Augmented Reality Based Personalized Learning in Autism Spectrum Disorder Reading Skills

Abstract

Reading can be a way to increase students’ social interaction and writing, especially for those continuing to a higher level of education. More recent research shows that children with Autism Spectrum Disorder (ASD) have a variety of reading profiles that differ across a range of reading sub-skills. Meanwhile, the school teaching system provides one-size-fits-all learning for students with different abilities. This study aimed to test the use of personalized learning model-based Augmented Reality (AR) in supporting the learning of early reading skills in aspects of reading words, syllables, and letters. Four students with ASD participated in various cross-behavioral investigations. Data was collected based on early reading skills in words, syllables, and letters using the Early Grade Reading Assessment (EGRA) instrument, which had been adjusted according to the children’s abilities (three groups of children). Participants went through baseline-1 sessions, AR interventions, and baseline-2. The results show increased student learning outcomes in reading words and syllables. However, there was no significant increase in learning outcomes regarding recognizing letters.

Keywords

augmented reality personalized learning autism spectrum disorders early reading

Introduction

Reading skills are part of cognitive abilities, combining information interactions and individual knowledge bases to achieve reading goals (Mart, 2012; Richards & Burns, 2012). Meanwhile, reading skills allow readers to create new knowledge and share it with others (Thompson et al., 2022; Zajic et al., 2020). The importance of mastering reading skills is to have the knowledge base to listen to others speak, and they have a high success rate in writing while they have strong knowledge bases through reading in general (Richards & Burns, 2012). Furthermore, reading supports academic competence to continue to the following educational stages (Kaya, 2015). Therefore, reading skills become a knowledge base to master for every individual.

Reading skills are essential for children with Autism Spectrum Disorder (ASD), although the approach and implementation timing can differ for each individual. Reading skills can help children with ASD in various aspects of development and social interactions (McIntyre et al., 2017; Omar & Bidin, 2015a). Reading can also help improve communication skills in speaking and understanding written language (Gilakjani & Sabouri, 2016). Although some children with ASD may struggle with verbal communication, they can still learn to communicate through the written word. Learning to read can help expand the vocabulary and language understanding of children with ASD (Jackson & Hanline, 2020) This can help them interact with others and understand instructions and information in various contexts.

Most reading studies involving children with ASD have reported specific challenges with reading comprehension and the ability to read full words (Brown et al., 2013; Nally et al., 2018; Zuccarello et al., 2015). However, the investigation highlights that child with ASD showed variation profile reading in various sub-reading skills, including phonological awareness, decoding, fluency, and comprehension (Johnels et al., 2019; McIntyre, Solari, Grimm, et al., 2017; Nation et al., 2006; Singh et al., 2017). Challenges in language skills, especially reading skills in children with ASD, are generally influenced by disturbances in cognitive development (Nally et al., 2018).

In its implementation, improving students’ daily and academic skills is a challenge in the special school (SLB) teaching system (Petersen, 2016). SLB is an educational institution specifically designed to provide education and support to students with special educational needs, including ASD students. It is important to understand that every child with ASD is unique, with different preferences, needs, and learning styles (Nilsen & Bacso, 2017). The personalized learning approach recognizes these differences and seeks to present learning material in a way that suits the individual characteristics of each child. However, the number of educators must be commensurate with the number of students (Zhou et al., 2021). Under these conditions, there is a concern between the individual, the learning, and the support they need. Technology can be an option to overcome this gap. Previous research said that using technology in learning with personalization can be varied and modified based on characteristics and student needs, so the lecturer can easily apply personalized learning (Kurniawati, 2020).

Augmented Reality (AR) is a tool for the learning process, enabling the user to see and combine virtual objects with a natural environment to create interaction between the real world and virtual objects (Antão et al., 2020; Ayoub & Pulijala, 2019). A virtual object can help children with ASD interpret realistic images and stimulate children’s cognitive process to achieve knowledge transfer and process (Kolomoiets & Kassim, 2018). AR also can attract attention and increase the involvement of children with ASD in the learning materials (Mohd et al., 2019).

AR has been widely used in the world of education, such as Interactive Textbooks where pictures in textbooks can produce 3D animation or additional information when scanned using an AR application (Howorth et al., 2019; Khoirunnisa et al., 2023; Kolomoiets & Kassim, 2018; McMahon et al., 2016). Then, Virtual Tours, where users can “visit” these places virtually and get contextual information when interacting with the AR environment (Andri et al., 2018; Safitri et al., 2017). Alternatively, as a practice simulation by providing practical guidance. For example, users can learn how to assemble a product or paint with the help of AR instructions displayed over actual object (Squire & Klopfer, 2007; Wang et al., 2018).

Overall, using AR in education can increase student engagement and understanding of the subject matter. AR technology interests children, including elementary school-age children (Danaei et al., 2020; Howorth et al., 2019; Tang et al., 2019). Objects or images displayed in AR help keep their attention and make learning more fun. AR allows abstract or complex concepts to be illustrated more visually and interactively (Padeliadu & Antoniou, 2014). This can help children better understand the subject by seeing it concretely. AR can also be adapted to the needs of each child. Content, speed, and difficulty level can be adjusted according to the child’s level of understanding and ability (Bartoli et al., 2014; Maglione et al., 2012).

AR has been used to improve ASD children’s reading skills, but only limited to Interactive Textbooks. The research found targeted different aspects of reading, among others phonological awareness (understanding corresponding alphabet-sound in reading and spelling) (Antão et al., 2020), increasing vocabulary (understanding words that used to be read and tied to spoken words) (Hashim et al., 2022; McMahon et al., 2016; Tang et al., 2019), Furthermore, other research combines several skills: phonic awareness, vocabulary improvement, and reading comprehension (Howorth et al., 2019; Kolomoiets & Kassim, 2018). In its implementation, the majority of researchers only focus on ASD children in general, while the individual conditions of ASD have different reading abilities.

Previous research on improving reading skills using AR for students with ASD has been carried out a lot, but most of its implementation only focused on children with ASD in general; meanwhile, heterogeneity was found among students with ASD in reading skills. Based on this description, the authors researched implementing personalized augmented reality to improve ASD students’ reading skills. With this in mind, the author tries to offer a new perspective on the implementation of AR, which is implemented through personalized learning models and the Synthetic Structural Analytical (SAS) (Djauzak, 1996) reading method, which is a learning approach that focuses on the analysis and synthesis of language elements such as phonemes, graphemes, morphemes, and syntax. This approach can help children with ASD understand the detailed structure of language, which is essential in developing reading skills (Muammar et al., 2018).

Reading Skills Interventions for Children With ASD

Reading is a very complex skill prerequisite for adapting to social life, where much information is communicated in written form (Rayner et al., 2012). The analysis conducted by Williamson et al. (2012) revealed three profiles of reading comprehension towards the views of children with ASD: text-bound comprehension, strategic comprehension, and imaginative understanding. Factors influencing comprehension across all three profiles are based on differences in language and knowledge, action strategies (e.g., use of comprehension strategies, ability to make certain types of inferences), and text factors provided.

ASD describes a person with social communication differences (Lord et al., 2018). They find it easier to learn vocabulary but have difficulties understanding reading incredibly complex reading (Reutebuch et al., 2015) and experience the challenge of comprehension skills to answer questions (Bethune & Wood, 2013). Children with disabilities tend to have to be assisted by technology in illustrating readings so they can easily understand the contents of the reading (Munir et al., 2018).

Many media have been applied to the reading skills of students with ASD, such as Computer-Presented Video Models (Kinney et al., 2003), Computer-Assisted Instruction (Coleman-Martin et al., 2005; McKissick et al., 2018), Collaborative Virtual Environment (Moore et al., 2005) Multimedia Graphic and Text (Omar & Bidin, 2015b), Multimedia in Education for Special Education (MESE) (Munir et al., 2018), Augmented Reality Storybooks (Danaei et al., 2020; Khoirunnisa et al., 2023), Educational Game (Munir et al., 2021), and Interactive Multimedia (Paramesti et al., 2021). The latest media development used in the education of children with special needs is Augmented Reality (AR) technology. AR is a high-tech product that allows users to view and combine virtual objects with natural environments (Ayoub & Pulijala, 2019). AR uses real-world settings but allows users to interact with virtual objects (Antão et al., 2020). Observation of virtual objects that represent realistic images stimulates children’s cognitive processes leading to a qualitatively new level in information exchange and processing (Kolomoiets & Kassim, 2018). Many reviews of AR-based interventions for students with ASD were conducted. Apart from assisting in life skills, AR has also begun to be used to help improve the academic skills of students with ASD (Wong et al., 2019). AR can attract attention and increase the involvement of students with ASD in the material being delivered (Mohd et al., 2019). AR, especially for students with ASD, has a positive effect on improving different domains such as social interaction, social communication skills, verbal and nonverbal communication, facial emotion recognition procedures, attention skills, or functional life (Berenguer et al., 2020).

The application of technology in learning allows innovation to occur so that learning becomes effective. One way to achieve effective learning is to combine the real world with the virtual world and be interactive so that students with ASD can focus on the material being studied, and this can increase student motivation.

Purpose

This research aims to determine the effect of AR-based personalized learning on the reading skills of children with ASD. This AR-based learning activity includes reading training using the SAS method and drag-and-drop quizzes. The specific research questions are as follows:

RQ 1

What is the effect of using AR-based personalized learning using the SAS reading method on improving word recognition in children with ASD?

RQ 2. What is the effect of using AR-based personalized learning using the SAS reading method on increasing the ability to recognize syllables in children with ASD?

RQ 2

What is the effect of using AR-based personalized learning using the SAS reading method on increasing the ability to recognize letters in children with ASD?

Methodology

Participants

Four boys with ASD (RA, RD, RG, and RR) were enrolled in this study from special schools in Indonesia. The characteristics of these children are almost the same: difficulty making eye contact, using unusual language, and being easily distracted. The first participant called RA was eight years old, the second participant called RD was ten years old, the third participant called RG was 13 years old, and the fourth participant called RR was 12 years old. All children were selected based on the following: (a) diagnosed with ASD, (b) will or still learning reading skills, (c) have no physical disabilities that hinder the implementation of activities, (d) can understand the researcher’s directions even though communication is limited and (e) consent to participate in this study.

Two weeks before the start of this study, all participants were given a test from the Early Grade Reading Assessment (EGRA) (Gove & Wetterberg, 2011) to determine their ability to recognize letters, syllables, and words. For each correct question participant get a score of 1 and the scores are totaled. Sometimes children are not able to answer a single question correctly on a given subtask. For example, the student may not be able to identify a single letter sound or read a single word. This results in a sub-reading skill score of zero. Zero scores are referred to as the proportion of students who cannot perform even one item correctly for a given sub-reading skill. Of the four participants, there was heterogeneity in early reading ability by the results of a recent study on the reading ability of children with ASD. Early reading refers to individuals’ ability, especially children, to understand and read words and texts at the early stages of reading development (Davidson & Ellis Weismer, 2014; Grindle et al., 2013). This is an essential stage in learning to read, where children begin to recognize letter sounds and develop an understanding of how written words relate to audible sounds. Condition of RA, RD, RG, and RR, they have a difficulty reading syllables, combining them into words, and pronouncing words without assistance.

RD could read more words without assistance than other participants and read several syllables correctly. It can be seen in Table 1 that RD has the highest score in the “without assistance” column in all skills. Some of the obstacles faced by RD in word reading skills are the pronunciation of “zebra”, “kaktus”, “rumput”, “lantai”, “harimau”, and “dompet”. This is due to the pronunciation of the syllable “ut” in “rumput”, “ai” in “lantai”. “au” in “harimau”, and “pet” in “dompet” are still unclear. For word “wastafel” cannot be the word at all.

Table 1.

The Results of the Preliminary Study on the Reading Skills.

Participant	Reading Words		Reading Syllables		Read Letters
Participant	Without Assistance	With Assistance	Without Assistance	With Assistance	Without Assistance	With Assistance
RA	7	8	9	16	20	10
RD	8	7	14	11	25	5
RG	6	9	13	12	23	7
RR	6	9	10	15	19	11

RG has intermediate abilities among the other participants. In the aspect of reading words, nine words still need help, including “bebek”, “pisau”, “nanas”, “kasur”, “cicak”, “gelas”, “lilin”, “koboi”, and “raket”. RG has difficulty pronouncing words that start with “k”, “s”, “t” and “n”, so the words cannot be read clearly.

RA and RR have similarities in reading skills. In reading the letters in random order, it can be seen from Table 1 that the RA and RR scores in the “with assistance” column had the highest scores compared to other participants. RA and RR have similarities in that they cannot distinguish letters that are similar in shape (such as p and q, b and p). In the skills of reading words and syllables, the ability was still below the other two participants.

Setting

The intervention was carried out in a one-on-one arrangement for each participant at special school. In the early stages, apperception and conditioning were carried out to implement personalized learning models assisted by AR media. The apperception activity aims to create student readiness for the material to be studied, while conditioning students aims to build readiness in participating in the learning process using the AR-assisted personalized learning model. At the core stage, it is a series of learning processes to achieve predetermined learning objectives. The learning objective to be achieved is to be able to read words, syllables, and letters correctly. The implementation of this activity is carried out interactively and fun with the direct involvement of students. Children with ASD use AR by scanning images in the books provided, then the system will display 3D objects equipped with sound. Lastly, the teacher conducts an attitude assessment by observing students during the intervention. Assessments are carried out to see how students respond to all instructions in the application and the extent of interest in the application created. This assessment was not carried out in depth and only served as an additional resource for researchers. In addition, the teacher conducts a knowledge assessment in the form of an oral test using the EGRA instrument. All activities are carried out in the classroom using the help of a tablet to run the AR program. (Figure 1).

Figure 1.

The intervention set up for a mobile AR application.

AR for Reading

AR development is carried out using the Linear Sequential Model by Pressman (Pressman, 2010). This mobile AR application allows students to be involved in learning and given the freedom to determine the words they will learn. This application is equipped with group selection to get material content according to children’s abilities. When the child scans the pictures in the book on the tablet screen, there is a “recognize” button to carry out the SAS reading method. In the first condition, the word will appear according to the scanned image (Figure 2(a)). When clicking the “recognize” button, the word will be parsed into syllables (Figure 2(b)), then it will be broken down into letters (Figure 2(c)), and so on. To make it easier for children with ASD to learn, this mobile AR application is equipped with 3D sounds and objects. In addition, knowledge development is carried out by answering questions on completing words and choosing letters.

Figure 2.

Display of the SAS method in AR applications. (a) Word forms, (b) parsed into syllables, and (c) parsed again into letters.

Data Collection

The data was obtained from the number of correct answers in each aspect of reading skills, namely words, syllables, and letters. Assessment is done by dividing the number of correct answers by the total number of questions. Group one is students who have higher abilities compared to groups two and three. Each group has a minimum mastery of each aspect of reading skills which can be seen in Table 2.

Table 2.

The Number of Words, Syllables, and Letters on the Instrument.

Group	Number of Words	Minimum Mastery	Number of Syllables	Minimum Mastery	Number of Letters	Minimum Mastery
1	15	10	32	20	30	24
2	15	8	30	15	30	24
3	15	4	30	10	30	24

Note. Group: 1-high reading ability, 2-medium reading ability, 3-low reading ability; number of words: The number of words learned; number of syllables: Number of syllables studied; number of letters: The number of letters learned; minimum mastery: The minimum amount of mastery of each reading sub-skill.

Questions are ordered randomly to minimize the child memorizing words, syllables, or letters. If the participant answers correctly, it will be recorded on the score sheet as the correct answer without assistance or correct with assistance. The assistance provided by the researcher is to use the correct pronunciation, then the child can follow again and again. If the participant does not answer correctly or does not answer, it will be recorded as an incorrect answer. The waiting time for response is around 10 seconds. The assessment was carried out at baseline-1, AR intervention, and baseline-2.

Procedure

Experimental method with multiple baselines across participant design was used to test the effects of an AR-based personalized learning program with several ASD students who have different initial reading abilities. Baseline measurement data was performed for each subject before the intervention began. This is done to get a good understanding of the behavior or initial characteristics of each subject. The intervention was carried out six times with one-on-one sessions. After baseline measurements are taken, intervention is done at different times for each subject. After the intervention, baseline measurements were carried out again to determine changes in students’ reading skills.

Baseline-1

At baseline-1, each participant completed three assessments on the aspects of the word, syllable, and letter recognition. There is no feedback when the assessment is conducted, but additional instructions are given in answering questions. Additional instructions given such as the pronunciation of the word “nanas”, the researcher mentions the word “na”, then students follow the pronunciation. Participants were instructed to answer questions; if they could not answer, they would be skipped and asked again at the end of the assessment. Apart from assessing students’ abilities, repeated assessments consider the participant’s condition.

Intervention AR

Participants were trained to use the mobile AR application to scan images in books so that AR content (3D objects, sound, and text) is displayed. Personalized learning model procedures are applied to help learn adapted to the child’s abilities. The activity steps carried out by participants for each meeting are based on the syntax of the personalized learning model.

The first syntax is to convey the purpose and prepare the participants. The researcher conveys the learning objectives directly to the participants, and the participants listen to what the researcher says. After that, the AR application will display a video as an apperception step. The video contains an introduction to the alphabet in the form of a song.

The second syntax is organizing Student Choice and Voice. The activities carried out in the AR application are that participants are directed to choose groups according to the results of the initial reading assessment. After that, participants can select material in the book to study according to their interests, which is then scanned using the AR application.

In the book, 15 words can be learned by students. The third syntax is Accessing Information. The teacher asks students/participants to scan the book with a marker as the study material. When students select the content, they want to learn, they begin to scan the book, and a 3D image and text will appear. Then, students read the words in the AR application. Then, the teacher gives students/participants the freedom to choose material in the book according to the participants’ interests and provides the opportunity to ask questions.

The fourth syntax is Developing Knowledge. The activity carried out by participants is answering quizzes in the form of drag-and-drop and multiple choice. Such as, spelling and identifying the letters of a word.

The last syntax is Giving feedback. The activity carried out in the AR application is giving feedback based on the answers to practice questions. This feedback is complemented by a sound corresponding to positive feedback or encouragement to try again. Participants attended sessions six times, with assessments carried out in each session using the EGRA instrument. At each meeting, the three initial reading skills are assessed at the end of the intervention.

Baseline-2

At baseline-2, a reassessment was carried out after the AR intervention was implemented. The assessment was carried out three times using the EGRA instrument. Baseline-2 was carried out to see the effects after implementing personalized AR-based learning for the reading skills of children with ASD. The baseline-2 assessment procedure is the same as in baseline-1, where the questions are ordered randomly to minimize the child’s memorization of words, syllables, or letters. If the participant answers correctly, it will be recorded on the score sheet as the correct answer without assistance or correct with assistance. If the participant does not answer correctly or does not answer, it will be recorded as an incorrect answer.

Results

RQ 1: What is the Effect of Using AR-Based Personalized Learning Using the SAS Reading Method on Improving Word Recognition in Children With ASD?

RA

The average correct response at baseline-1 for the word reading aspect was 44%. After using the AR intervention, the average RA correct in answering questions began to increase significantly. During the intervention, the correct average for aspects of word reading was 62%. RA achieved the criteria for material mastery in the aspect of word reading, with a total of 11 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RA increased word mastery, with the correct average for this aspect of word reading being 76%. That way, until the end of the study, RA mastered 12 of the 15 words learned. RA results on the aspect of word reading are presented in Figure 3(a).

Figure 3.

Ability to read words phases A1-B-A2 for each participant.

RD

The average correct response at baseline-1 for the word reading aspect was 51%. After using the AR intervention, the average correct RD in answering questions began to increase significantly. During the intervention, the correct average for aspects of word reading was 76%. RD achieved the criteria for material mastery in the aspect of word reading, with a total of 13 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RD increased word mastery, with the correct average for this aspect of word reading being 89%. That way, until the end of the study, RD mastered 14 of the 15 words learned. The RD results on the aspect of word reading are presented in Figure 3(b).

RG

The average correct response at baseline-1 for the word reading aspect was 44%. After using the AR intervention, the average RG correct in answering questions began to increase significantly. During the intervention, the correct average for aspects of word reading was 82%. RG achieved the criteria for material mastery in the aspect of word reading, with a total of 13 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RG increased word mastery with a correct average for this aspect of word reading of 84%. That way, until the end of the study, RG mastered 13 of the 15 words learned. The RG results on the aspect of word reading are presented in Figure 3(c).

RR

The average correct response at baseline-1 for the word reading aspect was 42%. After using the AR intervention, the average correct RR in answering questions began to increase significantly. During the intervention, the correct average for aspects of word reading was 62%. RR achieved the criteria for material mastery in the aspect of word reading, with a total of 11 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RG increased word mastery with a correct average for this aspect of word reading of 78%. That way, until the end of the study, RR mastered 12 of the 15 words learned. The RR results on the aspect of reading words are presented in Figure 3(d).

RQ 2: What is the Effect of Using AR-Based Personalized Learning Using the SAS Reading Method on Increasing the Ability to Recognize Syllables in Children with ASD?

RA

The average correct response at baseline-1 for the reading aspect of the syllable was 37%. After using the AR intervention, the average RA correct in answering questions began to increase quite a bit. During the intervention, the correct average for aspects of reading syllables was 55%. RA achieved the criteria for material mastery in the aspect of reading syllables, with 17 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed an increase in RA mastery of syllables, with the correct average for this aspect of reading syllables was 71%. That way, until the end of the study, RA mastered 18 of the 30 syllables studied. RA results on the aspect of reading syllables are presented in Figure 4(a).

Figure 4.

Ability to read syllables phases A1-B-A2 for each participant.

RD

The average correct response at baseline-1 for the reading aspect of the syllable was 59%. After using the AR intervention, the average correct RD in answering questions increased quite a bit. During the intervention, the correct mean for aspects of reading syllables was 76%. RD achieved the criteria for material mastery in the aspect of reading syllables, with 23 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RD had an increased mastery of syllables, with a correct average for this aspect of reading syllables of 88%. That way, until the end of the study, RD mastered 24 of the 32 syllables studied. The RD results on the aspect of reading syllables are presented in Figure 4(b).

RG

The average correct response at baseline-1 for the reading aspect of the syllable was 59%. After using the AR intervention, the average RG correct in answering questions increased quite a bit. During the intervention, the correct mean for aspects of reading syllables was 76%. RG achieved the material mastery criteria in reading syllables, with 21 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RG had an increased mastery of syllables, with a correct average for this aspect of reading syllables of 88%. That way, until the end of the study, RG mastered 23 of the 30 syllables studied. RG’s results on the aspect of reading syllables are presented in Figure 4(c).

RR

The average correct response at baseline-1 for the reading aspect of the syllable was 48%. After using the AR intervention, the average correct RR in answering questions began to increase quite a bit. During the intervention, the correct mean for aspects of reading syllables was 58%. RR achieved the criteria for material mastery in the aspect of reading syllables, with 17 words mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed an increase in RR mastery of syllables, with a correct average for this aspect of reading syllables of 69%. That way, until the end of the study, RR mastered 18 of the 30 syllables studied. The RR results on the aspect of reading syllables are presented in Figure 4(d).

RQ 3: What is the Effect of Using AR-Based Personalized Learning Using the SAS Reading Method on Increasing the Ability to Recognize Letters in Children with ASD?

RA

The average correct response at baseline-1 for aspects of reading letters was 73%. After using the AR intervention, the average RA correct in answering questions began to increase quite a bit. During the intervention, the correct average for aspects of reading letters was 87%. RA achieved the criteria for material mastery in the aspect of reading letters, with a total of 28 letters mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RA did not increase mastery of letters. That way, until the end of the study, RA mastered 28 of the 30 letters learned. RA results on aspects of reading letters are presented in Figure 5(a).

Figure 5.

Ability to read phases A1-B-A2 letters for each participant.

RD

The average correct response at baseline-1 for aspects of reading letters was 88%. After using the AR intervention, the average correct RD in answering questions increased quite a bit. During the intervention, the correct average for aspects of reading letters was 87%. RA achieved the criteria for material mastery in the aspect of reading letters, with a total of 29 letters mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RD did not increase in mastery of letters. That way, until the end of the study, RD mastered 29 of the 30 letters learned. The results of the RD on the aspect of reading letters are presented in Figure 5(b).

RG

The average correct response at baseline-1 for aspects of reading letters was 82%. After using the AR intervention, the average RG correct in answering questions began to increase quite a bit. During the intervention, the correct average for aspects of reading letters was 93%. RG achieved the criteria for material mastery in the aspect of reading letters, with a total of 28 letters mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results showed that RG did not increase in mastery of letters. That way, until the end of the study, RG mastered 28 of the 30 letters learned. The RG results on the aspect of reading letters are presented in Figure 5(c).

RR

The average correct response at baseline-1 for aspects of reading letters was 71%. After using the AR intervention, the average correct RR in answering questions began to increase quite a bit. During the intervention, the correct average for aspects of reading letters was 83%. RR achieved the criteria for material mastery in the aspect of reading letters, with a total of 27 letters mastered in the final intervention session. After that, a reassessment was conducted a few days after the intervention session. The results show an increase in RR mastery of letters, with the correct average for this aspect of reading letters is 91%. That way, until the end of the study, RR mastered 28 of the 30 letters learned. The results of the RR on the aspect of reading letters are presented in Figure 5(d).

Discussion

From the testing results conducted in the learning cycle, researchers see that students are enthusiastic in every part of the learning activity. Such as following the singing from the video being displayed, then clicking on the 3D object in the application, and following the sound made by the 3D object. During the learning process, students play an active role, such as asking questions and following directions in AR media. The teacher tries to create an exciting learning atmosphere to arouse students’ curiosity in every learning activity. The use of AR in learning is something new for the participants. Therefore, it is natural that technical problems occur during the learning process. Analysis of learning outcomes data was carried out to determine changes in the level of reading ability at the beginning of research participants while participating in a series of learning activities in this study. Analysis was carried out on each initial reading ability score achieved by students.

The results show that in letter reading skills, the improvements that occur are not visible. This is because initial reading learning at school starts with reading letters, so participants are already trained in this sub-skill. This is also supported by the fact that ASD children tend to be sensitive to details and patterns. This helps them identify and remember the letters and patterns involved in reading.

However, the ability to read words and syllables from the initial to the final baseline experienced a quite visible increase. This shows that using personalized AR to improve ASD students’ early reading skills, especially in reading words and syllables, helps improve ASD students’ early reading skills. AR can make learning more interactive and exciting (Howorth et al., 2019; Kolomoiets & Kassim, 2018; McMahon et al., 2016). ASD children may be more engaged in learning when it involves elements that catch their attention, such as moving pictures or sounds. Besides that, research conducted by Shemshack & Spector (2021), this personalized learning model is suitable for children with special needs, ensuring that schools accommodate the needs of students with different individual needs, interests, and goals. The research results of Basham, Hall, Jr., & Stahl (2016) also support this finding that personalized learning provides tremendous growth for students with special needs. In addition, the results of this study align with the findings of Rastegarmoghadam and Ziarati (2017) that the implementation of personalized learning is more effective with the use of technology to facilitate the organization of learning with a large number of students.

In Conclusion, this study provides an example of how AR technology is used to help children with ASD learn to read early. From the positive impact of using personalized learning model-based AR, in the future, we can explore how AR can meet each individual’s learning preferences by involving content presentation, level of difficulty, or other more interactive elements based on the learner’s profile. Besides that, consider the content provided so that it is appropriate for ASD children and that learning does not become frustrating. Likewise, assessments can be dynamically provided according to student learning progress and ensure continuous and personalized evaluation of skills and knowledge. To increase the participation of students with ASD in inclusive schools, for students who already have good reading skills, there needs to be a collaborative learning experience by sharing personalized content, interacting with each other, and collectively enhancing their understanding of the skills being studied. In addition, the learning outcomes analysis shows that this mobile AR application is exciting and effective in supporting early reading learning for children with ASD. However, more research is needed to expand and develop our findings about designing multiple representations and different types of interactions in AR technologies that are friendly to children with ASD.

Limitations

Although the study results showed an increase in the initial reading ability of ASD students after using an AR-assisted personalized learning-based curriculum, this cannot be separated from the limitations in its implementation. First, lack of teacher involvement in this research. The teacher is also not directly involved in this intervention, only as a companion to the participants. This is due to limited time for research making it impossible to carry out teacher training. Second, the material provided needs to be expanded, so students do not feel bored and can explore more material. Third, everyday environmental factors, the family, need to be analyzed so students can improve their reading skills by studying independently at home.

Implications for Future Research

In the context of learning to read, future research can further explore how reading skills can contribute to improving social interaction and writing skills in ASD children. Furthermore, how the use of reading skills can help children with ASD attend inclusive schools. Then, future research can continue to explore the effectiveness of AR-based learning models with a personalized approach. This includes exploring AR strategies and features that are most effective in understanding words, syllables, and letters and examining further impacts on learning outcomes in the context of reading skills. Overall, future research could involve further exploring the application of AR in the education of children with ASD and how this might specifically affect different aspects of their reading skills.

Footnotes

Acknowledgments

We would like to appreciate and thank PMDSU Scholarship under Direktorat Jenderal Pendidikan Tinggi, Kementerian Pendidikan dan Kebudayaan Republik Indonesia for support throughout this research, and special school’s teachers and students for lending their precious time aiding in the success of the research.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Kementerian Pendidikan, Kebudayaan, Riset, dan Teknologi under PMDSU.

ORCID iD

Azizah Nurul Khoirunnisa

Author Biographies

Azizah Nurul Khoirunnisa is a PhD candidate in curriculum development study program, at Universitas Pendidikan Indonesia. Her research focuses on multimedia in education, assistive technology, and personalized learning.

Munir is a Professor of Computer Science, Faculty of Mathematics and Sciences Education, Universitas Pendidikan Indonesia. He received a Ph.D degree in Information Science from Universiti Kebangsaan Malaysia in 2001 on ―Multimedia in Education for Literacy. His research interests are in Multimedia for Education, ICT for Education and e-learning.

Faaizah Shahbodin is a Professor of the Department of Interactive Media, Faculty of ICT, UTeM, Malaysia. Her research focuses on Multimedia in Education, Problem-based Learning, Assistive Technology, and Learning Technologies.

Laksmi Dewi is an Associate Prof in the Curriculum Development study program, at Universitas Pendidikan Indonesia. Her research focuses on curriculum development and educational technology.

References

Andri

Hazim Alkawaz

Bibo Sallow

(2018). Adoption of mobile augmented reality as a campus tour application. International Journal of Engineering & Technology, 7(4.11), 64–69. https://doi.org/10.14419/ijet.v7i4.11.20689

Antão

J. Y. F. d. L.

Abreu

L. C. de

Barbosa

R. T. d. A.

Crocetta

T. B.

Guarnieri

Massetti

Antunes

T. P. C.

Tonks

Monteiro

C. B. d. M.

(2020). Use of augmented reality with a motion-controlled game utilizing alphabet letters and numbers to improve performance and reaction time skills for people with autism spectrum disorder. Cyberpsychology, Behavior, and Social Networking, 23(1), 16–22. https://doi.org/10.1089/cyber.2019.0103

Åsberg Johnels

Carlsson

Norbury

Gillberg

Miniscalco

(2019). Current profiles and early predictors of reading skills in school-age children with autism spectrum disorders: A longitudinal, retrospective population study. Autism: The International Journal of Research and Practice, 23(6), 1449–1459. https://doi.org/10.1177/1362361318811153

Ayoub

Pulijala

(2019). The application of virtual reality and augmented reality in oral & maxillofacial surgery. BMC Oral Health, 19(1), 238. https://doi.org/10.1186/s12903-019-0937-8

Bartoli

Garzotto

Gelsomini

Oliveto

Valoriani

(2014). Designing and evaluating touchless playful interaction for ASD children. In Proceedings of the 2014 Conference on Interaction Design and Children (pp. 17–26). ACM Digital Library. https://doi.org/10.1145/2593968.2593976

Berenguer

Baixauli

Gómez

Andrés

M. d. E. P.

De Stasio

(2020). Exploring the impact of augmented reality in children and adolescents with autism spectrum disorder: A systematic review. In International Journal of Environmental Research and Public Health (Vol. 17(17), 6143–6215). MDPI AG. https://doi.org/10.3390/ijerph17176143

Bethune

K. S.

Wood

C. L.

(2013). Effects of wh-question graphic organizers on reading comprehension skills of students with autism spectrum disorders. Education and Training in Autism and Developmental Disabilitites, 48(2), 236–244. https://www-jstor-org-443.web.bisu.edu.cn/stable/23880642

Brown

H. M.

Oram-Cardy

Johnson

(2013). A meta-analysis of the reading comprehension skills of individuals on the autism spectrum. Journal of Autism and Developmental Disorders, 43(4), 932–955. https://doi.org/10.1007/s10803-012-1638-1

Coleman-Martin

M. B.

Heller

K. W.

Cihak

D. F.

Irvine

K. L.

(2005). Using computer-assisted instruction and the nonverbal reading approach to teach word identification. Focus on Autism and Other Developmental Disabilities, 20(2), 80–90. https://doi.org/10.1177/10883576050200020401

10.

Danaei

Jamali

H. R.

Mansourian

Rastegarpour

(2020). Comparing reading comprehension between children reading augmented reality and print storybooks. Computers & Education, 153(6), 103900. https://doi.org/10.1016/j.compedu.2020.103900

11.

Davidson

M. M.

Ellis Weismer

(2014). Characterization and prediction of early reading abilities in children on the autism spectrum. Journal of Autism and Developmental Disorders, 44(4), 828–845. https://doi.org/10.1007/s10803-013-1936-2

12.

Djauzak

(1996). Metodik khusus pengajaran bahasa Indonesia di Sekolah dasar. Dirjen Pendidikan Dasar Departemen Pendidikan Nasional.

13.

Gilakjani

A. P.

Sabouri

N. B.

(2016). Learners’ listening comprehension difficulties in English language learning: A literature review. English Language Teaching, 9(6), 123. https://doi.org/10.5539/elt.v9n6p123

14.

Gove

Wetterberg

(2011). The early grade reading assessment: Applications and interventions to improve basic literacy. RTI Press. https://doi.org/10.3768/rtipress.2011.bk.0007.1109

15.

Grindle

C. F.

Carl Hughes

Saville

Huxley

Hastings

R. P.

(2013). Teaching early reading skills to children with autism using mimiosprout early reading. Behavioral Interventions, 28(3), 203–224. https://doi.org/10.1002/bin.1364

16.

Hashim

H. U.

Yunus

M. M.

Norman

(2022). ‘AReal-Vocab’: An augmented reality English vocabulary mobile application to cater to mild autism children in response towards sustainable education for children with disabilities. Sustainability, 14(8), 4831. https://doi.org/10.3390/su14084831

17.

Howorth

S. K.

Rooks-Ellis

Flanagan

M. W.

(2019). Augmented reality supporting reading skills of students with autism spectrum disorder. Intervention in School and Clinic, 55(2), 71–77. https://doi.org/10.1177/1053451219837635

18.

Jackson

E. M.

Hanline

M. F.

(2020). Using a concept map with RECALL to increase the comprehension of science texts for children with autism. Focus on Autism and Other Developmental Disabilities, 35(2), 90–100. https://doi.org/10.1177/1088357619889933

19.

Kaya

(2015). The role of reading skills on reading comprehension ability of Turkish EFL students. Üniversitepark Bülten, 4(1-2), 37–51. https://doi.org/10.12973/unibulletin.412.4

20.

Khoirunnisa

A. N.

Munir Dewi

(2023). Design and prototype development of augmented reality in reading learning for autism. Computers, 12(3), 55. https://doi.org/10.3390/computers12030055

21.

Kinney

E. M.

Vedora

Stromer

(2003). Computer-presented video models to teach generative spelling to a child with an autism spectrum disorder. Journal of Positive Behavior Interventions, 5(1), 22–29. https://doi.org/10.1177/10983007030050010301

22.

Kolomoiets

T. H.

Kassim

D. A.

(2018). Using the augmented reality to teach of global reading of preschoolers with autism Spectrum disorders. Educational Dimension, 51(304–315), 304–315. https://doi.org/10.31812/pedag.v51i0.3678

23.

Kurniawati

L. A.

(2020). Personalized learning: Empowering a student with autism spectrum disorder in basic grammar course. Yavana Bhasha: Journal of English Language Education, 1(1), 1. https://doi.org/10.25078/yb.v1i1.1372

24.

Lord

Elsabbagh

Baird

Veenstra-Vanderweele

(2018). Autism spectrum disorder. Lancet (London, England), 392(10146), 508–520. https://doi.org/10.1016/S0140-6736(18)31129-2

25.

Maglione

M. A.

Gans

Das

Timbie

Kasari

(2012). Nonmedical interventions for children with ASD: Recommended guidelines and further research needs. Pediatrics, 130(Suppl 2), S169–S178. https://doi.org/10.1542/peds.2012-0900O

26.

Mart

C. T.

(2012). Developing speaking skills through reading. International Journal of English Linguistics, 2(6), 91–96. https://doi.org/10.5539/ijel.v2n6p91

27.

McIntyre

N. S.

Solari

E. J.

Gonzales

J. E.

Solomon

Lerro

L. E.

Novotny

Oswald

T. M.

Mundy

P. C.

(2017). The scope and nature of reading comprehension impairments in school-aged children with higher-functioning autism spectrum disorder. Journal of Autism and Developmental Disorders, 47(9), 2838–2860. https://doi.org/10.1007/s10803-017-3209-y

28.

McIntyre

N. S.

Solari

E. J.

Grimm

R. P.

E Lerro

E Gonzales

Mundy

P. C.

(2017). A comprehensive examination of reading heterogeneity in students with high functioning autism: Distinct reading profiles and their relation to autism symptom severity. Journal of Autism and Developmental Disorders, 47(4), 1086–1101. https://doi.org/10.1007/s10803-017-3029-0

29.

McKissick

B. R.

Davis

L. L.

Spooner

Fisher

L. B.

Graves

(2018). Using computer-assisted instruction to teach science vocabulary to students with autism spectrum disorder and intellectual disability. Rural Special Education Quarterly, 37(4), 207–218. https://doi.org/10.1177/8756870518784270

30.

McMahon

D. D.

Cihak

D. F.

Wright

R. E.

Bell

S. M.

(2016). Augmented reality for teaching science vocabulary to postsecondary education students with intellectual disabilities and autism. Journal of Research on Technology in Education, 48(1), 38–56. https://doi.org/10.1080/15391523.2015.1103149

31.

Mohd

C. K. N. C. K.

Shahbodin

Suparjoh

Khidir

N. A. S. M.

(2019). Application of augmented reality in learning bakery for autism spectrum disorder. International Journal of Engineering and Advanced Technology, 9(1), 2616–2620. https://doi.org/10.35940/ijeat.A9853.109119

32.

Moore

David

Cheng

Yufang

McGrath

Paul

Powell

Norman J.

(2005). Collaborative Virtual Environment Technology for People With Autism. Focus on Autism and Other Developmental Disabilities, 20(4), 231–243. https://doi.org/10.1177/10883576050200040501

33.

Muammar

Mustadi

(2018). Model pembelajaran keterampilan berbicara berbasis komunikatif bagi siswa sekolah dasar: Teori dan praktik. Sanabil Publishing.

34.

Munir

Setiawan

Nugroho

E. P.

Kusnendar

Wibawa

A. P.

(2018). The effectiveness of multimedia in education for special education (MESE) to improve reading ability and memorizing for children with intellectual disability. International Journal of Emerging Technologies in Learning, 13(8), 254–263. https://doi.org/10.3991/ijet.v13i08.8291

35.

Munir Kaosar

R. N.

Rasim Murtadha

Shahbodin

Riza

L. S.

(2021). Expert system using the educational game to determine children’s autism levels using forward chaining. Linguistics and Culture Review, 5(S1), 1149–1172. https://doi.org/10.37028/lingcure.v5nS1.1499

36.

Nally

Healy

Holloway

Lydon

(2018). An analysis of reading abilities in children with autism spectrum disorders. Research in Autism Spectrum Disorders, 47(3), 14–25. https://doi.org/10.1016/j.rasd.2017.12.002

37.

Nation

Clarke

Wright

Williams

(2006). Patterns of reading ability in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 36(7), 911–919. https://doi.org/10.1007/s10803-006-0130-1

38.

Nilsen

E. S.

Bacso

S. A.

(2017). Cognitive and behavioural predictors of adolescents’ communicative perspective‐taking and social relationships. Journal of Adolescence, 56(1), 52–63. https://doi.org/10.1016/j.adolescence.2017.01.004

39.

Omar

Bidin

(2015). The impact of multimedia graphic and text with autistic learners in reading. Universal Journal of Educational Research, 3(12), 989–996. https://doi.org/10.13189/ujer.2015.031206

40.

Padeliadu

Antoniou

(2014). The relationship between reading comprehension, decoding, and fluency in Greek: A cross-sectional study. Reading & Writing Quarterly, 30(1), 1–31. https://doi.org/10.1080/10573569.2013.758932

41.

Paramesti

V. M.

Shahbodin

Munir Riza

L. S.

(2021). Multimedia to reading and counting for autistic children using aba and pecs methods. Turkish Journal of Physiotherapy and Rehabilitation, 32(2), 2856–2885. https://www.turkjphysiotherrehabil.org

42.

Petersen

(2016). Perspectives of special education teachers on general education curriculum access. Research and Practice for Persons with Severe Disabilities, 41(1), 19–35. https://doi.org/10.1177/1540796915604835

43.

Pressman

R. S.

(2010). Software engineering: A practitioner’s approach. Pressman. https://www.mhhe.com/pressman

44.

Rastegarmoghadam

Ziarati

(2017). Improved modeling of intelligent tutoring systems using ant colony optimization. Education and Information Technologies, 22(3), 1067–1087. https://doi.org/10.1007/s10639-016-9472-2

45.

Rayner

Pollatsek

Ashby

Clifton

Jr. (2012). Psychology of reading (2nd ed.). Routledge.

46.

Reutebuch

C. K.

El Zein

Kim

M. K.

Weinberg

A. N.

Vaughn

(2015). Investigating a reading comprehension intervention for high school students with autism spectrum disorder: A pilot study. Research in Autism Spectrum Disorders, 9(11), 96–111. https://doi.org/10.1016/j.rasd.2014.10.002

47.

Richards

J. C.

Burns

(2012). The Cambridge guide to pedagogy and practice in second language teaching. In Burns

Richards

J. C.

(Eds.). Cambridge University Press. https://doi.org/10.1017/9781009024778

48.

Safitri

Yusra

D. S.

Hermawan

Ripmiatin

Pradani

(2017). Mobile tourism application using augmented reality. In 2017 5th International Conference on Cyber and IT Service Management (CITSM) (pp. 1–6). IEEE Xplore. https://doi.org/10.1109/CITSM.2017.8089305

49.

Shemshack

Kinshuk Spector

J. M.

(2021). A comprehensive analysis of personalized learning components. Journal of Computers in Education, 8(4), 485–503. https://doi.org/10.1007/s40692-021-00188-7

50.

Singh

B. D.

Moore

D. W.

Furlonger

B. E.

Anderson

Busacca

M. L.

English

D. L.

(2017). Teaching reading comprehension skills to a child with autism using behaviour skills training. Journal of Autism and Developmental Disorders, 47(10), 3049–3058. https://doi.org/10.1007/s10803-017-3229-7

51.

Squire

Klopfer

(2007). Augmented reality simulations on handheld computers. The Journal of the Learning Sciences, 16(3), 371–413. https://doi.org/10.1080/10508400701413435

52.

Tang

T. Y.

Winoto

(2019). Automatic object recognition in a light-weight augmented reality-based vocabulary learning application for children with autism. In Proceedings of the 2019 3rd International Conference on Innovation in Artificial Intelligence (pp. 65–68). ACM Digital Library. https://doi.org/10.1145/3319921.3319945

53.

Thompson

J. L.

Plavnick

J. B.

Skibbe

L. E.

Bak

M. Y. S.

Doher

(2022). Beyond tier three: Individualized levels of support during Headsprout® early reading instruction for children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 52(5), 2132–2148. https://doi.org/10.1007/s10803-021-05072-y

54.

Wang

Ong

S. K.

Nee

A. Y. C.

(2018). Enhancing mechanisms education through interaction with augmented reality simulation. Computer Applications in Engineering Education, 26(5), 1552–1564. https://doi.org/10.1002/cae.21951

55.

Williamson

Carnahan

C. R.

Jacobs

J. A.

(2012). Reading comprehension profiles of high-functioning students on the autism spectrum: A grounded theory. Exceptional Children, 78(4), 449–469. https://doi.org/10.1177/001440291207800404

56.

Wong

K.-T.

Hanafi

H. F. B.

Abdullah

Noh

N. M.

Hamzah

M. B.

(2019). A prototype of augmented reality animation (ARA) E-courseware: An assistive technology to assist autism spectrum disorders (Asd) students master in basic living skills. International Journal of Innovative Technology and Exploring Engineering, 9(1), 3487–3492. https://doi.org/10.35940/ijitee.A4962.119119

57.

Zajic

M. C.

Solari

E. J.

Grimm

R. P.

McIntyre

N. S.

Mundy

P. C.

(2020). Relationships between reading profiles and narrative writing abilities in school-age children with autism spectrum disorder. Reading and Writing, 33(6), 1531–1556. https://doi.org/10.1007/s11145-020-10015-7

58.

Zhou

Zhang

(2021). Study on the personalized learning model of learner-learning resource matching. International Journal of Information and Education Technology, 11(3), 143–147. https://doi.org/10.18178/ijiet.2021.11.3.1503

59.

Zuccarello

Di Blasi

F. D.

Zingale

Panerai

Finocchiaro

Trubia

Buono

Zoccolotti

(2015). Reading decoding and comprehension in children with autism spectrum disorders: Evidence from a language with regular orthography. Research in Autism Spectrum Disorders, 17(13), 126–134. https://doi.org/10.1016/j.rasd.2015.06.013