Beyond Compliance: A Practical Guide for Accessible Adaptive Learning with AI in ADDIE

Learners bring a wide range of accessibility requirements to their learning environments, shaped by diverse physical, sensory, cognitive, and technological needs. While accessibility guidelines and universal design frameworks offer broad recommendations, they do not always account for the specific values or lived experiences of different societal groups in need of accommodations. A review of the literature reveals a rich discussion of accessibility best practices, yet there remains limited guidance on how to reflect the ethical considerations and nuanced preferences that various disability communities hold for specific accessibility features, particularly during times of emergency remote instruction, where accessibility gaps were magnified (Russ & Hamidi, 2021).

For example, Deaf and hard-of-hearing (DHH) users in multiple studies reported wanting prosodic information added to captions, and experimental comparisons show tone-inclusive captions can be preferred and judged more informative (Pataca et al., 2023). Similarly, students from low-income backgrounds may value downloadable or low-bandwidth course materials that allow offline access, as they may not always have stable internet or up-to-date devices at home (Vardeh, 2023). In such cases, accessibility is not only about disability compliance but also about honoring learners' lived realities and promoting equitable participation across socioeconomic lines.

This preference is rooted not only in usability but in cultural values tied to representation and trust in digital content. Such distinctions illustrate how accessibility is not simply a checklist, but a value-informed process shaped by the needs of diverse learners. However, instructional designers often lack tools to intentionally surface and address these context-dependent preferences during the early stages of design.

The ADDIE framework offers a useful structure for this work. In its Analyze phase, instructional designers investigate the learner population, context, and constraints to identify both needs and barriers. In the Design phase, these insights are translated into objectives, strategies, and formats that anticipate and integrate accessibility from the outset. As artificial intelligence becomes more integrated into the instructional design process (Luo et al., 2024), its potential to support value-sensitive reflection is worth exploring. Generative AI tools such as ChatGPT can serve as reflective partners, prompting designers to interrogate their assumptions, identify whose needs are being centered, and brainstorm design solutions that prioritize inclusive values. When used during the Analyze and Design phases of the ADDIE model, these tools can help instructional designers go beyond compliance to proactively anticipate accessibility barriers and align learning experiences with the specific values of their learners.

To support this deeper, values-informed approach to accessibility, this chapter introduces the Anticipatory Accessibility Reflection Guide (AARG), a practitioner-focused tool designed to assist instructional designers in embedding accessibility considerations from the outset of their design work. Grounded in Value Sensitive Design (VSD), the guide offers structured reflection prompts, AI-assisted ideation strategies, and checkpoints specifically suited for adaptive and branching learning contexts. The AARG is designed to support professionals such as instructional designers, higher education faculty, course developers, educational technologists, accessibility specialists, and design researchers.

This chapter provides an overview of its purpose. In the sections that follow, the chapter defines key terms, outlines the theoretical foundations of the guide, reviews relevant literature, and illustrates how the AARG can be applied during the Analyze and Design phases of the ADDIE model. Additional resources are provided for those seeking to deepen their engagement within accessibility design practices.

Content Discussion

Introduction

To apply the AARG with confidence, it’s significant to understand the foundational frameworks that shape its design. This section provides an overview of the ADDIE instructional design model, the ethical grounding of Value Sensitive Design theory, and the evolving role of artificial intelligence in instructional design. This section establishes the theoretical and practical underpinnings necessary to contextualize how AARG supports proactive, value-informed accessibility decisions during the Analyze and Design phases. By unpacking these elements, readers can better understand how course architecture, value reflection, and AI-supported ideations intersect to inform early design choices. This foundation ensures that the guide is not used in isolation, but rather as a practical extension of established models that collectively support accessible, value-informed learning environments.

Overview of the ADDIE Model

Originally developed by the U.S. military in the 1970s, ADDIE has since been used as a design model for instructional designers conducting work in various disciplines (Kadakia & Owens, 2020). The ADDIE model framework allows designers to effectively develop modules, simulations, microlearning, or programs for learning experiences by following its iterative workflow process. The acronym ADDIE stands for Analyze, Design, Develop, Implement, and Evaluate. See Figure 1 for an overview of each phase.

Figure 1

An Overview of the ADDIE Phases

Due to its widespread adoption, enduring relevance, and adaptability across diverse instructional design contexts, the ADDIE framework was selected for the development of the Anticipatory Accessibility Reflection Guide. As Szabo (2022) explains, ADDIE has become especially vital in online learning environments, where systematic planning and adaptation are necessary to ensure andragogical effectiveness and learner engagement. This chapter focuses specifically on the Analyze and Design phases of ADDIE, as these represent critical junctures where learner needs, values, and course architecture are established. Accessibility considerations introduced at these early stages are more likely to influence the overall trajectory of course development, whereas interventions introduced during later phases often result in reactive adjustments that emphasize compliance rather than meaningful accessibility design principles.

Integrating Artificial Intelligence (AI) into Instructional Design

The integration of artificial intelligence (AI), particularly generative AI tools like ChatGPT, is reshaping instructional design by accelerating course development tasks. Recent studies indicate that a majority of instructional designers are now leveraging AI to draft objectives, assessments, and course content, with reported gains in efficiency and ideation (McNeill, 2024). While these tools are not substitutes for instructional expertise, they offer practical support for brainstorming and prototyping during the early phases of course design.

Beyond automation, AI also enables dynamic personalization, aligning well with the demands of adaptive learning environments. Kasztelnik (2024) emphasizes that AI technologies such as machine learning and natural language processing facilitate curriculum adjustments based on learner data. Mitre and Zeneli (2024) further emphasize AI’s role in promoting inclusive higher education through tools like chatbots, adaptive learning systems, and assistive technologies, which can personalize content and reduce barriers for students with disabilities.

Overall, these sources emphasize that the benefits of AI are contingent upon ethical use, human oversight, and a critical understanding of AI’s limitations. Thomson Reuters, 2025 Challenges such as prompt engineering, data bias, and lack of contextual nuance remain barriers to fully autonomous implementation (McNeill, 2024). This chapter frames generative AI not as a replacement for design expertise, but as a reflective partner that supports anticipatory thinking about accessibility. A perspective supported by recent findings that position ChatGPT as a performance-empowering collaborator capable of streamlining course prototyping while still requiring human instructional insight for effective adaptation (Choi et al., 2024).

Value Sensitive Design Theory

Value Sensitive Design (VSD) is a theory developed to address the ethical dimensions of technology design (Friedman & Hendry, 2019). Rooted in the field of human-computer interaction, VSD offers a structured approach for incorporating human values such as autonomy, culture, and dignity into the design process. While widely applied in disciplines such as computer science and human-computer interaction, the application of VSD in instructional design remains relatively underexplored, with few studies offering practical frameworks or tools tailored to instructional planning models (Nguyen et al., 2025; Sadek & Mougenot, 2024). VSD presents a valuable framework for guiding reflective, accessible design practices in education.

This chapter focuses on three core VSD constructs, value elicitation, value translation, and value embodiment (Friedman & Hendry, 2019), to inform instructional decision-making within the Analyze and Design phases of the ADDIE model. These constructs were selected for their relevance to early-stage planning, where foundational decisions shape the trajectory of a course and determine the extent to which accessibility is embedded or overlooked. For instance, Burge and Mazzuca (2024) demonstrated how accessibility consultants, when included early in course design teams, contributed critical insights into learner needs that helped inform more inclusive and sustainable design decisions. Value elicitation involves identifying the values of key stakeholders, particularly learners with disabilities, and understanding how those values, such as usability, autonomy, and learner agency, inform design priorities. Value translation addresses the process of operationalizing these values into specific design goals. Finally, the value embodiment focuses on ensuring that these values are materially represented in the technical and structural elements of the learning environment. This may include navigation design, interface logic, or decision-making sequences in branching scenarios. This emphasis on value embodiment aligns with Morgan’s (2024) findings, where deliberately integrating UDL principles into course design led to increased student engagement, suggesting that when accessibility values are reflected in structural and technical features of a course, learners experience greater inclusion and participation. Together, these constructs form the foundation of the Anticipatory Accessibility Reflection Guide introduced in this chapter.

Applying VSD to Instructional Design

To establish a value-informed approach in instructional design, it is essential to integrate the constructs of Value Sensitive Design, value elicitation, value translation, and value embodiment into the Analyze and Design phases of the ADDIE model. This approach accesses

potential value tensions and identifies blind spots early in the design process of learning experiences.

Analyze Phase

The Analysis phase of instructional design sets the foundation. In this phase, instructional designers and design teams explore who their learners are, what they value, and what barriers might exist in accessing the learning environment, especially when building adaptive learning scenarios (e.g., branching simulations). VSD can help teams proactively consider learning styles, values, and accessibility needs. This is an increasingly urgent priority given that many instructional designers report feeling underprepared to address accessibility and would benefit from structured frameworks that guide universal planning from the outset (Lomellini et al., 2024). VSD’s conceptual and empirical investigations can be incorporated here to:

Identify Stakeholders. Consider both direct stakeholders (e.g., learners, instructors) and indirect stakeholders (e.g., IT staff, accessibility coordinators, academic advisors). VSD encourages designers to examine how each group’s values, responsibilities, and constraints influence the learning experience. For example, an accessibility coordinator may prioritize Web Content Accessibility Guidelines (WCAG) compliance and screen reader compatibility, while IT staff may value system security and server load efficiency. These perspectives can shape what tools are approved for use or how assistive technologies are integrated. Gathering input from these groups through interviews, surveys, or informal check-ins could highlight blind spots in the design decisions. Gray et al. (2023) discovered that learning experience designers often rely on collaborative dialogue and context-driven inquiry to navigate competing values and institutional constraints.

Analyze Context. Evaluate the broader socio-technical landscape in which the learning experience will occur. This includes understanding what technologies are available to both

instructors and students, what digital tools are supported by the institution, how cultural values influence communication styles or engagement, and what institutional policies shape accessibility support. For example, limited bandwidth in rural areas may affect a learner’s ability to interact with media-rich content, or an institution's LMS may not support certain accessibility plug-ins. Recognizing these contextual factors allows design teams to anticipate constraints and make informed, value-driven decisions from the outset.

Design Phase

In the Design phase, professionals translate their findings from the analysis into tangible learning solutions that reflect the values and needs of their stakeholders. This phase focuses on designing the structure, content, and interaction strategies of the learning experience. The designers translate analytical insights into pedagogically sound, accessible, and learner-centered solutions by applying VSD principles. Applying VSD at this stage means deliberately aligning instructional strategies and technology choices with the values uncovered during analysis. For instance, a commitment to learner agency may be expressed by offering multiple modes of engagement, such as providing information through text, video, and audio, to accommodate varied preferences in communicating their competency of a learning concept

Designers are also encouraged to create and evaluate low- or high-fidelity prototypes, storyboards, or wireframes in collaboration with a diverse range of stakeholders. These artifacts can serve as testing grounds to assess whether the learning experience genuinely reflects the intended values. By embedding VSD into the Design phase, professionals can make choices that are not only instructionally sound, but also socially responsive and ethically grounded.

During this phase, designers can:

• Translate Values into Design Features: Abstract values such as equity or learner agency are mapped to specific instructional elements. For example, the value of equity might be addressed by designing multiple modes of engagement (e.g., text, audio, video), while learner agencies could inform the use of branching scenarios that allow for autonomous decision-making.

• Develop and Evaluate Design Artifacts: Low- or high-fidelity prototypes, storyboards, and wireframes are created and tested with stakeholders to assess how well the design reflects the intended values.

• Use Participatory Design Methods: Engage stakeholders, especially underrepresented or marginalized learners, in co-design activities to surface potential value tensions and ensure the design does not unintentionally exclude or disadvantage any group.

• Ensure Ethical and Inclusive Use of Technology: Decisions about authoring tools, LMS features, and media are guided not only by functionality but also by their capacity to embody values like accessibility, privacy, and cultural responsiveness.

By incorporating VSD into the Design phase, instructional designers can make deliberate, transparent choices that honor the complexity of human values, resulting in learning experiences that are not only instructionally effective but also ethically robust and socially responsive.

Anticipatory Accessibility Reflection Guide (AARG)

The AARG consists of reflection prompts, AI-assisted ideation strategies, and checkpoints specifically tailored to adaptive and branching learning contexts. These elements are structured to help design teams:

• Identify whose values are being centered or overlooked.

• Translate those values into concrete design decisions.

• Use generative AI tools, such as ChatGPT, to challenge assumptions and explore alternative approaches.

• Evaluate whether early design choices are likely to support accessibility across diverse learner populations.

Reflection prompts are aligned with common moments in instructional planning, such as defining learning objectives, selecting media formats, or structuring learner pathways. For example, when selecting a branching scenario structure, the guide may ask: How does this decision support or limit learner agency, especially for those with cognitive or mobility-related access needs? The AARG is not a rigid checklist, but rather a flexible scaffold intended to support value-informed design thinking across a range of contexts.

Step 1: Analyze Phase

The AARG is organized to mirror the progression of the ADDIE model, with emphasis on the Analyze and Design phases. Each phase includes a set of reflection prompts, value alignment checkpoints, and optional AI-assisted strategies that guide teams through thoughtful consideration of accessibility and learner values. The prompts are meant to be adaptable— whether used during solo planning, team design meetings, or in early stakeholder consultations.

Below is the first part of the guide, focused on the Analyze Phase. Each reflection item is designed to help professionals uncover whose values are present in the decision-making process and whose values may be missing.

Reflection Prompts and Considerations

1. Who are the direct and indirect stakeholders affected by this learning experience, and what are their roles in shaping accessibility? Consider learners, instructors, support staff, and others who influence the design and delivery of the course. What are their goals, concerns, and constraints? Are there stakeholder voices that are not currently represented in the planning process? (Supports - Needs Analysis)

2. What values do these stakeholders hold regarding access, autonomy, and usability? Reflect on how different groups may prioritize various aspects of the learning experience. For example, learners with disabilities may value flexible pacing or screen reader compatibility, while instructional staff may prioritize usability within the LMS. Where do values align, and where might they come from? (Supports - Instructional Storyboards)

3. What institutional or technical constraints might limit the expression of these values? Analyze policies, platforms, or infrastructures that may shape or restrict design decisions. Are there limitations in available tools or institutional policies that unintentionally exclude certain learners? (Supports - Accessibility Audit Reports)

4. How might generative AI tools support value exploration during this phase? Use AI as a brainstorming partner to surface underrepresented perspectives or challenge default assumptions. For instance, you might prompt ChatGPT to suggest accessibility considerations for multilingual learners in low-bandwidth settings or to simulate the perspective of a student navigating a course with cognitive disabilities. (Supports - Accessibility Design Brainstorm Artifact)

5. What assumptions are already embedded in the current course structure, and how do they reflect or exclude stakeholder values? If applicable, consider existing learning materials, for instance, excluding stakeholder values. Are assessment formats varied to accommodate different learning strengths? This reflection helps instructional designers identify where values are already embodied intentionally or not before redesign begins. (Supports - Content/Assessment Gap Analysis Report)

Table 1

VSD Constructs Corresponding to Analyze Phase Reflection Prompts

Step 2: Design Phase

The Design phase builds on the insights gathered during the analysis phase, translating learner needs into tangible course structures, learning objectives, instructional strategies, and assessment plans. During this phase, accessibility must move from intention to integration, shaping the blueprint of the learning experience to participation and meaningful engagement.

Reflection Prompts and Considerations

1. How are accessibility values reflected in the design of learning objectives, instructional materials, and assessments? Examine whether learning goals are inclusive of diverse abilities and flexible learning pathways. Are the assessments designed in multiple formats or modalities? Do course materials reflect principles of Universal Design for Learning (UDL)? Are learners given choices in how they demonstrate mastery? (Supports – Universal Design for Learning)

2. What modes of interaction and engagement are prioritized, and who might be excluded by default design choices? Consider the diversity of learners in terms of communication preferences, sensory needs, and cultural norms. Are discussion boards, synchronous meetings, group projects, or video assignments accessible to all learners? Who might be unintentionally left out or disadvantaged? (Supports – Interaction/Engagement Map)

3. Have you designed for variability in access, attention, and cognition? Ensure that cognitive load is considered across content sequencing, navigation structure, and visual hierarchy. Is the pacing adaptable? Are instructions clear and chunked? Do learners have access to previews, summaries, or scaffolds that aid memory and focus? (Supports – Cognitive Load and Navigation Review)

4. How are learning technologies and tools selected with accessibility in mind? Interrogate whether chosen platforms and tools meet accessibility standards (e.g., WCAG 2.1, Section 508) and whether alternatives exist for students with limited connectivity, assistive technology needs, or privacy concerns. Have you tested the tools through multiple access lenses? (Supports – Accessibility Compliance Audit)

Table 2

VSD Constructs Corresponding to Design Phase Reflection Prompts

Suggestions for Pilot Testing of AARG

The pilot testing for AARG should focus on how the guide influences the thinking and practices of faculty and instructional designers. A small group of instructors and design staff should be invited to integrate AARG into an active design course or as part of a redesign process. Data should be gathered through reflection logs and post-use surveys to capture immediate impressions of usability and adaptability, supplemented by brief interviews to explore how the guide influenced their decisions regarding accessibility. Additionally, design artifacts should be analyzed to identify evidence of shifts in value-based considerations.

Further Exploration

The following curated resources support continued learning around accessible instructional design, adaptive learning technologies, and the ethical integration of AI. These materials are intended to help instructional designers expand upon the concepts introduced in this chapter and apply them in diverse instructional contexts. To continue advancing the field of accessibility practices and ethical educational design, it is important to study the longitudinal impact of accessibility-centered design. Investigating the long-term academic, emotional, and professional outcomes of learners in courses designed with proactive accessibility frameworks is essential to understanding the impact of these intentional design practices.

Author Note

Dr. Rinki Suryavanshi. https://orcid.org/0000-0001-6591-0345

Dr. Aurelia O’Neil. https://orcid.org/0009-0000-8595-4038

There is no known conflict of interest to disclose.

References

Burge, P., & Mazzuca, J. (2022). Accommodation dispensing or course design consulting?: Views and experiences of accessibility consultants contributing to inclusive course redesigns in postsecondary education. Canadian Journal of Educational Administration and Policy, (201), 75-89. https://doi.org/10.7202/1095484ar

Choi, G. W., Kim, S. H., Lee, D., & Moon, J. (2024). Utilizing generative AI for instructional design: Exploring strengths, weaknesses, opportunities, and threats. Techtrends, 68(4), 832-844. https://doi.org/10.1007/s11528-024-00967-w

Friedman, B., & Hendry, D. G. (2019). Value sensitive design: Shaping technology with moral imagination. MIT Press

Kadakia, C., & Owens, L. (2020). Designing for modern learning: Beyond ADDIE and SAM (1st ed.). ATD Press.

Kasztelnik, K. (2024). Artificial intelligence-assisted curriculum development: Innovations in designing educational content for the 21st century learner. Journal of Higher Education Theory and Practice, 24(11), 51–59.

Lomellini, A., Lowenthal, P. R., Trespalacios, J., & Snelson, C. (2024). Instructional designers’ perceptions of accessible and inclusive online course design. Online Learning, 28(4). https://doi.org/10.24059/olj.v28i4.3719

Luo, T., Muljana, P. S., Ren, X., & Young, D. (2025). Exploring instructional designers' utilization and perspectives on generative AI tools: A mixed methods study. Educational Technology Research and Development, 73(2), 741-766. https://doi.org/10.1007/s11423- 024-10437-y

McNeill, L. (2024). Automation or innovation? A generative AI and instructional design snapshot. The 9th IAFOR International Conference on Education in Hawaii 2024: Official Conference Proceedings, 187-194. https://doi.org/10.22492/issn.2189-1036.2024.17

Mitre, X., & Zeneli, M. (2024). Using AI to improve accessibility and inclusivity in higher education for students with disabilities. Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/ITHET61869.2024.10837607

Morgan, A. (2024). Enhancing access in an online course using universal design for learning (UDL) and scenario-based learning (SBL). Techtrends, 68(5), 904-913. https://doi.org/10.1007/s11528-024-00981-y

Nguyen, H., Nguyen, V., Ludovise, S., & Santagata, R. (2025). Value-sensitive design of chatbots in environmental education: Supporting identity, connectedness, well-being and sustainability. British Journal of Educational Technology, https://doi.org/10.1111/bjet.13568

Pataca, Calua., Watkins, M., Peiris, R., Lee, S., & Huenerfauth, M. (2023). Visualization of Speech Prosody and Emotion in Captions: Accessibility for Deaf and Hard-of-Hearing Users. Association for Computing Machinery, 831, (1-15). https://doi.org/10.1145/3544548.35815

Russ, S., & Hamidi, F. (2021). Online learning accessibility during the COVID-19 pandemic. Proceedings of the 18th International Web for All Conference, 8, 1-7. https://doi.org/10.1145/3430263.3452445

Sadek, M., & Mougenot, C. (2024). Challenges in value-sensitive AI design: Insights from AI practitioner interviews. International Journal of Human-Computer Interaction, 1-18. https://doi.org/10.1080/10447318.2024.2439021

Szabo, D. A. (2022). Adapting the ADDIE instructional design model in online education. Studia Universitatis Babeş-Bolyai Psychologia-Paedagogia, 67(1), 125-140. https://doi.org/10.24193/subbpsyped.2022.1.08

Thomson Reuters. (2025, August 28). Ethics of Artificial Intelligence. https://www.thomsonreuters.com/en/insights/articles/ethics-of-artificial-intelligence

Vardeh, M. (2023). Internet access for low-income students (30522790). [Doctoral Dissertation, California State University]. ProQuest Dissertations & Theses Global.

Zhou, X., Kok, C., Quintana, R. M., Delahay, A., & Wang, X. (2023). How learning experience designers make design decisions: The role of data, the reliance on subject matter expertise, and the opportunities for data-driven support. Proceedings of the Tenth ACM Conference of Learning at Scale, 132-143. https://doi.org/10.1145/3573051.3593388