ADIDAS: An Examined Approach for Enhancing Cognitive Load and Attitudes towards Synchronous Digital Learning Amid and Post COVID-19 Pandemic

SARS-CoV-2 (COVID-19) has disrupted university education and turned it into distance learning for at least one semester in many countries, including the Kingdom of Saudi Arabia (KSA). However, there was an issue with university students’ cognitive load at this critical time, because education totally stopped for about a month and then resumed remotely. This research draws on the cognitive load theory, particularly the extraneous load, to develop an instructional design model called ADIDAS. The model includes six stages, namely: analyse (A), design (D), improve (I), do (D), Assess (A), and Share (S). Thirty-four experts in instructional technology models have reviewed the ADIDAS model in Arab university contexts, producing a consensus about its suitability for use in distance learning amid the COVID-19 pandemic. Following the consensus of the experts, the model was applied to a sample of 527 students at King Faisal University, KSA. The results confirmed significant statistical differences with a very large effect size in relation to the attitude towards synchronous digital learning (SDL) and cognitive load pre and post ADIDAS. Students had a positive attitude towards SDL and a low cognitive load during the educational process pre adoption of the ADIDAS model, compared to post ADIDAS. The current research results have numerous implications for post the COVID-19 pandemic, especially in Arab countries and similar contexts.


Introduction
Due to the impact of SARS-CoV-2 (COVID- 19), education in the Kingdom of Saudi Arabia (KSA), like many other nations worldwide, has shifted from traditional learning, i.e., face-to-face, to synchronous digital learning (SDL), with virtual classrooms using numerous tools, e.g., Blackboard Collaborate, Microsoft Teams, Zoom, and social media in some cases [1][2][3]. Before COVID-19, in the KSA, utilizing SDL in the higher education context was limited to some courses [4,5]. Furthermore, numerous studies have highlighted the various benefits of SDL: enhancing learning outcomes, learning satisfaction, learning performance, learning motivation, and positive attitudes towards education [6][7][8]. Amid COVID-19, in the first quarter of 2020, in a short time, the Ministry of Education in the KSA instructed all universities to turn all traditional classroom courses from a face-to-face setting to an online one [9]. Due to the sudden shift to SDL, educators were forced to utilise SDL as the sole tool for teaching and learning communication with their students [10]. Thus, this became a daunting job for educators who had presented and developed their courses remotely with individual efforts [11].
Accordingly, digital learning was expected to significantly improve cognitive achievement, skills, competencies, attitudes, and learning outcomes for higher education students Instructional design is the science and art of developing, evaluating, and maintaining learning situations that facilitate and improve teaching performance using IDMs [37]. In addition, numerous studies have benefited from instructional design during the COVID-19 pandemic [32,35]. Hence, adopting instructional design models is still one of the best practices to be mastered by higher education educators, especially during online learning amid global health emergencies, i.e., COVID-19 [38,39]. The approach of educational design has gradually evolved, observing a shift from a traditional prescriptive, normative definition to the current one based on diverse systems that are antipathetic to unique recipes and stringent specifications [35]. IDMs direct organized planning for learning elements (contents, activities, assessments, and others) [39]. Additionally, IDMs need a symmetry of sense and intuition, momentum to perform, and an ability to reflect on the accepted actions [29]. In general, all instructional design approaches depend on (at least) five significant steps: (1) analysis of the setting and student needs; (2) design of a bunch of specifications for an influential and applicable learning environment; (3) development of contents, activities, assess, and learners; (4) implementation of learning methods and strategies; (5) evaluation of all results [34,36,40]. In typical situations, besides the educators, the implementation of instructional design requires the presence of at least two people: (1) the instructional designer who designs the storyboard and (2) the e-learning developer who converts the developed storyboard into a product [34].
Many articles have examined the multiple IDMs that educators can use to deliver SDL amid COVID-19 [32,35,37]. According to Sangsawang [32], university educators' adoption of IDMs could prioritise their students' needs, feelings, and challenges with the IDMs' designs during the transition amid COVID-19. According to Wang [37], utilising an instructional design model enhances adaptability and good planning, emphasising what it takes for an educator to serve their students. According to Hanafi, Yusuf et al. [41] and Xie [42], university educators' adoption of IDMs improved students' satisfaction with distance learning and enriched courses through extra activities. Though several studies have been conducted on IDMs and cognitive load among higher education students, especially in the context of Arab counties such as KSA, a lack of studies have been undertaken regarding educators' utilisation of instructional design to address cognitive load among higher education students during synchronous digital learning amid COVID-19.
The current study has two main objectives. First, it develops a new instructional design model for higher education educators to create positive attitudes towards SDL and enhance cognitive load amid global health emergencies, i.e., COVID-19. This novel instructional model is called ADIDAS, which will be discussed later, and has been examined by experts. Second, the study also examines the newly developed model with a sample of undergraduate students and compares their attitudes towards SDL and cognitive load pre and post ADIDAS. Therefore, the current study endeavours to answer two main questions. First, what is the ADIDAS model structure that educators can adopt for SDL amid emergencies, i.e., COVID-19? Second, to what extent does the ADIDAS model make a difference in attitudes towards SDL and cognitive load among higher education students amid emergencies, i.e., COVID-19, compared to pre-pandemic circumstances?
Our hypotheses are that (1) there are significant differences in cognitive load among higher education students, pre and post ADIDAS-model adoption amid and (2) there are significant differences in attitudes towards SDL among higher education students, pre and post ADIDAS-model adoption amid COVID-19.

Development of ADIDAS Model
To develop a draft of the ADIDAS model and answer the first research question, the literature review related to digital learning, especially amid COVID-19, instructional design models, and cognitive load as well as learning theories from various databases that we have reviewed and analysed (see Appendix A). After screening the literature, we focused our analysis on the ADDIE model [43], interaction analysis model (IAM) [44], and model of self-regulated learning (SRL) [45] to develop the first draft of the ADIDAS model.
The ADDIE model is a systematic, practical heuristic framework for synchronous online learning course development, which opens perfect possibilities and gives good outcomes. The IAM is one of the most frequently used instruments in the study of knowledge construction, and the extent of its use makes it one of the most coherent. Models of SRL depict learning as an activity that a learner self-regulates, whether learning alone (unsupervised) or in the presence of instructors or peers (supervised). These models were chosen to develop the first draft of ADIDAS. These models helped us to develop the ADI-DAS model to achieve SDL goals, making it more interactive. They helped us apply the educational theories through SDL and optimal investment of learning elements, making the learner focus and rely on their efforts, raising the learner's motivation, and providing sufficient space for the learner to interact with the learning elements. This motivates the learner to be creative and innovate and, thus, able to achieve comprehensive evaluation. Additionally, the ADIDAS model was based on the cognitive load theory [26]. Cognitive load theory ensures that learners acquire sufficient information and have secure dealing with novel information not to cause cognitive overload. It also considered Piaget's cognitive constructivist theory [45] and the sociocultural constructivist learning theory [46].
The ADIDAS model was directed to 50 expert specialists in instructional technology, digital learning, cognitive psychology, and information technology in Middle Eastern countries. The purpose of this was to examine the validity of utilizing SDL. It is worth noting that responses were collected from only 34 experts, as 16 did not respond. The profile of these respondents is shown in Table 1. Experts were identified through personal networks and recommendations from different colleagues. The procedures yielded 49 items with five dimensions, guiding the learning practices that help educators to provide SDL with low cognitive load. Each dimension of six constructs contained four factors: D1: learners/recipients (L), D2: the content of learning (C), D3: technology/apps (T), D4: evaluation (E), and D5: reviewing/modification (R). The 49 items that were included within the ADIDAS model for representing six constructs: Analyse (A) (9 items, α = 0.854); Design (D) (8 items, α = 0.862); Improve (I) (8 items, α = 0.799); Do (D) (10 items, α = 0.857); Assess (A) (9 items, α = 0.7986); Share (S) (5 items, α = 0.814). (See Figure 1 and Table 2). In Table 2, we explain the dimensions of the models and factors of each dimension as well as the items (with their original sources). Do (D) (10 items, α = 0.857); Assess (A) (9 items, α = 0.7986); Share (S) (5 items, α = 0.814). (See Figure 1 and Table 2). In Table 2, we explain the dimensions of the models and factors of each dimension as well as the items (with their original sources).  The ADIDAS model was circulated via experts' private e-mails and social media on 1 October 2020 and maintained for three weeks. Day to day, the investigators reviewed and observed the responses. Cronbach's alpha coefficient was used to assess the ADIDAS factors' reliability.

Cognitive Load Scale
The procedures yielded 16 items with three factors measuring the students' cognitive load. The scale showed good reliability and was derived from [29,30,48]. The 16 items that comprised the questionnaire for the study represent three factors: main cognitive load (6 items, α = 0.871), extraneous cognitive load (5 items, α = 0.793), and closely related cognitive load (5 items, α = 0.801). Each item was operationalised on a five-interval Likert scale, with students selecting one of five options to indicate the degree to which they reflect their cognitive load.

Attitude Scale
The procedures yielded 10 items with three factors measuring the students' attitudes towards synchronous digital learning (SDL). The scale showed good reliability and was derived from [49,50]. The 10 items that comprised the questionnaire for the study represent three factors: knowledge development (3 items, α = 0.779), skills development (3 items, α = 0.832), and learning attitudes (4 items, α = 0.824). Each item was operationalised on a five-interval Likert scale, with students selecting one of five options to indicate the degree to which they reflect their attitude.

Research Population and Sample
The research population included all university students enlisted in King Faisal University (KFU) colleges in Al-Ahsaa, Eastern Province, KSA. The colleges at KFU relied considerably on online platforms and virtual classrooms to manage content, lectures, and exams amid the COVID-19 pandemic. Thirteen educators at King Faisal University were trained to utilise the ADIDAS model for SDL. The research team targeted 600 participating students, concerning their perceptions of cognitive load and attitudes towards SDL before and after using the ADIDAS model. According to Hill [51], the sample size calculation must be based on the total number of items, which should be at least five responses for each item. The items used in the current study were 49; hence, the sample should not be less than 245 responses. Furthermore, Muthén [52] added that a sample should be more than 150. In this research, our sample size was appropriate, since there were 527 valid responses for analysis.
In total, 527 valid responses from students were received. Most students in the group were female (77.61% females, 22.39% males). Most respondents were between 18 and 20 (98.4%), and 6.7% of students had never used technology as an educational tool.
The educators provided the surveys to students via their private networks, i.e., What-sApp, email, etc.. There was no power bias or authority over students. They were informed that the survey was just for scientific research and that their responses would be unidentified. Participants were voluntary and unnamed, and all the essential safeguards were utilised on site to assure data confidentiality. All personally identifiable information about participants was removed from the publicly available analysis, to ensure that answers could not be recognised. Further, distinct items such as name, age, etc., were optional.

Data Collection and Analysis
This research adopted a quantitative research methodology to develop the model and examine its effects on students' cognitive load and attitudes towards SDL. The experts' responses regarding the model were analysed and are presented in Table 1. Descriptive statistics were used to analyse the profile as well as cognitive load and attitudes towards SDL items. The responses of students, in relation to students' cognitive load and attitudes towards SDL before and after the adoption of the ADIDAS model, were analysed by paired sample t-test using SPSS version 25. Eta squared was adopted to test the effect size. This gives a sign of the size of the variances between pre-and post-ADIDAS model adoption.

Students' Cognitive Load
To answer the second research question and examine the first research hypothesis, a comparison was made between pre-and post-ADIDAS adoption. First, descriptive statistics was adopted to analyse this using mean and standard deviation ( Table 3). The results showed that the mean for pre-model adoption was between 1.403 (S.D. 0.586) and 1.504 (S.D. 0.616). On the other side, the mean for post-model adoption was between 4.184 (S.D. 0.917) and 4.534 (S.D. 0.784). These results show a significant difference between the pre-and post-ADIDAS model, which will be examined using the paired sample t-test.  Table 4 compares the three domains of cognitive load (CL): main (CL), extraneous (CL), and closely related (CL) pre and post implementation of the ADIDAS model. The results of the paired sample t-test showed significant differences in the three cognitive loads: main (CL), extraneous (CL), and closely related (CL) between pre-model and post-model adoption. Students achieved better results and had lower levels in the three domains of cognitive load after the implementation of the ADIDAS model. Students' cognitive load (CL) pre-model adoption has the information presented density, traditional teaching methods, provided unnecessary information, and non-related activities; these did not contribute to the learning process. The effect size was very large, as confirmed by eta squared (Table 4); this means that the difference between pre-and post-model adoption was very large. Indeed, the better results were for post-model adoption (see Table 4). This supports the first research hypothesis (H1).

Students' Attitude towards Synchronous Digital Learning
To answer the second research question and examine the second research hypothesis, a comparison was undertaken between pre-and post-model adoption. First, descriptive statistics was adopted to analyse this using mean and standard deviation ( Table 5 . 0. 893). These results show a significant difference between preand post-model adoption, which will be examined using the paired sample t-test.  Table 6 compares attitudes towards synchronous digital learning (SDL): knowledge development, skills development, and learning attitudes before and after the implementation of the ADIDAS model. The paired sample t-test showed significant differences in students' attitudes towards SDL between pre-model and post-model adoptions. After implementing the ADIDAS model, students' attitude synchronous digital learning SDL was positive compared to pre-model adoption. There was a lot of improvement in knowledge development, skills development, and learning attitude post-model adoption, compared to pre-model adoption. The effect size was very large, as confirmed by eta squared (Table 6); this means that the difference between pre-and post-model adoption was very large. Indeed, better results in attitude towards SDL were posted for the model adoption. This supports the second research hypothesis (H2).

Discussions
The current study was set to achieve two main objectives. Firstly, it was set to develop an instructional design model to create positive attitudes towards SDL and enhance cognitive load among higher education students amid COVID-19. Secondly, it examined the newly developed model with a sample of undergraduate students and compared their attitudes towards SDL and cognitive load pre and post adoption of the new model. The study was conducted on undergraduate students at King Faisal University, Saudi Arabia, amid the COVID-19 pandemic. Overall, the results showed significant statistical differences in cognitive load and attitude towards SDL post adoption of the ADIDAS model compared to pre-model adoption. These results are consistent with previous studies [25,53], which show that cognitive load and attitude towards SDL depend on the chosen method of presenting information, student motivation, and involvement in the learning process.
The results confirmed that SDL amid the COVID-19 pandemic, as provided by the ADI-DAS model, enriched the educational environment. The ADIDAS model provided appropriate methods for presenting and organising content, which reduced students' cognitive load and increased levels of mental achievement, consistent with previous studies [25,47]. The ADIDAS model was based on the ADDIE model, which has been built according to the cognitive load theory; the information display methods and diversity of SDL gave the learners the necessary information to exclude redundant and repetitive activities unrelated to the content, which are regular in studies [28,31]. Furthermore, the transformation of learners from mere recipients of information contributed to alleviating the accidental cognitive load on working memory, increasing the learning process, and continuing to focus their attention. This increased memory capacity and facilitated understanding of the information presented to the students, keeping it in their memory, which is similar to the results of other studies [27,53]. Moreover, the ADIDAS model was designed according to Piaget's cognitive constructivist theory principles and modern instructional models, which were created during COVID-19, that ensured effective learning and enhanced mental load distribution during SDL. Accordingly, the students had low cognitive load levels while learning within the ADIDAS model [43,44].
Additionally, the results show that the students taught by the ADIDAS model intend to use this learning type in the future, which agrees with other studies [28,48]. Therefore, the ADIDAS model has drawn on the interaction analysis model (IAM), which can explain why distance learning attracts students' attention, offers an effective learning environment, and increases students' motivation to learn the topics, all of which is compatible with other studies [27,30]. Likewise, the underlying reason that turned students' attitudes positive may be that students came across a different education style other than the traditional SDL and interacted with their educators, peers, and content, which matches with other studies [29,47]. The ADIDAS model enabled students to engage more in an interactive learning environment with learners, content, evaluation, and technology, which is consistent with sociocultural constructivist learning theory [32,54]. Moreover, the ADIDAS model has been built according to self-regulated learning (SRL) models, which helped to direct the students' attention selectively by providing more engaging and interactive learning elements that increased the positive attitudes among students, which is compatible with previous studies [36,37].
The current article includes some limitations that could be handled in future exertions, including the relationship between synchronous digital learning (SDL) sustainability and cognitive load and attitudes towards digital learning among higher education students in the KSA. This would include that the data were assembled solely from a tiny sample of students in higher education institutions in the KSA. The findings' generalisability to elsewhere in the Gulf, the Middle East, or another geographical location should be approached with caution. Likewise, this article applied the quantitative analysis method, so future research could integrate qualitative and quantitative approaches to discover further reasons for and associations between the suggested factors. In addition, the effect of other mediating and moderating variables (students' gender or educators' experience, competencies, skills, digitalisation, etc.) can be combined in future research.

Conclusions
The present research developed a new instructional design model, entitled ADIDAS, and reviewed it with experts in the Arab countries' context. The model has been provided to higher education educators as a guide for delivering SDL during emergencies, i.e., amid the COVID-19 pandemic. The model was applied for SDL amid the COVID-19 pandemic on 527 students in the colleges of KFU, KSA. The results showed that the students' cognitive load and attitudes towards SDL were better after adopting the ADIDAS model. There were significant differences pre-and post-model adoption, and the size difference between pre-and post-model adoption was very large. The results for post-model adoption were better. The results confirmed a lower level of cognitive load and a positive attitude towards the use of SDL after the ADIDAS model. The use of instructional design models such as the ADIDAS model for SDL contributes to digital learning sustainability amid the COVID-19 pandemic.  Data Availability Statement: Data are available upon request from researchers who meet the eligibility criteria. Kindly contact the first author privately through email.

Conflicts of Interest:
The authors declare no conflict of interest.

Appendix A PRISMA Flow Diagram
Inclusion/exclusion criteria: Please note that the final review included articles published in English and excluded articles published in Arabic. Additionally, it included articles related to online learning in undergraduate higher education and excluded articles about other educational grades. Likewise, it included articles that related to the ADDIE model, interaction analysis model (IAM), and models of self-regulated learning (SRL), while excluding instructional design models. Hence, it included articles on using instructional design models, while excluding others. Furthermore, it included articles on enhancing cognitive load in online learning, while excluding others. Furthermore, it included articles on cognitive load theory, Piaget's cognitive constructivist theory, and the sociocultural constructivist learning theory, while excluding other learning theories' articles.
Data Availability Statement: Data are available upon request from researchers who meet the eligibility criteria. Kindly contact the first author privately through email.

Conflicts of Interest:
The authors declare no conflicts of interest. Inclusion/exclusion criteria: Please note that the final review included articles published in English and excluded articles published in Arabic. Additionally, it included articles related to online learning in undergraduate higher education and excluded articles about other educational grades. Likewise, it included articles that related to the ADDIE model, interaction analysis model (IAM), and models of self-regulated learning (SRL), while excluding instructional design models. Hence, it included articles on using instructional design models, while excluding others. Furthermore, it included articles on enhancing cognitive load in online learning, while excluding others. Furthermore, it included articles on cognitive load theory, Piaget's cognitive constructivist theory, and the sociocultural constructivist learning theory, while excluding other learning theories' articles.