear on the screen. The Memory Game: Prototype IV player uses the keyboard of the tablet to write the words. If the To represent the alternative versions of this game, a version player gets stuck at some point, the skip button on the top right with pictures was developed (Multimedia Appendix 2). In this corner can be used, and a new category starts. There were 10 version, sky objects were used, such as sun, moon, cloud, and different categories in the first version of the game. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 5 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 3. Screenshot of the Category game from Prototype I. Category Game: Prototype IV Category Game: Prototype II Given that the users did not request any new updates after The category game was separated into two parts. In the first Prototype III, this game was not modified. part (Multimedia Appendix 3), the player must choose five correct words about a given category, between 20 available Space Game: Prototype I options. After the fifth correct answer, the category changes. In the Space game (Figure 4), the primary objective is to test When five categories are completed, the second part, which is the time management and concentration levels of children with the same as the previous Category games, starts. The player a theme that is not causally related to their courses. The player’s gives examples of the same categories in the first part. primary goal is collecting gems while trying not to hit tiny Moreover, a skip button is added to the screen. Session time spaceships that are programmed to go in a random direction. and the number of correct and false answers were saved in the The initially available time slot is 90 seconds, and the timer database. For the second part (Multimedia Appendix 4), the counts down. Whenever the player taps into the screen, a number of correct answers, number of typos, number of false spaceship rotates into that point and moves toward the point words, and session time were recorded. until it reaches the tapped position. A gem appears every 5 seconds at a random location. If the player cannot collect the Category Game: Prototype III gem in 10 seconds, the gem disappears. Every 0.4 seconds, an In this prototype (Multimedia Appendix 5), the input system of enemy spaceship spawns off-screen and flies straight through the open-ended questions was modified. The player does not the screen in a random direction. The player gains 2, 4, or 6 use the keyboard of the tablet keyboard anymore but uses an seconds randomly after collecting a gem. Whenever a player on-screen keyboard aligned in alphabetical order. collides with another spaceship, the player loses 3, 5, or 7 seconds randomly. The game is over when the time is up. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 6 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 4. Screenshot of the Space game from Prototype I. which was added to the screen’s bottom right. Additionally, the Space Game: Prototype II default enemy spawn frequency was decreased. The waiting No modification was applied in the Space game except forming time between two enemies’ spawn was changed to 0.7 seconds. a database for game statistics. The number of collected gems, Moreover, visual feedback on collecting gems and crashing into the number of crashes with the enemy, and session time were enemies was added. Players could also see how much time they adjusted to be kept in the database. gained (in blue) or lost (in red) during the game. Space Game: Prototype III Math Game: Prototype I After analyzing the first two prototypes, two games—Memory The Math game (Figure 5) is a simple addition and subtraction game and Space game— were updated and enhanced with the game. In the first version, the math question appeared below adaptive difficulty level. The games’ design or mechanics were the screen, and four kites floated up in the air. Questions are not modified. The purpose of the second interview with the created randomly with a set of selected numbers between 0 to students was to determine their behavior and success in games 15. The player’s primary goal is tapping on the kite containing with adaptive difficulty levels. the correct answer before exiting from the screen. There was no penalty for choosing wrong answers to see the tapping Space Game: Prototype IV strategy of the children. They could tap all four to see the next All the objects in the scene, except the user interface, were question and still get total points. enlarged. The control mechanism was changed with a joystick, https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 7 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 5. Screenshot of the Math game from Prototype I. difficulty of the game, as it is introduced in the Flow study of Math Game: Prototype II Czikszentmihalyi [16] and enhanced by Thomas and Young A timer was added as a countdown, and both the timer and [17] (Multimedia Appendix 6). For the adaptive difficulty, a difficulty could be set before the game starts. The difficulty rule-based method was selected due to the time constraints of level modified the kites’ speed, and a maximum limit of numbers the students. They could only play both games twice, once each for the math operations could also be arranged. The number of for the standard and adaptive versions. Since repetitive testing correct and false solutions, unanswered questions, session time, is no longer an option, methods that use training could not be and difficulty level were kept in the database. used. The rule-based difficulty method was chosen to modify the difficulty for a couple of actions instead of every action. Math Game: Prototype III These two alternatives would give the same results for large As an option, the multiplication operation was added to the sample sizes in theory; however, the rule-based method was game, and the scoring system was modified so that the player chosen to ensure that every student gets the same results for does not gain total points after their first guess. After each guess, their successful or unsuccessful set of actions. In addition, since the scoring system changes (ie, in the second guess, 8 points; there will be only one game play session per student, the in the third guess, 4 points; and a the last guess, only 1 point is adaptive difficulty should affect the game while the game is given to the player). continuing, unlike the study of Hendrix et al [18] where the players’ previous game sessions changed the game difficulty. Math Game: Prototype IV The system was designed to change only one variable to get Given that the users did not request any new updates after meaningful results from the adaptive difficulty levels. For the Prototype III, this game was not modified. Memory game, the correct answer’s display time was the Adaptive Difficulty Mechanism selected variable, while for the Space game, enemy spawn The adaptive difficulty is designed to check consecutive frequency was selected. These variables affect games directly, successful and unsuccessful moves by the player and adjust the but they cannot be noticed by the player explicitly, unlike the game difficulty according to the player’s game play level. A number of letters in the Memory game or the type of enemies simple adaptive difficulty system was developed for the Memory in the Space game, so that the students can give their honest game and Space game. The number of games was decreased to opinion about how they feel about the game difficulty without two since these games would be played by the students twice, any a priori clue. The flow of adaptive difficulty change is and the students had time constraints. Besides, these two games shown in Multimedia Appendix 7. Since the students could play had repeatability, and they were liked by children the most. This these games only for a limited time, a single change in the adaptive difficulty was added to these games to keep the players variables was adjusted by around 10%. The difference in these in the golden path or optimal game play corridor, which variables could affect the outcome quickly after the game gets prevents the players from getting bored or frustrated by the easier or more difficult. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 8 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer counts faster or slower according to the success of the players Memory Game With Adaptive Difficulty System in the game. The rules of the adaptive difficulty are simple, as For all the students with dyslexia or dysgraphia, a letter version they are explained in Figure 6 in pseudocode. If the player can of the Memory game was developed. The Memory game’s letter give two correct answers (C ) in a row, the appearing time CCA version (Multimedia Appendix 8) contains only four consonants: (t ) decreases by half a second. If the player provides more p, q, b, and d. These are the most commonly confused letters LA than one false solution, false grid, or incorrect letter in a row for children with dyslexia when reading and writing. Although (C is used for all), the appearing time increases by half a the Turkish alphabet does not contain the letter q, children are CFA familiar with this letter from the keyboard. The adaptive system second (A is used for given answer). Changes in the emerging modifies the first appearing time of the letters on the screen, time are applied to the next question. and the timer counts down seemingly from five. However, it Figure 6. Adaptive difficulty memory. The first interview results show that students gave an average the number of crashes (C ) is three more than the number of CE of 4.45 correct answers and 2.35 false answers for two grid collected gems (C ). CG difficulties in 60 seconds, which led to the selection of two Waiting time decreases by 0.03 seconds when the number of consecutive correct or false answers as the adaptive difficulty collected gems is two more than the number of crashes. Both changes condition. According to the results, the expected counters reset when the spawning frequency changes (E is average difficulty change was more than two for the increment CE used for the crashing enemy event, and E is used for the and more than one for the decrement during 100 seconds of CG playtime. Due to these initial outcomes, the amount of the collect gem event). According to the first interview results, change was set as 0.5 seconds so that neither students can realize students collected gems an average of 0.997 times. They crashed the difference directly nor face substantial difficulty changes. the enemies an average of 1.317 times per 10 seconds, and their The number of correct and false answers before a change in average playtime was 94.06 seconds. Due to these results, the appearing time and the number of changes in appearing time number of events (crashing enemies or collecting gems) is stored were added to the database. as one variable. Instead of the consecutive events, to change the difficulty, the total number of crashes or collected gems is Space Game With Adaptive Difficulty System expected to be more than the other. To decrease the enemy Adaptive difficulty in the Space game only affects the enemy spawn period, the number of collisions is expected to be three spaceships’ spawn frequency (Figure 7). If the player is more than collected gems instead of two since the students were successful, the frequency increases, and vice versa. The default more likely to crash into an enemy rather than collect a gem. waiting time (T ) between the spawn of two enemy spaceships ES The change amount of the enemy spawn period was selected as is 0.4 seconds. The waiting time increases by 0.03 seconds when 0.03 seconds since the Space game is a more dynamic game that requires more attention than the Memory game. The change was kept below 10% to avoid a profound difference. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 9 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 7. Adaptive difficulty space. Data Analysis Participants For both participant groups, the box plot of SUS and TAM In this study, 10 educators participated in the study—7 of them questionnaires were plotted using SPSS Statistics 25 software were special education teachers, 2 of them were occupational (IBM Corp), while the individual plots were plotted with the therapists, and 1 was a psychologist. The mean age of educators Minitab software’s trial version (Minitab, LLC). For each was 33 (SD 14.37) years, and 7 of the educators were female. interview and questionnaire, the average and SD of the positive Children participants were selected among the Albatros Special statements for each game type were calculated. Student’s game Education Center (Albatros Özel Eğitim Merkezi in Turkish), statistics, including the average play time, the number of correct which provides education for children with SpLD. The final answers, and difficulty-based changes, were also presented in student group includes 23 students in the age group of 7 to 11 detail. In addition, the Wilcoxon signed rank tests for both SUS years. Before the data acquisition started, permission of their and TAM questionnaires were calculated. This test was applied parents was taken both verbally and written. Additionally, to compare the standard and adaptive versions of the games children’s consent was taken verbally before the acquisitions, since data were not normally distributed. and the voluntary nature of the study was emphasized. Only 3 children could not complete the study since they had changed Results their school before the end of the study. The mean age of the 23 children participants was 8.6 (SD 1.13) years. Most of the Overview children were male (18/23, 80%). A total of 20 children were In this section, data acquired from the interviews are analyzed. diagnosed with the SpLD by a doctor. The rest of the participant Throughout this study, two interviews with both participant children had taken the Albatros Special Education Center’s test. groups (students and educators) were done. Answers to the Given that the special education center’s psychologists evaluated following questions were examined: them as having SpLD, they were also included in the study. • Are proposed games suitable for training purposes? Permissions and Ethics Approval • Are children with SpLD comfortable with different types This study was performed by two participant groups—students of the proposed games? and their educators. This study was not designed as a clinical • Can an adaptive difficulty system provide a better game trial but as a two-phase usability study for the Master’s thesis experience for children with SpLD? of OY, supervised by ES. The Ethical Approval of Research The Questionnaire Results of the Students was approved by the Middle East Technical University Human Subjects Ethics Committee (ID of the permission: There are 10 questions in the SUS, where half of the questions 28620816/398). The participants or their parents, if they were consist of positive statements. The TAM questionnaire includes children, were informed both with verbal and written 19 questions about games to understand participants’ acceptance communication about the purpose of the study, possible benefits level of the proposed games (Table S1 in Multimedia Appendix of the study to children with SpLD, and the voluntary nature of 9). The playtime of the games was recorded, and the average participation. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 10 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer playing times are shown in Table S2 in Multimedia Appendix most common five words were colorful, ship, car, control, and 9. enemy. The Comments and Suggestions of the Students Game Statistics of the Students The comments of children (Multimedia Appendix 10) were In this section, statistics of the games are shown. Tables S3 and examined in three sections: comments on current games, future S4 in Multimedia Appendix 9 show the data of students in the game ideas, and general feedback. For the first part, children Memory game from the first interview. Number of false grids were asked to comment on existing game ideas, the difficulty refers to when the player fills a grid that is supposed to be level of games, visuals, and all related things to the games. Most empty. Number of false color/arrow markings refers to cases children said that the Math game was too simple in terms of when the player marks a grid with a wrong color or arrow visualization and the question types. They all said that there option. The difficulty of the Memory game changes with the should be other operations (multiplication and division), and number of grids since they should be memorized to be filled. there should be more themes such as planes and cars besides Students played this game three times: the first color version kite themes. Multiplication was added to the final version of with low difficulty to get used to the game, then again the color the game. Most of the students complained about the number version with greater difficulty, and finally the arrow version. of enemies and their small sizes in the Space game, which were The level of difficulty was set according to the age of the fixed in the final prototype. Most of the students wanted to play students. the Space game level by level rather than in an endless fashion. Tables S5-S7 in Multimedia Appendix 9 show the students’ In addition, they said that they wanted to see “boss creatures” game statistics from the first interview of the Space game, Math at the end of each level. Moreover, most children suggested game, and Word game, respectively. alternative control mechanisms for the Space game, such as buttons, touch joystick, or dragging. After these suggestions, a Statistics of both parts of the Category game are shown in Tables touch joystick was added in the last version. There were not S8 and S9 in Multimedia Appendix 9. For part two, an irrelevant many common comments on the Word game. Finally, students answer to the given category is counted as False Answer, while were asked if they had any other comments in general. Three misspelled words are counted as Typo. Table S10 in Multimedia of them suggested that music or audial feedback in the game Appendix 9 shows the students’ data in the Memory game from will make it more fun. Most of the children proposed an in-game the second interview, where they played each version of the achievement system. Finally, one of the children suggested a game once. The Space game statistics from the second interview ranking system so that players can understand if they won with the students are shown in Table S11 in Multimedia something or not. Multimedia Appendix 11 shows the most Appendix 9. common words or phrases in the comments of students. The Table 1 shows how often the adaptive difficulty system adjusts the difficulty while the students play the games. Table 1. Adaptive difficulty changes during game: mean and SD results of in-game adaptive difficulty changes for both games in the second session with students. Game Difficulty up, mean (SD) Difficulty down, mean (SD) Memory game 2.58 (1.41) 1.25 (0.97) Space game 1.52 (1.20) 1.56 (0.98) Wilcoxon signed rank test results can be seen in Table 2. This a z score of 1.96. The asymptotic significance column shows test was applied to compare the games’ standard and adaptive the significance of the difference. A value of less than 0.05 is versions since data were not normally distributed. Column Z considered significant. Figures 8 and 9 show the box plots of refers to the z score, which is closer to 0 when the groups are playing time for the Space game and Memory game, evenly distributed. A P value of .05 is approximately equal to respectively. Table 2. Wilcoxon signed rank test: results of test between data of two versions of both games from the second session with students. Game Z score Asymptotic significance (two-tailed) Memory game: correct answers –0.567 0.571 Memory game: false grid –0.138 0.890 Memory game: false letter –0.536 0.592 Space game: crashed gem –2.173 0.030 Space game: crashed enemy –0.924 0.355 Space game: playtime –1.884 0.060 https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 11 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 8. Box plot of playing time for the Space game from the second session with students. Figure 9. Box plot of correct answers for the Memory game from the second session with students. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 12 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figures 10 and 11 show the box plots of the false answers of The Questionnaire Results of the Educators the Memory game and the difference between the correct and The TAM questionnaire results of educators from both false answers, respectively. Figures 12 and 13 show the box interviews are displayed in Table S12 in Multimedia Appendix plots of the SUS and TAM questionnaires, while Multimedia 9. Table S13 in Multimedia Appendix 9 shows the Wilcoxon Appendix 12 and 13 display the individual plots of the SUS and signed rank test results of educators for both questionnaires. TAM questionnaires. Figure 10. Box plot of false answers for the Memory game from the second session with students. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 13 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 11. Box plot of the difference between the correct and false answers for the Memory game from the second session with students. Figure 12. Box plot of SUS scores of the educators from the first (October) and second (March) sessions. SUS: System Usability Scale. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 14 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 13. Box plot of TAM questionnaire answers of educators from the first (October) and second (March) sessions. TAM: Technology Acceptance Model. classroom, as homework or as auxiliary material, only one The Comments and Suggestions of the Educators educator said she would not use the games until the parents The educators’ feedback (Multimedia Appendix 14) was also were sufficiently aware of this system and its benefits. analyzed under prototyped game ideas, potential game ideas, Multimedia Appendix 15 shows the most common words or and general feedback. For the first segment, educators were phrases in the comments used by the educators. The most asked to comment on current games, game difficulty levels, common five words were feedback, audial feedback, graphics, and game-related features. Instead of words, one of well-thought, visual elements, and reward. The only negative the educators proposed that game pictures could be used for the feedback was “Will not use,” which was said due to the first part of the category. Another educator suggested that visuals educator’s concern about parents’ unfamiliarity with tablet use should enrich the Math game without creating distractions. in special education. Other teachers suggested that creating a shape by collecting a gem in the Space game would support children’s sketching Discussion ability. She also said that a star or a diamond should be gained instead of points since these symbols correspond to the special Principal Findings education center’s current motivation system. In this study, five different serious games and their prototypes were developed to provide training material for children with Educators were also asked if there was any game they want to any type of SpLD. The proposed games were evaluated in a see in this set of games. One educator suggested a game that two-phase study consisting of the SUS and TAM questionnaires, enables the players to write letters by tracing a line or an area and open-ended questions for 23 children with SpLD and 10 with fingers. Another educator said that a game based on finding educators. The main findings of this study showcase the synonyms or antonyms would be helpful to children with potentials of using serious games in SpLD training, the impact dyslexia. Rhythmic counting is indicated as another possible of adding adaptive difficulty systems that enable personalized addition to the Math game. A “Spot 7 differences” type of puzzle training experiences, and the outcomes of player-oriented design and jigsaw puzzles were also suggested. Finally, an educator where each prototype was structured based on the player said that a game to improve players’ auditory memory would statistics and suggestions. be beneficial. Three educators said the lack of feedback makes it difficult to feel like they are making progress. Two educators Results on comments and technology acceptance of the students believed that the degrees of difficulty should be varied. Two (Table S1 in Multimedia Appendix 9) show that children other educators indicated that holding scores on a leaderboard evaluated the games as easy-to-use and easy-to-learn. list and displaying previous scores would be helpful and Easy-to-use was a design criterion, and the results demonstrate encouraging to children. One of the educators indicated that a that it was accomplished. Since the game was designed with a guide for educators or parents would be helpful. Finally, single control mechanism and a single goal, children did not although most educators favored using this method in the https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 15 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer feel uncomfortable or overwhelmed, even when the game itself to play these games was 3.6 per week. This average can be was challenging (like the Space game). Children with SpLD accepted as high since the students’ permission to play games aged 7 to 11 years were comfortable with different input types was limited. Almost half of the kids indicated that they were (dragging, tapping, and joystick) when using only one of them allowed to play games only during the weekend or every other during the game. The single goal also mattered because most day, and they said that they would play these games in that of these children did not experience more complex games before allocated limited time. these interventions (Multimedia Appendix 16). To habituate A two-phase SUS and TAM evaluation was provided in this the children with SpLD to serious games, not only difficulty study, although the SUS scores between initial and final versions levels but also the complexity of games were increased in a of the games were demonstrating similar adjective ratings (ie, stepwise fashion. best imaginable, in our case) as mentioned by Bangor et al [20]. The results of the questionnaires showed that the outcomes of However, it was seen that in the second interview, the average the feedback-related questions were relatively low. There was of the SUS was higher, and the SD was lower, which can be no audio feedback in the proposed games, and visual feedback considered to demonstrate the condensed results regarding the was reduced to prevent any possible distraction. After this result, outcome of the games, where the hesitations and fluctuations the final version of the games was modified to have slightly from the data were less than the initial versions. more visual feedback. The questionnaire results of the educators A rule-based adaptive difficulty system was also proposed in supported these initial findings. Educators remarked that the serious games that adjust difficulty during the game, in contrast games were easy to learn and play, but there was a lack of to other studies in the literature that used a system that changes feedback in the final version of the games, although the games difficulty between two play sessions [19] and used predefined included several visual feedbacks. The children’s average points levels, which is limited [21]. The Space game’s adaptive version indicated that verbal games (the Word and Category games) did not change the results considerably for the students. had the lowest scores (8.36 and 8.08 out of 10, respectively), Moreover, the game’s speed change was practically equal, which which may stem from the duration of the games. Both the Word means the users nearly had the same statistics at the same and Category games took more than four times longer than other difficulty level. It can be interpreted in a couple of ways. The games during game play. Since playtime did not include any first possible reason for this result can be the adaptive rules or level mechanism in these games, it became boring for children effects not being well-adjusted, which is too difficult to achieve. after a while. In addition, children gave a correct answer in 54.8 In that case, the players will usually play in normal difficulty. seconds (Word game) and 38 seconds (the second part of the If adaptive changes affect the game difficulty, too many players Category game), which can be accepted as exceedingly long will probably increase or decrease their pace to the standard time durations for students with dyslexia. The point difference difficulty level. The other possible reason for this result can be between the Category and Word games is based on the visual the game difficulty being too balanced. The game could have differences. The Word game has a child avatar and a kitchen such a difficulty that players can collect the gem and crash into illustration, which is a more child-oriented visual theme. The enemies at almost the same rate. However, this is not a desired Word game, which only includes a blackboard and some outcome since children should encounter this situation after a writings, was liked by the children, and they completed the successful game period to feel their progress. The Wilcoxon game in their given time period. This result conflicts with test for the Memory game results showed that children had Shabbir et al’s [19] rules of serious games for children with better performance in the game’s adaptive version, but it was dyslexia. Shabbir et al [19] claim that not using a plain not statistically significant. Since the adaptive version increases background can cause a distraction for children. However, the its difficulty while players show good performance, children children’s questionnaires and comments show that game-related played a more challenging version and showed better static background images or avatars increased the immersion. performance, which means that the flow strategy worked. Due The kitchen image in the background or the child avatar that to the increased difficulty level, children’s performance could asks questions was not mandatory for the Word game, but not be statistically different but was still better. children stated that they loved both versions. Some of the children suggested a similar visual design for the Category Especially for adaptive games, more tests should be done for game. different variables. This study includes the modification of only one variable—enemy spawn frequency for the Space game and The average points by both participant groups were between question screen time for the Memory game—which gives 7.9 and 8.84 out of 10. When these results were combined with promising results for the future, especially in the Memory game. the questionnaire results, it can be easily said that both Children did not realize they were playing faster, and they participant groups liked all games, and they can accept all parts showed slightly better performance in adaptive versions. More of the system as a training source. Unlike other studies on tests can be applied by changing grid numbers or changing the serious games for children with SpLD, this study proposes a predefined letters in the left panel. During the test of the adaptive collection of games to educate children with any type of SpLD. games, only 1 student said that he understood the change in the All of the games can be played by children regardless of their Memory game and explained the correct reason, while others SpLD types. A specially designed set of games for these children could not explain the change they felt in both games, misjudged can provide a complete training source and better experience them, or did not feel any change at all. Many students showed rather than only playing a game directly related to their disorder that the games influenced them by expressing sadness, joy, or or difficulty. The average number of days when students want surprise during the games—mainly in the Space game. Except https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 16 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer for 1 student, the students were eager to replay the games, and children with SpLD. Both educators and students participated they demanded to continue to play after the interventions were in the game-based interventions, and their feedback was complete. In the first interview, 1 student stated that playing recorded. Both participant groups were enthusiastic about the the Space game for the second time would be unnecessary, so proposed “all-in-one” scheme. Besides, different design she did not want to replay the game. The same student did not requirements that have been listed in previous studies were make such a request in other games or during the second investigated with an additional adaptive difficulty level. These interview. The students wanted different visual themes in the adjustments and personalized approaches increased children Math game, showing that the game could be diversified visually, with dyslexia’s immersion in the games. One of the study’s not mechanically. The “boss creatures” concept was the most contributions was the use of serious games to target various highlighted suggestion of the students, which was proposed as SpLD types. Each child participant had more than one form of an improvement several times. It may be good to add them as SpLD, and playing games that focus on dyslexia or dyscalculia positive feedback in the games that are designed in levels instead has been beneficial to them instead of playing games that focus of endless games. Besides, although the game ends due to the on only one of these difficulties. player’s mistake in endless games, in level-based games, the Furthermore, playing games like the Memory and Space games player’s success leads to a new stage, which changes the feeling and education-oriented question-and-answer games resulted in at the end of the game. Thus, boss creatures can be good for increased positive feedback and excitement. Furthermore, the strengthening the feeling at the end of the episode. Students’ students mentioned that they preferred to play in a general comments on serious games were usually game-related system rather than a system focusing only on an area where suggestions, and they were not interested in the games’ they already had difficulties. The teachers’ overall impressions educational purpose. This is because children did not see these were highly positive, and most of them intend to use these games games as educational material, but they accepted these games as part of their curriculum’s supplementary material. This as standard games, which can be considered in line with the integration will enable the games to be tested for a longer period fundamental purpose of the serious game concept. so that new enhancements can be added to the proposed games with additional evaluations to measure their educational impact. Conclusion This study proposed and tested five serious games, each having a different number of prototypes, specifically developed for Acknowledgments The authors thank Dislev Türkiye Disleksili Çocuklar Vakfı (Dislev Turkish Association of Children with Dyslexia) for their tremendous help during data acquisition. The authors also thank the children and their educators for participating in this study and giving valuable and detailed feedback. Conflicts of Interest None declared. Multimedia Appendix 1 The workflow diagram of this study. [PNG File , 29 KB-Multimedia Appendix 1] Multimedia Appendix 2 Screenshot of the picture version of the Memory game from Prototype IV. [PNG File , 83 KB-Multimedia Appendix 2] Multimedia Appendix 3 Screenshot of the first part of the Category game from Prototype II. [PNG File , 564 KB-Multimedia Appendix 3] Multimedia Appendix 4 Screenshot of the second part of the Category game from Prototype II. [PNG File , 439 KB-Multimedia Appendix 4] Multimedia Appendix 5 Screenshot of the second part of the Category game with an alternative keyboard layout from Prototype III for the theme “Clothes". [PNG File , 477 KB-Multimedia Appendix 5] https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 17 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Multimedia Appendix 6 The optimal game play corridor and the golden path figure are adapted from Thomas 2010. [PNG File , 109 KB-Multimedia Appendix 6] Multimedia Appendix 7 Visual diagram of the adaptive difficulty system. [PNG File , 33 KB-Multimedia Appendix 7] Multimedia Appendix 8 Screenshot of the letter version of the Memory game from Prototype III. [PNG File , 123 KB-Multimedia Appendix 8] Multimedia Appendix 9 Supplementary files and tables on game statistics and results of the questionnaires. [DOCX File , 20 KB-Multimedia Appendix 9] Multimedia Appendix 10 Student feedback illustration map. [PNG File , 53 KB-Multimedia Appendix 10] Multimedia Appendix 11 Student feedback word cloud. [PNG File , 112 KB-Multimedia Appendix 11] Multimedia Appendix 12 Individual plot of System Usability Scale scores of educators from the first (October) and second (March) sessions. [PNG File , 13 KB-Multimedia Appendix 12] Multimedia Appendix 13 Individual plot of Technology Acceptance Model questionnaire answers of educators from the first (October) and second (March) sessions. [PNG File , 14 KB-Multimedia Appendix 13] Multimedia Appendix 14 Educator feedback illustration map. [PNG File , 47 KB-Multimedia Appendix 14] Multimedia Appendix 15 Educator feedback word cloud. [PNG File , 96 KB-Multimedia Appendix 15] Multimedia Appendix 16 Photo of a child during the intervention. [PNG File , 313 KB-Multimedia Appendix 16] References 1. Lipkin PH, Okamoto J, Council on Children with Disabilities, Council on School Health. The Individuals With Disabilities Education Act (IDEA) for children with special educational needs. Pediatrics 2015 Dec;136(6):e1650-e1662. [doi: 10.1542/peds.2015-3409] [Medline: 26620061] 2. Balcı E. Disleksi hakkında öğretmen görüşleri ve karşılaştıkları sorunlar. Ege Eğitim Dergisi 2019;20(1):162-179. [doi: 10.12984/egeefd.453922] https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 18 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer 3. Bellotti F, Ott M, Arnab S, Berta R, de Freitas S, Kiili K, et al. Designing serious games for education: from pedagogical principles to game mechanisms. 2011 Oct 1 Presented at: The 5th European Conference on Games Based Learning; 2011; Athens, Greece. 4. Franceschini S, Gori S, Ruffino M, Viola S, Molteni M, Facoetti A. Action video games make dyslexic children read better. Curr Biol 2013 Mar 18;23(6):462-466 [FREE Full text] [doi: 10.1016/j.cub.2013.01.044] [Medline: 23453956] 5. Bartolomé NA, Zorrilla AM, Zapirain BG. Dyslexia diagnosis in reading stage though the use of games at school. 2012 Jul 30 Presented at: 17th International Conference on Computer Games; July 30-August 1, 2012; Louisville, KY. [doi: 10.1109/cgames.2012.6314544] 6. Rauschenberger M, Rello L, Baeza-Yates R, Bigham JP. Towards language independent detection of dyslexia with a web-based game. In: Proceedings of the 15th International Web for All Conference. 2018 Apr 01 Presented at: W4A '18; April 2018; Lyon, France p. 1-10. [doi: 10.1145/3192714.3192816] 7. Avila-Pesantez DF, Vaca-Cardenas LA, Avila RD, Padilla NP, Rivera LA. Design of an augmented reality serious game for children with dyscalculia: a case study. In: Botto-Tobar M, Pizarro G, Zúñiga-Prieto M, D'Armas M, Sánchez MZ, editors. Technology Trends: 4th International Conference, CITT 2018, Babahoyo, Ecuador, August 29–31, 2018, Revised Selected Papers. Cham: Springer; 2019:165-175. 8. Holz H, Brandelik K, Beuttler B, Brandelik J, Ninaus M. How to train your syllable stress awareness - a digital game-based intervention for German dyslexic children. Int J Serious Games 2018 Sep 25;5(3):37-59. [doi: 10.17083/ijsg.v5i3.242] 9. Sood M, Toornstra A, Sereno M, Boland M, Filaretti D, Sood A. A digital app to aid detection, monitoring, and management of dyslexia in young children (DIMMAND): protocol for a digital health and education solution. JMIR Res Protoc 2018 May 17;7(5):e135 [FREE Full text] [doi: 10.2196/resprot.9583] [Medline: 29773528] 10. Flogie A, Aberšek B, Kordigel Aberšek M, Sik Lanyi C, Pesek I. Development and evaluation of intelligent serious games for children with learning difficulties: observational study. JMIR Serious Games 2020 Apr 16;8(2):e13190 [FREE Full text] [doi: 10.2196/13190] [Medline: 32297864] 11. Yıldırım O. Development of adaptive serious games for the training of children with specific learning disabilities. OpenMETU. 2020 Sep 10. URL: https://open.metu.edu.tr/handle/11511/69207 [accessed 2021-05-27] 12. Tobler-Ammann BC, Surer E, de Bruin ED, Rabuffetti M, Borghese NA, Mainetti R, et al. Exergames encouraging exploration of hemineglected space in stroke patients with visuospatial neglect: a feasibility study. JMIR Serious Games 2017 Aug 25;5(3):e17 [FREE Full text] [doi: 10.2196/games.7923] [Medline: 28842388] 13. Tobler-Ammann BC, Surer E, Knols RH, Borghese NA, de Bruin ED. User perspectives on exergames designed to explore the hemineglected space for stroke patients with visuospatial neglect: usability study. JMIR Serious Games 2017 Aug 25;5(3):e18 [FREE Full text] [doi: 10.2196/games.8013] [Medline: 28842390] 14. Brooke J. SUS: a “quick and dirty” usability scale. In: Jordan PW, Thomas B, McClelland IL, Weerdmeester B, editors. Usability Evaluation In Industry. United Kingdom: Taylor & Francis; 1996. 15. Davis, Jr FD. A technology acceptance model for empirically testing new end-user information systems: theory and results. DSpace@MIT. 1985. URL: https://dspace.mit.edu/bitstream/handle/1721.1/15192/14927137-MIT.pdf [accessed 2021-05-27] 16. Czikszentmihalyi M. Flow: The Psychology of Optimal Experience. New York, NY: Harper & Row; 1990. 17. Thomas JM, Young RM. Annie: automated generation of adaptive learner guidance for fun serious games. IEEE Trans Learning Technologies 2010 Oct;3(4):329-343. [doi: 10.1109/tlt.2010.32] 18. Hendrix M, Bellamy-Wood T, McKay S, Bloom V, Dunwell I. Implementing adaptive game difficulty balancing in serious games. IEEE Trans Games 2019 Dec;11(4):320-327. [doi: 10.1109/tg.2018.2791019] 19. Shabbir N, Bhatti Z, Hakro DN. Serious game user interface design rules for dyslexic children. 2019 Dec 14 Presented at: 13th International Conference on Mathematics, Actuarial Science, Computer Science and Statistics; December 14-15, 2019; Karachi, Pakistan p. 1-6. [doi: 10.1109/macs48846.2019.9024786] 20. Bangor A, Kortum P, Miller J. Determining what individual SUS scores mean: adding an adjective rating scale. J Usability Stud 2009 May 01:114-123. 21. Cuschieri T, Khaled R, Farrugia VE, Martinez HP, Yannakakis GN. The iLearnRW game: support for students with dyslexia in class and at home. 2014 Presented at: 6th International Conference on Games and Virtual Worlds for Serious Applications; September 9-12, 2014; Valletta, Malta p. 1-2. [doi: 10.1109/vs-games.2014.7012167] Abbreviations IDEA: Individuals with Disabilities Education Act LD: learning disorder SpLD: specific learning difficulties SUS: System Usability Scale TAM: Technology Acceptance Model https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 19 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Edited by N Zary; submitted 07.04.21; peer-reviewed by CW Ong, J Brooke; comments to author 22.04.21; revised version received 30.04.21; accepted 21.05.21; published 31.05.21 Please cite as: Yildirim O, Surer E Developing Adaptive Serious Games for Children With Specific Learning Difficulties: A Two-phase Usability and Technology Acceptance Study JMIR Serious Games 2021;9(2):e25997 URL: https://games.jmir.org/2021/2/e25997 doi: 10.2196/25997 PMID: ©Oguzcan Yildirim, Elif Surer. Originally published in JMIR Serious Games (https://games.jmir.org), 31.05.2021. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Serious Games, is properly cited. The complete bibliographic information, a link to the original publication on https://games.jmir.org, as well as this copyright and license information must be included. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 20 (page number not for citation purposes) XSL FO RenderX http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JMIR Serious Games JMIR Publications

Developing Adaptive Serious Games for Children With Specific Learning Difficulties: A Two-phase Usability and Technology Acceptance Study

JMIR Serious Games , Volume 9 (2) – May 31, 2021

Loading next page...
 
/lp/jmir-publications/developing-adaptive-serious-games-for-children-with-specific-learning-53346JyfC0

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
JMIR Publications
Copyright
Copyright © The Author(s). Licensed under Creative Commons Attribution cc-by 4.0
ISSN
2291-9279
DOI
10.2196/25997
Publisher site
See Article on Publisher Site

Abstract

Background: Specific learning difficulties (SpLD) include several disorders such as dyslexia, dyscalculia, and dysgraphia, and the children with these SpLD receive special education. However, the studies and the educational material so far focus mainly on one specific disorder. Objective: This study’s primary goal is to develop comprehensive training material for different types of SpLD, with five serious games addressing different aspects of the SpLD. The second focus is measuring the impact of adaptive difficulty level adjustment in the children’s and their educators’ usability and technology acceptance perception. Receiving feedback from the children and their educators, and refining the games according to their suggestions have also been essential in this two-phase study. Methods: A total of 10 SpLD educators and 23 children with different types of SpLD tested the prototypes of the five serious games (ie, Word game, Memory game, Category game, Space game, and Math game), gave detailed feedback, answered the System Usability Scale and Technology Acceptance Model (TAM) questionnaires, and applied think-aloud protocols during game play. Results: The games’ standard and adaptive versions were analyzed in terms of average playtime and the number of false answers. Detailed analyses of the interviews, with word clouds and player performances, were also provided. The TAM questionnaires’ average and mean values and box plots of each data acquisition session for the children and the educators were also reported via System Usability Scale and TAM questionnaires. The TAM results of the educators had an average of 8.41 (SD 0.87) out of 10 in the first interview and an average of 8.71 (SD 0.64) out of 10 in the second interview. The children had an average of 9.07 (SD 0.56) out of 10 in the first interview. Conclusions: Both the educators and the children with SpLD enjoyed playing the games, gave positive feedback, and suggested new ways for improvement. The results showed that these games provide thorough training material for different types of SpLD with personalized and tailored difficulty systems. The final version of the proposed games will become a part of the special education centers’ supplementary curriculum and training materials, making new enhancements and improvements possible in the future. (JMIR Serious Games 2021;9(2):e25997) doi: 10.2196/25997 KEYWORDS serious games; adaptive games; specific learning difficulty; usability; system usability scale; technology acceptance model; training; development; adaptation; gaming; learning disability; children; education; teacher https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 1 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer included six mini-games, was created to diagnose dyslexia in Introduction children aged 6 to 7 years. Each of the games in the app tested a different ability such as word forming, syllabic memory, verbal Special education is a form of education that is customized to work memory, auditory memory, and word reading. Teachers the needs of children who have a disability, illness, or difficulty. found the app helpful in identifying children with potential In the United States, special education is offered in public dyslexia and correcting exercises. Nonetheless, the app was schools, and this right is secured by the Individuals with language-dependent, and children should have been able to read Disabilities Education Act (IDEA) [1]. IDEA categorizes and write. Another app [6] for screening dyslexia in children disabilities and conditions demanding special education into 13 aged 7 to 12 years was also developed to solve the language types, including autism, attention-deficit/hyperactivity disorder, dependency. There are also findings in the SpLD literature for and specific learning difficulty (SpLD). A specific learning dyscalculia. A serious virtual reality game was created to help impairment (also known as a specific learning disability or children with dyscalculia learn math [7]. Three separate SpLD) is a condition in which children learn specific concepts scenarios and game plays were included in the game, each with at a slower rate than would be expected for their age or three different difficulty levels. The participants in the study educational level. Learning disability (or learning disorder; LD) were 40 students aged 7 to 9 years. The suggested game was is not associated with SpLD, although it is a medical term used played by half of the children at random, while the other half, for diagnosis and is often referred to as LD. To avoid any the control group, was handled with the conventional Domino misunderstanding, the word SpLD will be used in this study to system. Children who played the proposed game took refer to specific learning difficulties. SpLD is not linked to substantially less time to fit simple mathematical operations to intelligence, which means that children with SpLD may have the correct answer when compared with the control group. average or superior intelligence but learn slower than their peers. SpLD is divided into five groups by the Learning Disabilities In a study performed in Germany [8], a mobile game–based Association of America as dyslexia (reading disability), intervention on syllable stress and literacy was developed for dyscalculia (math disability), dysgraphia (writing disability), German children with dyslexia. The study focused mainly on oral and written language disorder and specific reading the design criteria of the game-based intervention, while the comprehension deficit, and nonverbal learning disability. quantitative analysis was planned to be performed on children by using the mobile game 20 minutes a day, 5 days per week. Dyslexia is the most well-known form of SpLD, and it causes Similarly, Sood et al [9] also focused on developing games spelling mistakes, poor reading, and false reading in children. where the objectives were detecting, monitoring, and managing Children with dyscalculia have difficulties with numbers and dyslexia in young children (aged 4-18 years). The study abstract principles such as counting and basic numerical presented the details of the protocols and game design principles, operations. Misspelling, writing letters in wrong forms, and but the games were not tested with the control group yet. Flogie writing letters of various sizes in a word are all symptoms of et al [10] developed intelligent game interventions and used dysgraphia in children. Oral and written language disorder and them with 51 children with learning difficulties in the specific reading comprehension deficit affect children’s ability mainstream Slovenian education system. The results showed to articulate and comprehend sentences. Dyspraxia can show that intelligent game interventions provide personalized itself in various behaviors, from a lack of coordination between education and can be helpful while designing the specialized the lips and tongue to problems with handwriting. There is no curriculum. agreement on the types of SpLD, and the coverage of SpLD changes in every country. Since the proposed games in this This study is adapted from author OY’s Master’s thesis [11], study are intended for the Turkish training system and given supervised by author ES. This study concentrates on serious that these SpLD types are accepted as SpLD in Turkey [2] and games’ usability and technology acceptance results in the are suggested by educators in special education centers with training of children with SpLD. As previously mentioned, whom the authors collaborated, they are accepted as SpLD types serious games, or games with more than just entertainment value in this paper. In Turkey, 41,600 people have been diagnosed [12,13], have recently become an important research topic with SpLD, and 82% of dyslexic students drop out of school among researchers interested in children’s education with SpLD. before attending university. To date, the only verified therapy To the best of the authors’ knowledge, this is the first systematic for SpLD is special education. In addition to the regular school research covering all types of SpLD and measuring the impact education, children with SpLD attend extra classes taught by of using adaptive difficulty via quantitative and qualitative special education teachers in groups or individually. This special approaches. This research’s primary aim is to create a collection education can include reading, writing, algebra, developing of specially developed games for use in special education for motor coordination skills, and language education activities. children with any form of SpLD. Usability tests were used to investigate the potential of this series of games. Another aim is The impact of games on children’s training with SpLD has been to demonstrate how the adaptive difficulty system affects discovered recently, and the number of studies on this topic has children’s playing experience. An adaptive difficulty framework increased. Investigating the influence of daily video game was developed to assess its impact on children with SpLD. To playing on children with SpLD was the initial research focus that end, five serious games for children with SpLD were in SpLD research [3]. In a study [4], researchers collaborated designed and developed. Two of the five games were improved with 20 children with dyslexia to see if video games, Rayman with an adaptive difficulty system to increase the children’s Raving Rabbids, a promotional Wii video game, influenced experience when playing the games. Children with SpLD and their reading abilities. In another study [5], an app, which https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 2 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer their educators evaluated all five games, and their responses to multitasking such as picking up stars in the sky while solving questionnaires—System Usability Scale (SUS) [14] and multiplication problems on the ground. Technology Acceptance Model (TAM) [15]—and their Given that children with SpLD will play the games, the themes comments on the games were classified and analyzed. The of the games were decided cautiously. Games do not focus on results demonstrated the games’ usability as enhanced training a single type of SpLD but cover various types of SpLD. The material in special education. Both the children with SpLD and first game’s theme was the spelling for children with dyslexia their educators gave highly positive feedback regarding playing or dysgraphia. Simple mathematical operations were the second the established games, which was also repeated on the responses game’s theme for children with dyscalculia. The third game’s to the SUS and TAM questionnaires. This study was split into theme was writing examples of specific categories developed two parts to implement the children’s and educators’ for children with dysgraphia or dyslexia. Two more themes recommendations into the framework and thoroughly analyze were added for different purposes (ie, a memory game targeted the outcomes. at improving visual memory and a space game requiring time management and constant concentration). When choosing Methods themes, the educators showed what kind of practices they do in their special education and suggested game ideas, so the Overview opinions of educators were also taken into consideration during For this study, five different serious games were designed and the game design. The children had only 30 minutes per week, developed to train children with any type of SpLD. The details and they had to play all the games and answer the study of the games and their versions are explained in detail in the questions in 2 weeks, so the number of themes was limited due following subsections. The games were developed in Unity to the time limitations of the children. (version 2018.3.0f2), a cross-platform game engine that allows Developed Games its users to build games in 2D and 3D environments. For this study, an ASUS Zenpad 8 tablet is used during the data During this study, different versions of the five games were acquisition so that Unity’s Android device settings were used developed. The first version of the games were developed as during the game development procedure. After the game prototypes, but all the themes and rules were already development process, the games were tested twice by two well-defined. Educators examined the games’ prototype versions participant groups, including 10 educators and 23 students with so that the educators’ first interview could be considered an SpLD between the ages of 7 and 11 years. During the first initial quality check (Prototype I). Since the users requested no interview with the participant groups, the SUS and TAM modifications after Prototype III, only two prototypes were questionnaires were filled by the participants besides applying developed in the Word game. The Category game and Math the think-aloud methods during game play and answering game had three prototypes in total for the same reason (ie, no open-ended questions after playing the games. After the first updates were requested by the users). Four different prototypes interviews, a rule-based adaptive difficulty enhancement was were developed for the Memory game and Space game. added to two of the five games to measure the impact of Word Game: Prototype I adaptiveness in the developed serious games. During the second interview with the children, only open-ended questions and The word game (Figure 1) is focusing on spelling for children think-aloud methods were used. For the educators, the same with dyslexia or dysgraphia. In the game, a child character asks interview procedure was used as before. Multimedia Appendix 10 questions about a single predefined topic, and the answers 1 summarizes the prototyping and interview procedures of the are limited to 4 to 6 letters in Turkish. The player needs to drag study. and drop the letters to the appropriate yellow squares, representing a letter in the answer. If the letter is correct, it Design Criteria and Themes of the Games replaces the yellow square, and if it is a false letter, the letter At first, with special educators’ help, design criteria were does not fill the yellow square and returns to its initial position defined to develop games compatible with the children’s (bottom left corner). After entering all the correct letters, the educational background. The first criterion is about the games’ child avatar confirms “It is correct!” and asks the next question. target groups, including children with SpLD who know reading The game’s background image is chosen to keep the players and writing. Therefore, children’s age was limited to 11 years focused on the topic. For instance, “Kitchen” was the topic of since designing a decent game for both new readers and the first designed level, and the players were asked to answer secondary school students could be an unfeasible goal. The questions about cutlery or fruits, such as “What is the name of second criterion is that games should be easy-to-use given the the sweet fruit which has red, yellow, and green colors?” and children’s struggle with SpLD. Thus, games do not include “What do you use to eat your soup?” The player gets 10 points for each correct answer. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 3 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 1. Screenshot of the Word game from Prototype I. are two panels; on the left side of the screen, there are four Word Game: Prototype II colors (or four arrows each faced in another direction) and a A skip button was added to the screen so that unsolved questions counter. Nine boxes are aligned as a three-by-three grid on the could be passed. The number of correct answers, the number right side of the screen. When the game starts, initial colors of false letters, the number of skipped questions, and the total appear on three of the nine boxes randomly, and the counter time were recorded in the database. starts to go down. After the counter hits zero, from five, colors on the right side disappear. The player’s goal is to refill the Word Game: Prototype III three boxes with the correct colors as quickly as possible. The Given that the users did not request any new updates after player first needs to tap the correct color on the left panel and Prototype II, this game was not modified. tap the right panel’s accurate grid. To erase the grids, the player can tap on a filled grid. Whenever the player fills all three grids Word Game: Prototype IV correctly, “CORRECT!” feedback shows up at the counter’s Given that the users did not request any new updates after position. After 3 seconds, a new randomly generated memory Prototype II, this game was not modified. question is asked, where each correct answer is 10 points. Memory Game: Prototype I The Memory game (Figure 2) was created to improve the visual memory and attention of children with SpLD. In the game, there https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 4 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 2. Screenshot of the Memory game from Prototype I. lightning. Moreover, the appearing time can be set as 3, 4, or 5 Memory Game: Prototype II seconds before the game starts, increasing the difficulty. Thus, A hint button was added to the screen to show the colors or this game has four versions (colors, arrows, letters, and pictures), arrows once again to the players. Additionally, a timer was five difficulty options—one grid, two grids, three grids, four added to display the countdown. Both the difficulty and timer grids, and adaptive (two grids with variable appearing can be set just before starting the game by the educators or time)—with three different appearing time options (3, 4, and 5 parents. The difficulty setting affects the number of grids to be seconds), and an adjustable timer. remembered during the game. The number of correct answers, Category Game: Prototype I number of chosen false grids, number of grids filled with wrong color or direction, difficulty level, and session time were The Category game was developed to improve the thinking and recorded in the Memory game’s database. writing skills of children with dysgraphia or dyslexia—not handwriting but correctly ordering letters in a word. During the Memory Game: Prototype III game, a couple of categories are given, and the players are After analyzing the first two prototypes, two games—the expected to find words related to that category. For example, Memory game and Space game—were updated and enhanced in Figure 3, the category is “Things that we close,” and with the adaptive difficulty level. The games’ design or predefined words are “Window,” “Door,” and “Box.” In each mechanics were not modified. The purpose of the second category, there are 15 to 20 predefined answers. If the player interview with the students was to determine their behavior and writes down one of them, it appears on the screen with a random success in games with adaptive difficulty levels. color and position, and the player gains 10 points. If the answer is wrong or misspelled, it does not appear on the screen. The Memory Game: Prototype IV player uses the keyboard of the tablet to write the words. If the To represent the alternative versions of this game, a version player gets stuck at some point, the skip button on the top right with pictures was developed (Multimedia Appendix 2). In this corner can be used, and a new category starts. There were 10 version, sky objects were used, such as sun, moon, cloud, and different categories in the first version of the game. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 5 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 3. Screenshot of the Category game from Prototype I. Category Game: Prototype IV Category Game: Prototype II Given that the users did not request any new updates after The category game was separated into two parts. In the first Prototype III, this game was not modified. part (Multimedia Appendix 3), the player must choose five correct words about a given category, between 20 available Space Game: Prototype I options. After the fifth correct answer, the category changes. In the Space game (Figure 4), the primary objective is to test When five categories are completed, the second part, which is the time management and concentration levels of children with the same as the previous Category games, starts. The player a theme that is not causally related to their courses. The player’s gives examples of the same categories in the first part. primary goal is collecting gems while trying not to hit tiny Moreover, a skip button is added to the screen. Session time spaceships that are programmed to go in a random direction. and the number of correct and false answers were saved in the The initially available time slot is 90 seconds, and the timer database. For the second part (Multimedia Appendix 4), the counts down. Whenever the player taps into the screen, a number of correct answers, number of typos, number of false spaceship rotates into that point and moves toward the point words, and session time were recorded. until it reaches the tapped position. A gem appears every 5 seconds at a random location. If the player cannot collect the Category Game: Prototype III gem in 10 seconds, the gem disappears. Every 0.4 seconds, an In this prototype (Multimedia Appendix 5), the input system of enemy spaceship spawns off-screen and flies straight through the open-ended questions was modified. The player does not the screen in a random direction. The player gains 2, 4, or 6 use the keyboard of the tablet keyboard anymore but uses an seconds randomly after collecting a gem. Whenever a player on-screen keyboard aligned in alphabetical order. collides with another spaceship, the player loses 3, 5, or 7 seconds randomly. The game is over when the time is up. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 6 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 4. Screenshot of the Space game from Prototype I. which was added to the screen’s bottom right. Additionally, the Space Game: Prototype II default enemy spawn frequency was decreased. The waiting No modification was applied in the Space game except forming time between two enemies’ spawn was changed to 0.7 seconds. a database for game statistics. The number of collected gems, Moreover, visual feedback on collecting gems and crashing into the number of crashes with the enemy, and session time were enemies was added. Players could also see how much time they adjusted to be kept in the database. gained (in blue) or lost (in red) during the game. Space Game: Prototype III Math Game: Prototype I After analyzing the first two prototypes, two games—Memory The Math game (Figure 5) is a simple addition and subtraction game and Space game— were updated and enhanced with the game. In the first version, the math question appeared below adaptive difficulty level. The games’ design or mechanics were the screen, and four kites floated up in the air. Questions are not modified. The purpose of the second interview with the created randomly with a set of selected numbers between 0 to students was to determine their behavior and success in games 15. The player’s primary goal is tapping on the kite containing with adaptive difficulty levels. the correct answer before exiting from the screen. There was no penalty for choosing wrong answers to see the tapping Space Game: Prototype IV strategy of the children. They could tap all four to see the next All the objects in the scene, except the user interface, were question and still get total points. enlarged. The control mechanism was changed with a joystick, https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 7 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 5. Screenshot of the Math game from Prototype I. difficulty of the game, as it is introduced in the Flow study of Math Game: Prototype II Czikszentmihalyi [16] and enhanced by Thomas and Young A timer was added as a countdown, and both the timer and [17] (Multimedia Appendix 6). For the adaptive difficulty, a difficulty could be set before the game starts. The difficulty rule-based method was selected due to the time constraints of level modified the kites’ speed, and a maximum limit of numbers the students. They could only play both games twice, once each for the math operations could also be arranged. The number of for the standard and adaptive versions. Since repetitive testing correct and false solutions, unanswered questions, session time, is no longer an option, methods that use training could not be and difficulty level were kept in the database. used. The rule-based difficulty method was chosen to modify the difficulty for a couple of actions instead of every action. Math Game: Prototype III These two alternatives would give the same results for large As an option, the multiplication operation was added to the sample sizes in theory; however, the rule-based method was game, and the scoring system was modified so that the player chosen to ensure that every student gets the same results for does not gain total points after their first guess. After each guess, their successful or unsuccessful set of actions. In addition, since the scoring system changes (ie, in the second guess, 8 points; there will be only one game play session per student, the in the third guess, 4 points; and a the last guess, only 1 point is adaptive difficulty should affect the game while the game is given to the player). continuing, unlike the study of Hendrix et al [18] where the players’ previous game sessions changed the game difficulty. Math Game: Prototype IV The system was designed to change only one variable to get Given that the users did not request any new updates after meaningful results from the adaptive difficulty levels. For the Prototype III, this game was not modified. Memory game, the correct answer’s display time was the Adaptive Difficulty Mechanism selected variable, while for the Space game, enemy spawn The adaptive difficulty is designed to check consecutive frequency was selected. These variables affect games directly, successful and unsuccessful moves by the player and adjust the but they cannot be noticed by the player explicitly, unlike the game difficulty according to the player’s game play level. A number of letters in the Memory game or the type of enemies simple adaptive difficulty system was developed for the Memory in the Space game, so that the students can give their honest game and Space game. The number of games was decreased to opinion about how they feel about the game difficulty without two since these games would be played by the students twice, any a priori clue. The flow of adaptive difficulty change is and the students had time constraints. Besides, these two games shown in Multimedia Appendix 7. Since the students could play had repeatability, and they were liked by children the most. This these games only for a limited time, a single change in the adaptive difficulty was added to these games to keep the players variables was adjusted by around 10%. The difference in these in the golden path or optimal game play corridor, which variables could affect the outcome quickly after the game gets prevents the players from getting bored or frustrated by the easier or more difficult. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 8 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer counts faster or slower according to the success of the players Memory Game With Adaptive Difficulty System in the game. The rules of the adaptive difficulty are simple, as For all the students with dyslexia or dysgraphia, a letter version they are explained in Figure 6 in pseudocode. If the player can of the Memory game was developed. The Memory game’s letter give two correct answers (C ) in a row, the appearing time CCA version (Multimedia Appendix 8) contains only four consonants: (t ) decreases by half a second. If the player provides more p, q, b, and d. These are the most commonly confused letters LA than one false solution, false grid, or incorrect letter in a row for children with dyslexia when reading and writing. Although (C is used for all), the appearing time increases by half a the Turkish alphabet does not contain the letter q, children are CFA familiar with this letter from the keyboard. The adaptive system second (A is used for given answer). Changes in the emerging modifies the first appearing time of the letters on the screen, time are applied to the next question. and the timer counts down seemingly from five. However, it Figure 6. Adaptive difficulty memory. The first interview results show that students gave an average the number of crashes (C ) is three more than the number of CE of 4.45 correct answers and 2.35 false answers for two grid collected gems (C ). CG difficulties in 60 seconds, which led to the selection of two Waiting time decreases by 0.03 seconds when the number of consecutive correct or false answers as the adaptive difficulty collected gems is two more than the number of crashes. Both changes condition. According to the results, the expected counters reset when the spawning frequency changes (E is average difficulty change was more than two for the increment CE used for the crashing enemy event, and E is used for the and more than one for the decrement during 100 seconds of CG playtime. Due to these initial outcomes, the amount of the collect gem event). According to the first interview results, change was set as 0.5 seconds so that neither students can realize students collected gems an average of 0.997 times. They crashed the difference directly nor face substantial difficulty changes. the enemies an average of 1.317 times per 10 seconds, and their The number of correct and false answers before a change in average playtime was 94.06 seconds. Due to these results, the appearing time and the number of changes in appearing time number of events (crashing enemies or collecting gems) is stored were added to the database. as one variable. Instead of the consecutive events, to change the difficulty, the total number of crashes or collected gems is Space Game With Adaptive Difficulty System expected to be more than the other. To decrease the enemy Adaptive difficulty in the Space game only affects the enemy spawn period, the number of collisions is expected to be three spaceships’ spawn frequency (Figure 7). If the player is more than collected gems instead of two since the students were successful, the frequency increases, and vice versa. The default more likely to crash into an enemy rather than collect a gem. waiting time (T ) between the spawn of two enemy spaceships ES The change amount of the enemy spawn period was selected as is 0.4 seconds. The waiting time increases by 0.03 seconds when 0.03 seconds since the Space game is a more dynamic game that requires more attention than the Memory game. The change was kept below 10% to avoid a profound difference. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 9 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 7. Adaptive difficulty space. Data Analysis Participants For both participant groups, the box plot of SUS and TAM In this study, 10 educators participated in the study—7 of them questionnaires were plotted using SPSS Statistics 25 software were special education teachers, 2 of them were occupational (IBM Corp), while the individual plots were plotted with the therapists, and 1 was a psychologist. The mean age of educators Minitab software’s trial version (Minitab, LLC). For each was 33 (SD 14.37) years, and 7 of the educators were female. interview and questionnaire, the average and SD of the positive Children participants were selected among the Albatros Special statements for each game type were calculated. Student’s game Education Center (Albatros Özel Eğitim Merkezi in Turkish), statistics, including the average play time, the number of correct which provides education for children with SpLD. The final answers, and difficulty-based changes, were also presented in student group includes 23 students in the age group of 7 to 11 detail. In addition, the Wilcoxon signed rank tests for both SUS years. Before the data acquisition started, permission of their and TAM questionnaires were calculated. This test was applied parents was taken both verbally and written. Additionally, to compare the standard and adaptive versions of the games children’s consent was taken verbally before the acquisitions, since data were not normally distributed. and the voluntary nature of the study was emphasized. Only 3 children could not complete the study since they had changed Results their school before the end of the study. The mean age of the 23 children participants was 8.6 (SD 1.13) years. Most of the Overview children were male (18/23, 80%). A total of 20 children were In this section, data acquired from the interviews are analyzed. diagnosed with the SpLD by a doctor. The rest of the participant Throughout this study, two interviews with both participant children had taken the Albatros Special Education Center’s test. groups (students and educators) were done. Answers to the Given that the special education center’s psychologists evaluated following questions were examined: them as having SpLD, they were also included in the study. • Are proposed games suitable for training purposes? Permissions and Ethics Approval • Are children with SpLD comfortable with different types This study was performed by two participant groups—students of the proposed games? and their educators. This study was not designed as a clinical • Can an adaptive difficulty system provide a better game trial but as a two-phase usability study for the Master’s thesis experience for children with SpLD? of OY, supervised by ES. The Ethical Approval of Research The Questionnaire Results of the Students was approved by the Middle East Technical University Human Subjects Ethics Committee (ID of the permission: There are 10 questions in the SUS, where half of the questions 28620816/398). The participants or their parents, if they were consist of positive statements. The TAM questionnaire includes children, were informed both with verbal and written 19 questions about games to understand participants’ acceptance communication about the purpose of the study, possible benefits level of the proposed games (Table S1 in Multimedia Appendix of the study to children with SpLD, and the voluntary nature of 9). The playtime of the games was recorded, and the average participation. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 10 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer playing times are shown in Table S2 in Multimedia Appendix most common five words were colorful, ship, car, control, and 9. enemy. The Comments and Suggestions of the Students Game Statistics of the Students The comments of children (Multimedia Appendix 10) were In this section, statistics of the games are shown. Tables S3 and examined in three sections: comments on current games, future S4 in Multimedia Appendix 9 show the data of students in the game ideas, and general feedback. For the first part, children Memory game from the first interview. Number of false grids were asked to comment on existing game ideas, the difficulty refers to when the player fills a grid that is supposed to be level of games, visuals, and all related things to the games. Most empty. Number of false color/arrow markings refers to cases children said that the Math game was too simple in terms of when the player marks a grid with a wrong color or arrow visualization and the question types. They all said that there option. The difficulty of the Memory game changes with the should be other operations (multiplication and division), and number of grids since they should be memorized to be filled. there should be more themes such as planes and cars besides Students played this game three times: the first color version kite themes. Multiplication was added to the final version of with low difficulty to get used to the game, then again the color the game. Most of the students complained about the number version with greater difficulty, and finally the arrow version. of enemies and their small sizes in the Space game, which were The level of difficulty was set according to the age of the fixed in the final prototype. Most of the students wanted to play students. the Space game level by level rather than in an endless fashion. Tables S5-S7 in Multimedia Appendix 9 show the students’ In addition, they said that they wanted to see “boss creatures” game statistics from the first interview of the Space game, Math at the end of each level. Moreover, most children suggested game, and Word game, respectively. alternative control mechanisms for the Space game, such as buttons, touch joystick, or dragging. After these suggestions, a Statistics of both parts of the Category game are shown in Tables touch joystick was added in the last version. There were not S8 and S9 in Multimedia Appendix 9. For part two, an irrelevant many common comments on the Word game. Finally, students answer to the given category is counted as False Answer, while were asked if they had any other comments in general. Three misspelled words are counted as Typo. Table S10 in Multimedia of them suggested that music or audial feedback in the game Appendix 9 shows the students’ data in the Memory game from will make it more fun. Most of the children proposed an in-game the second interview, where they played each version of the achievement system. Finally, one of the children suggested a game once. The Space game statistics from the second interview ranking system so that players can understand if they won with the students are shown in Table S11 in Multimedia something or not. Multimedia Appendix 11 shows the most Appendix 9. common words or phrases in the comments of students. The Table 1 shows how often the adaptive difficulty system adjusts the difficulty while the students play the games. Table 1. Adaptive difficulty changes during game: mean and SD results of in-game adaptive difficulty changes for both games in the second session with students. Game Difficulty up, mean (SD) Difficulty down, mean (SD) Memory game 2.58 (1.41) 1.25 (0.97) Space game 1.52 (1.20) 1.56 (0.98) Wilcoxon signed rank test results can be seen in Table 2. This a z score of 1.96. The asymptotic significance column shows test was applied to compare the games’ standard and adaptive the significance of the difference. A value of less than 0.05 is versions since data were not normally distributed. Column Z considered significant. Figures 8 and 9 show the box plots of refers to the z score, which is closer to 0 when the groups are playing time for the Space game and Memory game, evenly distributed. A P value of .05 is approximately equal to respectively. Table 2. Wilcoxon signed rank test: results of test between data of two versions of both games from the second session with students. Game Z score Asymptotic significance (two-tailed) Memory game: correct answers –0.567 0.571 Memory game: false grid –0.138 0.890 Memory game: false letter –0.536 0.592 Space game: crashed gem –2.173 0.030 Space game: crashed enemy –0.924 0.355 Space game: playtime –1.884 0.060 https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 11 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 8. Box plot of playing time for the Space game from the second session with students. Figure 9. Box plot of correct answers for the Memory game from the second session with students. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 12 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figures 10 and 11 show the box plots of the false answers of The Questionnaire Results of the Educators the Memory game and the difference between the correct and The TAM questionnaire results of educators from both false answers, respectively. Figures 12 and 13 show the box interviews are displayed in Table S12 in Multimedia Appendix plots of the SUS and TAM questionnaires, while Multimedia 9. Table S13 in Multimedia Appendix 9 shows the Wilcoxon Appendix 12 and 13 display the individual plots of the SUS and signed rank test results of educators for both questionnaires. TAM questionnaires. Figure 10. Box plot of false answers for the Memory game from the second session with students. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 13 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 11. Box plot of the difference between the correct and false answers for the Memory game from the second session with students. Figure 12. Box plot of SUS scores of the educators from the first (October) and second (March) sessions. SUS: System Usability Scale. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 14 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Figure 13. Box plot of TAM questionnaire answers of educators from the first (October) and second (March) sessions. TAM: Technology Acceptance Model. classroom, as homework or as auxiliary material, only one The Comments and Suggestions of the Educators educator said she would not use the games until the parents The educators’ feedback (Multimedia Appendix 14) was also were sufficiently aware of this system and its benefits. analyzed under prototyped game ideas, potential game ideas, Multimedia Appendix 15 shows the most common words or and general feedback. For the first segment, educators were phrases in the comments used by the educators. The most asked to comment on current games, game difficulty levels, common five words were feedback, audial feedback, graphics, and game-related features. Instead of words, one of well-thought, visual elements, and reward. The only negative the educators proposed that game pictures could be used for the feedback was “Will not use,” which was said due to the first part of the category. Another educator suggested that visuals educator’s concern about parents’ unfamiliarity with tablet use should enrich the Math game without creating distractions. in special education. Other teachers suggested that creating a shape by collecting a gem in the Space game would support children’s sketching Discussion ability. She also said that a star or a diamond should be gained instead of points since these symbols correspond to the special Principal Findings education center’s current motivation system. In this study, five different serious games and their prototypes were developed to provide training material for children with Educators were also asked if there was any game they want to any type of SpLD. The proposed games were evaluated in a see in this set of games. One educator suggested a game that two-phase study consisting of the SUS and TAM questionnaires, enables the players to write letters by tracing a line or an area and open-ended questions for 23 children with SpLD and 10 with fingers. Another educator said that a game based on finding educators. The main findings of this study showcase the synonyms or antonyms would be helpful to children with potentials of using serious games in SpLD training, the impact dyslexia. Rhythmic counting is indicated as another possible of adding adaptive difficulty systems that enable personalized addition to the Math game. A “Spot 7 differences” type of puzzle training experiences, and the outcomes of player-oriented design and jigsaw puzzles were also suggested. Finally, an educator where each prototype was structured based on the player said that a game to improve players’ auditory memory would statistics and suggestions. be beneficial. Three educators said the lack of feedback makes it difficult to feel like they are making progress. Two educators Results on comments and technology acceptance of the students believed that the degrees of difficulty should be varied. Two (Table S1 in Multimedia Appendix 9) show that children other educators indicated that holding scores on a leaderboard evaluated the games as easy-to-use and easy-to-learn. list and displaying previous scores would be helpful and Easy-to-use was a design criterion, and the results demonstrate encouraging to children. One of the educators indicated that a that it was accomplished. Since the game was designed with a guide for educators or parents would be helpful. Finally, single control mechanism and a single goal, children did not although most educators favored using this method in the https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 15 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer feel uncomfortable or overwhelmed, even when the game itself to play these games was 3.6 per week. This average can be was challenging (like the Space game). Children with SpLD accepted as high since the students’ permission to play games aged 7 to 11 years were comfortable with different input types was limited. Almost half of the kids indicated that they were (dragging, tapping, and joystick) when using only one of them allowed to play games only during the weekend or every other during the game. The single goal also mattered because most day, and they said that they would play these games in that of these children did not experience more complex games before allocated limited time. these interventions (Multimedia Appendix 16). To habituate A two-phase SUS and TAM evaluation was provided in this the children with SpLD to serious games, not only difficulty study, although the SUS scores between initial and final versions levels but also the complexity of games were increased in a of the games were demonstrating similar adjective ratings (ie, stepwise fashion. best imaginable, in our case) as mentioned by Bangor et al [20]. The results of the questionnaires showed that the outcomes of However, it was seen that in the second interview, the average the feedback-related questions were relatively low. There was of the SUS was higher, and the SD was lower, which can be no audio feedback in the proposed games, and visual feedback considered to demonstrate the condensed results regarding the was reduced to prevent any possible distraction. After this result, outcome of the games, where the hesitations and fluctuations the final version of the games was modified to have slightly from the data were less than the initial versions. more visual feedback. The questionnaire results of the educators A rule-based adaptive difficulty system was also proposed in supported these initial findings. Educators remarked that the serious games that adjust difficulty during the game, in contrast games were easy to learn and play, but there was a lack of to other studies in the literature that used a system that changes feedback in the final version of the games, although the games difficulty between two play sessions [19] and used predefined included several visual feedbacks. The children’s average points levels, which is limited [21]. The Space game’s adaptive version indicated that verbal games (the Word and Category games) did not change the results considerably for the students. had the lowest scores (8.36 and 8.08 out of 10, respectively), Moreover, the game’s speed change was practically equal, which which may stem from the duration of the games. Both the Word means the users nearly had the same statistics at the same and Category games took more than four times longer than other difficulty level. It can be interpreted in a couple of ways. The games during game play. Since playtime did not include any first possible reason for this result can be the adaptive rules or level mechanism in these games, it became boring for children effects not being well-adjusted, which is too difficult to achieve. after a while. In addition, children gave a correct answer in 54.8 In that case, the players will usually play in normal difficulty. seconds (Word game) and 38 seconds (the second part of the If adaptive changes affect the game difficulty, too many players Category game), which can be accepted as exceedingly long will probably increase or decrease their pace to the standard time durations for students with dyslexia. The point difference difficulty level. The other possible reason for this result can be between the Category and Word games is based on the visual the game difficulty being too balanced. The game could have differences. The Word game has a child avatar and a kitchen such a difficulty that players can collect the gem and crash into illustration, which is a more child-oriented visual theme. The enemies at almost the same rate. However, this is not a desired Word game, which only includes a blackboard and some outcome since children should encounter this situation after a writings, was liked by the children, and they completed the successful game period to feel their progress. The Wilcoxon game in their given time period. This result conflicts with test for the Memory game results showed that children had Shabbir et al’s [19] rules of serious games for children with better performance in the game’s adaptive version, but it was dyslexia. Shabbir et al [19] claim that not using a plain not statistically significant. Since the adaptive version increases background can cause a distraction for children. However, the its difficulty while players show good performance, children children’s questionnaires and comments show that game-related played a more challenging version and showed better static background images or avatars increased the immersion. performance, which means that the flow strategy worked. Due The kitchen image in the background or the child avatar that to the increased difficulty level, children’s performance could asks questions was not mandatory for the Word game, but not be statistically different but was still better. children stated that they loved both versions. Some of the children suggested a similar visual design for the Category Especially for adaptive games, more tests should be done for game. different variables. This study includes the modification of only one variable—enemy spawn frequency for the Space game and The average points by both participant groups were between question screen time for the Memory game—which gives 7.9 and 8.84 out of 10. When these results were combined with promising results for the future, especially in the Memory game. the questionnaire results, it can be easily said that both Children did not realize they were playing faster, and they participant groups liked all games, and they can accept all parts showed slightly better performance in adaptive versions. More of the system as a training source. Unlike other studies on tests can be applied by changing grid numbers or changing the serious games for children with SpLD, this study proposes a predefined letters in the left panel. During the test of the adaptive collection of games to educate children with any type of SpLD. games, only 1 student said that he understood the change in the All of the games can be played by children regardless of their Memory game and explained the correct reason, while others SpLD types. A specially designed set of games for these children could not explain the change they felt in both games, misjudged can provide a complete training source and better experience them, or did not feel any change at all. Many students showed rather than only playing a game directly related to their disorder that the games influenced them by expressing sadness, joy, or or difficulty. The average number of days when students want surprise during the games—mainly in the Space game. Except https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 16 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer for 1 student, the students were eager to replay the games, and children with SpLD. Both educators and students participated they demanded to continue to play after the interventions were in the game-based interventions, and their feedback was complete. In the first interview, 1 student stated that playing recorded. Both participant groups were enthusiastic about the the Space game for the second time would be unnecessary, so proposed “all-in-one” scheme. Besides, different design she did not want to replay the game. The same student did not requirements that have been listed in previous studies were make such a request in other games or during the second investigated with an additional adaptive difficulty level. These interview. The students wanted different visual themes in the adjustments and personalized approaches increased children Math game, showing that the game could be diversified visually, with dyslexia’s immersion in the games. One of the study’s not mechanically. The “boss creatures” concept was the most contributions was the use of serious games to target various highlighted suggestion of the students, which was proposed as SpLD types. Each child participant had more than one form of an improvement several times. It may be good to add them as SpLD, and playing games that focus on dyslexia or dyscalculia positive feedback in the games that are designed in levels instead has been beneficial to them instead of playing games that focus of endless games. Besides, although the game ends due to the on only one of these difficulties. player’s mistake in endless games, in level-based games, the Furthermore, playing games like the Memory and Space games player’s success leads to a new stage, which changes the feeling and education-oriented question-and-answer games resulted in at the end of the game. Thus, boss creatures can be good for increased positive feedback and excitement. Furthermore, the strengthening the feeling at the end of the episode. Students’ students mentioned that they preferred to play in a general comments on serious games were usually game-related system rather than a system focusing only on an area where suggestions, and they were not interested in the games’ they already had difficulties. The teachers’ overall impressions educational purpose. This is because children did not see these were highly positive, and most of them intend to use these games games as educational material, but they accepted these games as part of their curriculum’s supplementary material. This as standard games, which can be considered in line with the integration will enable the games to be tested for a longer period fundamental purpose of the serious game concept. so that new enhancements can be added to the proposed games with additional evaluations to measure their educational impact. Conclusion This study proposed and tested five serious games, each having a different number of prototypes, specifically developed for Acknowledgments The authors thank Dislev Türkiye Disleksili Çocuklar Vakfı (Dislev Turkish Association of Children with Dyslexia) for their tremendous help during data acquisition. The authors also thank the children and their educators for participating in this study and giving valuable and detailed feedback. Conflicts of Interest None declared. Multimedia Appendix 1 The workflow diagram of this study. [PNG File , 29 KB-Multimedia Appendix 1] Multimedia Appendix 2 Screenshot of the picture version of the Memory game from Prototype IV. [PNG File , 83 KB-Multimedia Appendix 2] Multimedia Appendix 3 Screenshot of the first part of the Category game from Prototype II. [PNG File , 564 KB-Multimedia Appendix 3] Multimedia Appendix 4 Screenshot of the second part of the Category game from Prototype II. [PNG File , 439 KB-Multimedia Appendix 4] Multimedia Appendix 5 Screenshot of the second part of the Category game with an alternative keyboard layout from Prototype III for the theme “Clothes". [PNG File , 477 KB-Multimedia Appendix 5] https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 17 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Multimedia Appendix 6 The optimal game play corridor and the golden path figure are adapted from Thomas 2010. [PNG File , 109 KB-Multimedia Appendix 6] Multimedia Appendix 7 Visual diagram of the adaptive difficulty system. [PNG File , 33 KB-Multimedia Appendix 7] Multimedia Appendix 8 Screenshot of the letter version of the Memory game from Prototype III. [PNG File , 123 KB-Multimedia Appendix 8] Multimedia Appendix 9 Supplementary files and tables on game statistics and results of the questionnaires. [DOCX File , 20 KB-Multimedia Appendix 9] Multimedia Appendix 10 Student feedback illustration map. [PNG File , 53 KB-Multimedia Appendix 10] Multimedia Appendix 11 Student feedback word cloud. [PNG File , 112 KB-Multimedia Appendix 11] Multimedia Appendix 12 Individual plot of System Usability Scale scores of educators from the first (October) and second (March) sessions. [PNG File , 13 KB-Multimedia Appendix 12] Multimedia Appendix 13 Individual plot of Technology Acceptance Model questionnaire answers of educators from the first (October) and second (March) sessions. [PNG File , 14 KB-Multimedia Appendix 13] Multimedia Appendix 14 Educator feedback illustration map. [PNG File , 47 KB-Multimedia Appendix 14] Multimedia Appendix 15 Educator feedback word cloud. [PNG File , 96 KB-Multimedia Appendix 15] Multimedia Appendix 16 Photo of a child during the intervention. [PNG File , 313 KB-Multimedia Appendix 16] References 1. Lipkin PH, Okamoto J, Council on Children with Disabilities, Council on School Health. The Individuals With Disabilities Education Act (IDEA) for children with special educational needs. Pediatrics 2015 Dec;136(6):e1650-e1662. [doi: 10.1542/peds.2015-3409] [Medline: 26620061] 2. Balcı E. Disleksi hakkında öğretmen görüşleri ve karşılaştıkları sorunlar. Ege Eğitim Dergisi 2019;20(1):162-179. [doi: 10.12984/egeefd.453922] https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 18 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer 3. Bellotti F, Ott M, Arnab S, Berta R, de Freitas S, Kiili K, et al. Designing serious games for education: from pedagogical principles to game mechanisms. 2011 Oct 1 Presented at: The 5th European Conference on Games Based Learning; 2011; Athens, Greece. 4. Franceschini S, Gori S, Ruffino M, Viola S, Molteni M, Facoetti A. Action video games make dyslexic children read better. Curr Biol 2013 Mar 18;23(6):462-466 [FREE Full text] [doi: 10.1016/j.cub.2013.01.044] [Medline: 23453956] 5. Bartolomé NA, Zorrilla AM, Zapirain BG. Dyslexia diagnosis in reading stage though the use of games at school. 2012 Jul 30 Presented at: 17th International Conference on Computer Games; July 30-August 1, 2012; Louisville, KY. [doi: 10.1109/cgames.2012.6314544] 6. Rauschenberger M, Rello L, Baeza-Yates R, Bigham JP. Towards language independent detection of dyslexia with a web-based game. In: Proceedings of the 15th International Web for All Conference. 2018 Apr 01 Presented at: W4A '18; April 2018; Lyon, France p. 1-10. [doi: 10.1145/3192714.3192816] 7. Avila-Pesantez DF, Vaca-Cardenas LA, Avila RD, Padilla NP, Rivera LA. Design of an augmented reality serious game for children with dyscalculia: a case study. In: Botto-Tobar M, Pizarro G, Zúñiga-Prieto M, D'Armas M, Sánchez MZ, editors. Technology Trends: 4th International Conference, CITT 2018, Babahoyo, Ecuador, August 29–31, 2018, Revised Selected Papers. Cham: Springer; 2019:165-175. 8. Holz H, Brandelik K, Beuttler B, Brandelik J, Ninaus M. How to train your syllable stress awareness - a digital game-based intervention for German dyslexic children. Int J Serious Games 2018 Sep 25;5(3):37-59. [doi: 10.17083/ijsg.v5i3.242] 9. Sood M, Toornstra A, Sereno M, Boland M, Filaretti D, Sood A. A digital app to aid detection, monitoring, and management of dyslexia in young children (DIMMAND): protocol for a digital health and education solution. JMIR Res Protoc 2018 May 17;7(5):e135 [FREE Full text] [doi: 10.2196/resprot.9583] [Medline: 29773528] 10. Flogie A, Aberšek B, Kordigel Aberšek M, Sik Lanyi C, Pesek I. Development and evaluation of intelligent serious games for children with learning difficulties: observational study. JMIR Serious Games 2020 Apr 16;8(2):e13190 [FREE Full text] [doi: 10.2196/13190] [Medline: 32297864] 11. Yıldırım O. Development of adaptive serious games for the training of children with specific learning disabilities. OpenMETU. 2020 Sep 10. URL: https://open.metu.edu.tr/handle/11511/69207 [accessed 2021-05-27] 12. Tobler-Ammann BC, Surer E, de Bruin ED, Rabuffetti M, Borghese NA, Mainetti R, et al. Exergames encouraging exploration of hemineglected space in stroke patients with visuospatial neglect: a feasibility study. JMIR Serious Games 2017 Aug 25;5(3):e17 [FREE Full text] [doi: 10.2196/games.7923] [Medline: 28842388] 13. Tobler-Ammann BC, Surer E, Knols RH, Borghese NA, de Bruin ED. User perspectives on exergames designed to explore the hemineglected space for stroke patients with visuospatial neglect: usability study. JMIR Serious Games 2017 Aug 25;5(3):e18 [FREE Full text] [doi: 10.2196/games.8013] [Medline: 28842390] 14. Brooke J. SUS: a “quick and dirty” usability scale. In: Jordan PW, Thomas B, McClelland IL, Weerdmeester B, editors. Usability Evaluation In Industry. United Kingdom: Taylor & Francis; 1996. 15. Davis, Jr FD. A technology acceptance model for empirically testing new end-user information systems: theory and results. DSpace@MIT. 1985. URL: https://dspace.mit.edu/bitstream/handle/1721.1/15192/14927137-MIT.pdf [accessed 2021-05-27] 16. Czikszentmihalyi M. Flow: The Psychology of Optimal Experience. New York, NY: Harper & Row; 1990. 17. Thomas JM, Young RM. Annie: automated generation of adaptive learner guidance for fun serious games. IEEE Trans Learning Technologies 2010 Oct;3(4):329-343. [doi: 10.1109/tlt.2010.32] 18. Hendrix M, Bellamy-Wood T, McKay S, Bloom V, Dunwell I. Implementing adaptive game difficulty balancing in serious games. IEEE Trans Games 2019 Dec;11(4):320-327. [doi: 10.1109/tg.2018.2791019] 19. Shabbir N, Bhatti Z, Hakro DN. Serious game user interface design rules for dyslexic children. 2019 Dec 14 Presented at: 13th International Conference on Mathematics, Actuarial Science, Computer Science and Statistics; December 14-15, 2019; Karachi, Pakistan p. 1-6. [doi: 10.1109/macs48846.2019.9024786] 20. Bangor A, Kortum P, Miller J. Determining what individual SUS scores mean: adding an adjective rating scale. J Usability Stud 2009 May 01:114-123. 21. Cuschieri T, Khaled R, Farrugia VE, Martinez HP, Yannakakis GN. The iLearnRW game: support for students with dyslexia in class and at home. 2014 Presented at: 6th International Conference on Games and Virtual Worlds for Serious Applications; September 9-12, 2014; Valletta, Malta p. 1-2. [doi: 10.1109/vs-games.2014.7012167] Abbreviations IDEA: Individuals with Disabilities Education Act LD: learning disorder SpLD: specific learning difficulties SUS: System Usability Scale TAM: Technology Acceptance Model https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 19 (page number not for citation purposes) XSL FO RenderX JMIR SERIOUS GAMES Yildirim & Surer Edited by N Zary; submitted 07.04.21; peer-reviewed by CW Ong, J Brooke; comments to author 22.04.21; revised version received 30.04.21; accepted 21.05.21; published 31.05.21 Please cite as: Yildirim O, Surer E Developing Adaptive Serious Games for Children With Specific Learning Difficulties: A Two-phase Usability and Technology Acceptance Study JMIR Serious Games 2021;9(2):e25997 URL: https://games.jmir.org/2021/2/e25997 doi: 10.2196/25997 PMID: ©Oguzcan Yildirim, Elif Surer. Originally published in JMIR Serious Games (https://games.jmir.org), 31.05.2021. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Serious Games, is properly cited. The complete bibliographic information, a link to the original publication on https://games.jmir.org, as well as this copyright and license information must be included. https://games.jmir.org/2021/2/e25997 JMIR Serious Games 2021 | vol. 9 | iss. 2 | e25997 | p. 20 (page number not for citation purposes) XSL FO RenderX

Journal

JMIR Serious GamesJMIR Publications

Published: May 31, 2021

Keywords: serious games; adaptive games; specific learning difficulty; usability; system usability scale; technology acceptance model; training; development; adaptation; gaming; learning disability; children; education; teacher

There are no references for this article.