Empowering Educators: Operationalizing Age-Old Learning Principles Using AI

Abstract

This paper aims to provide a framework for educators to effectively navigate the complexities of artificial intelligence (AI) integration while maintaining the core principles of effective teaching and learning, specifically through the lenses of Dewey’s experiential learning, situated cognition, and distributed cognition. By examining these principles, we explore the essential role of the teacher in this context and the implications of ignoring AI. The integration of AI can enhance personalized learning experiences, facilitate adaptive learning systems, and support educators in fostering critical thinking and problem-solving skills among students. Additionally, we highlight the challenges and ethical concerns associated with AI use in education. We argue that AI should be viewed as a tool that enhances, rather than replaces, the role of educators, emphasizing the importance of maintaining the educator’s role in guiding and supporting student learning. This framework serves as a valuable resource for educators seeking to embrace AI’s potential while ensuring that teaching remains centered on student engagement and successfully achieving learning outcomes as we prepare the next generation to be an AI-ready workforce.

1. Introduction

In recent years, the intersection of artificial intelligence (AI) and education has sparked both excitement and apprehension. Educators, administrators, and researchers grapple with questions about preserving the essence of effective teaching and learning while at the same time achieving heights that were not possible without AI.

For decades, educational research has identified age-old principles that underpin effective teaching and learning (Driscoll, 1994). These principles have stood the test of time. However, translating them into scalable interventions has been a persistent challenge (Mollick & Mollick, 2023). Education scholars have advocated for a paradigm shift away from an excessive focus on content delivery, rote memorization, and standardized testing. Instead, they emphasize cultivating critical thinking, problem-solving abilities, and experiential learning. Then emerged the promise of AI, a powerful technology that employs deep learning models to produce human-like content in response to prompts (Lim et al., 2023). While there is a tendency to focus more on the technical algorithms of AI, it is important to note that cognitive computing has been critical to its transformative nature. Cognitive computing aims to emulate human thought processes in a computerized model, which is essential for developing AI systems that can interact more naturally with humans and understand complex data in a human-like manner (Singla, 2024; Mesmari & Mesmari, 2023). By leveraging the principles of human cognition, which is the core of teaching and learning, AI has the potential to enhance various aspects of our lives.

Consider the concept of “adaptive learning systems”, an instrumental educational approach. The idea is compelling: using technology to tailor instruction to individual learners’ needs, adapting content and pacing dynamically. Until recently, the technology lagged behind the vision. Implementing adaptive systems requires substantial resources, technical expertise, and infrastructure. As a result, only a handful of institutions could experiment with these approaches effectively. Advancements in AI have changed this landscape. Machine learning algorithms can now analyze vast amounts of data, personalize content, and provide real-time insights. Suddenly, adaptive learning is within reach for the average educator. AI can augment teaching practices, allowing educators to focus on what matters most: student learning outcomes.

To align with the demands of the age of AI, a paradigm shift is essential. This shift involves moving from a traditional, standardized approach to a more personalized, adaptive, and experiential-focused model. By leveraging AI, educators can create learning environments that are more adaptive, interactive, and personalized. This paper provides a framework that educators can use to decide how and when to use AI effectively. Through selected age-old learning principles, we examine how educators can navigate AI while acknowledging the challenges that accompany this transition. We also investigate the role of the teacher. Finally, we provide a practical framework for integrating AI into the curriculum. This article offers a theoretical analysis of a novel topic and can inspire future empirical research.

We argue that AI is a tool, an enabler, and not a replacement for educators. Educators remain the experts, possessing a nuanced understanding of their students, subject matter, and pedagogical context. The educator-centered framework we propose acknowledges this agency. It encourages educators to deliberate on when and how to integrate AI, aligning it with their teaching philosophy and goals. As we delve into the intricacies of AI in education, let us remember that it is educators who wield this transformative power. By operationalizing age-old learning principles through AI, we empower educators to create richer, more personalized learning experiences. Together, we can navigate this exciting frontier, ensuring that AI serves as a force for good in our classrooms. Embracing a paradigm shift in education is not merely an option but a necessity in the age of AI. Educators should not approach AI with a polarized viewpoint. The question is not AI, yes or no; it is AI, how? We need to step up and think about what the next generation of learners need to be engaged citizens of a rapidly transforming world. By adopting a framework that emphasizes personalized learning, competency development, and ethical considerations, educators can effectively navigate the complexities of AI integration at their own pace, ultimately enhancing the educational experience for all.

Before we dive into the other sections of this article, we would like to recognize that the great AI revolution or disruption of education followed the COVID-19 pandemic (Hodges & Ocak, 2023). During the pandemic, many educators had to quickly adjust, acquire new skills, and learn new technologies. This sudden overhaul led to increased burnout, with many educators feeling overwhelmed, burnt out, and emotionally exhausted (Geraci et al., 2023; Westphal et al., 2022; Pressley, 2021). The cumulative stress from these changes can lead to resistance to adopting new technologies, such as AI, as educators are tired and may not have the energy to learn one more new thing (Agyapong et al., 2023). We hope that the spirit of this article resonates with you, whether you are an early adopter, cautious, reluctant, or simply too tired. We acknowledge all starting points and hope that by the end of the article, you will feel empowered to take one step further.

2. Learning Principles to Navigate AI

In this section, we will discuss Dewey’s experiential learning, situated cognition, and distributed cognition. These learning principles fall under the broader umbrella of constructivism, which emphasizes the active role of learners in constructing their own understanding and knowledge through interactions, reflections, and experiences. Richey et al. (2011, p. 144) identify three key foundational underpinnings of constructivism:

• Learning results from a personal interpretation of an experience.
• Learning is an active process occurring in realistic and relevant situations.
• Learning results from an exploration of multiple perspectives.

Dewey’s experiential learning, situated cognition, and distributed cognition share a similar foundation that aligns with and may be enhanced by AI, including the importance of context, collaboration, reflection, and active engagement in the learning process. In this section, we examine each principle through four lenses to see the nodes of connections as well as the distinctions (Cennamo, 2011; Reigeluth, 1983). (1) Inputs: Theoretical foundations and focus. (2) Methods: The pedagogical techniques employed. (3) Conditions: Contextual factors influencing the effectiveness. (4) Outcomes: Expected results or type of learning gains. We also give practical examples of how AI can operationalize each principle.

2.1. Dewey’s Experiential Learning

Experiential learning was introduced as part of a discussion of educational innovation and advancement by balancing traditional, structured, rigid education against flexible curricula and contexts that promote freedom. This balance requires that learning experiences are meaningful and grounded in purpose (Thorburn, 2018). Education that enriches and expands the learner’s experience should do so both in vocation and through interaction with the world (Taylor, 2023).

In the foundational Experience and Education, Dewey presents a theoretical framework (Inputs) for experiential learning (Dewey, 1938; Roberts, 2003; Smith, 1959), emphasizing three core components: realism, interaction, and continuity. Realism, in the context of Dewey’s experiential learning, refers to the idea that the way we see and understand reality evolves based on our interactions and experiences. Realism in learning necessitates that education should involve real-world experiences and not just theoretical information (Fabbro, 2021). Interaction pertains to the learner’s relationship with experiences, context, and situations, while continuity addresses the sequential and cumulative nature of experiences over time. Dewey views continuity as an “experiential continuum” in which learners maintain a growing and evolving base of “significant knowledge” (Dewey, 1938). This growth is iterative, allowing for deepening understanding and enhancing the transferability of knowledge across various contexts (Roberts, 2003).

Dewey’s experiential learning employs active learning strategies (Methods) such as project-based learning, simulations, and hands-on experimentation (Dewey, 1938; Thorburn, 2018). These methods emphasize the importance of learner engagement and authentic problem-solving opportunities (Fabbro, 2021). The pedagogical techniques focus on creating meaningful interactions between learners and their environment, fostering reflection, and enabling the practical application of knowledge (Roberts, 2003; Thorburn, 2018). For experiential learning to be effective, several contextual factors (Conditions) must be present (Dewey, 1938). Learners need opportunities to take ownership of their learning process and access the necessary tools and resources. The learning environment should support authentic experiences, meaningful interaction, and provide structured freedom for exploration. Educator guidance remains crucial, offering support while allowing learners to construct their own understanding through experience (Dewey, 1938; Fabbro, 2021). The expected results (Outcomes) include a deeper understanding of the subject matter, enhanced critical thinking capabilities, and the development of adaptable learners capable of problem-solving in diverse contexts (Roberts, 2003; Thorburn, 2018). These outcomes emerge through the iterative process of experience, reflection, and application (Dewey, 1938).

Using AI to Operationalize Dewey’s Experiential Learning

To illustrate how AI can operationalize experiential learning, consider an Environmental Science course examining climate change and coastal ecosystems. In a traditional setting, students may conduct research using textbooks, journal articles, news reports, and other media sources on climate change. They could participate in a role-playing exercise, such as simulating a town hall meeting to discuss climate change strategies. While these methods employ active and project-based methods, they are constrained by limited class time, the educator’s capacity to provide individualized guidance, limited student agency, and fewer opportunities to independently shape discussions or explore alternative solutions. The lack of real-time feedback can reduce student engagement and pause continuity in ways that, if addressed, could result in deeper learning.

The integration of AI can transform this learning experience through two complementary approaches. For institutions with access to immersive technology, such as AI-driven augmented or virtual reality (AR/VR), the AI-enhanced version may look as follows: students engage with a sophisticated virtual coastal ecosystem simulation where they can manipulate environmental variables (such as sea levels, temperature changes, and pollution levels) to observe immediate and long-term impacts on marine life and coastal communities; interact with AI-powered virtual stakeholders (local residents, business owners, environmental scientists) who respond dynamically to students’ proposed solutions; access real-time data visualizations showing how their decisions affect biodiversity, economic stability, and community resilience; and collaborate with peers in virtual field studies, where AI guides them through data collection and analysis while adapting the complexity based on their level of understanding. Furthermore, the AI system provides educators with valuable data analytics, enabling them to make informed decisions about personalized interventions based on individual student needs and progress.

For educators with access to only generative AI, the same learning objectives can be achieved through creative thinking and prompt engineering. For example, the following prompt can be used to obtain ideas and instructions for creating an adaptive learning pathway: “Create a two-level learning pathway in Brightspace LMS for my sophomore-level course focusing on the objective ‘Discuss climate change impact on coastal ecosystems’. For each level, specify content modules and assignment instructions. Include suggestions for using Brightspace’s conditional release, intelligent agents, quiz tools, and discussion boards to create an adaptive learning experience”. The AI-generated pathway can then be reviewed and customized for content accuracy, appropriate scaffolding, clear learning progressions, and accessibility considerations.

Both immersive AI technology and generative AI tools demonstrate how technology can enhance experiential learning by creating dynamic, responsive learning environments that align with Dewey’s principles of realism, interaction, and continuity. These AI-enhanced approaches bridge the gap between theoretical understanding and practical application, making learning more authentic and personalized. Building on Dewey’s emphasis on meaningful experiences grounded in purpose, we now turn to situated cognition, which further explores how learning is inherently tied to the context and culture in which it occurs. This theoretical perspective will provide additional insights into how AI can support authentic, context-rich learning experiences.

2.2. Situated Cognition

Situated cognition emerged based on the understanding that knowing and doing cannot be separated. This means that learning is not just about abstract knowledge, but about how knowledge is applied in real-world situations (Kirshner & Whitson, 2021). Situated cognition emphasizes the importance of social, cultural, environmental, and other contextual factors in shaping learning and cognition (Choi & Hannafin, 1995). In their seminal work, Brown et al. (1989) present a theoretical framework (Inputs) for situated cognition that emphasizes three core components: authenticity, social interaction, and cultural embeddedness. Authenticity refers to the idea that knowledge is best transferred to new situations when it is learned and practiced in the contexts similar to those in which it will be used (Hedegaard, 1998). Social interaction recognizes that learning occurs through engagement with others, particularly experts and peers within a community of practice, rather than being transmitted in isolation (Roth & Jornet, 2013). Cultural embeddedness acknowledges that learning is shaped by and inseparable from the cultural practices and tools of a particular domain (Kirshner & Whitson, 2021; Roth & Jornet, 2013). Collins et al. (1989) and Hennessy (1993) emphasize that knowledge construction occurs through apprenticeship-like experiences where novices learn from experts in authentic settings. An additional aspect of both experiential learning and situated cognition is the idea that cognition is embodied—meaning that our physical experiences and sensations play a significant role in how we think and learn (Robbins & Aydede, 2008; Dawson, 2014).

Situated cognition employs pedagogical techniques (Methods) such as cognitive apprenticeship, authentic problem-solving, and community-based learning (Roth & Jornet, 2013). These methods emphasize the importance of learning within real-world contexts and through genuine participation in domain practices, focusing on creating opportunities for learners to engage with experts, use domain-specific tools, and participate in authentic activities (Robbins & Aydede, 2008; Dawson, 2014).

For situated learning to be effective, several contextual factors (Conditions) must be present. Learners need access to authentic contexts and communities of practice. The learning environment should support legitimate peripheral participation, where learners can gradually move from observing to fully participating in domain practices. Expert guidance and community support are crucial, providing scaffolding while learners develop expertise through authentic engagement (Hedegaard, 1998).

The expected results (Outcomes) include a deep understanding of domain practices, the development of expert-like thinking and problem-solving capabilities, and successful integration into communities of practice. These outcomes emerge through sustained participation in authentic activities and the gradual development of expertise within meaningful contexts (Choi & Hannafin, 1995; Hennessy, 1993).

Using AI to Operationalize Situated Cognition Theory

Most teachers recognize that quality teaching relies on their ability to personalize a learning experience to the individual context of a student, with the goal of improving application of knowledge in real-world situations. Teachers should provide their real-time feedback and context-relevant resources in the process of facilitating authentic learning experiences. Until the advent of AI, however, this level of teaching has been difficult to accomplish at a realistic scale. There are a couple important ways that AI can be used to operationalize situated cognition.

Firstly, AI can be used to generate environments and scenarios that simulate real-world situations and dynamically respond to learner input. Such immersive learning environments can provide learners with the opportunity to apply their knowledge in a social and cultural context that closely mirrors the situations in which they will use the knowledge in real life. These immersive learning environments can be further enhanced with the use of virtual agents using natural language processing (NLP) (Dai & Ke, 2022; Dai, 2024), which is a collection of computational techniques aiming to learn, understand, and produce human languages (Hirschberg & Manning, 2015). Virtual agents, also called chatbots or large language models in the literature, interact with learners through dialogue and coaching in the same way an expert would work with an apprentice or a preceptor with a resident (Dai, 2024). AI can present learners with complex, real-world problems and case-based activities that require the application of multiple areas of knowledge. This can help learners understand how different pieces of knowledge can be integrated and applied in the professional domain. The cognitive assistance of virtual agents is designed to guide inquiry and thinking rather than provide answers (Dai & Ke, 2022). Secondly, AI-powered adaptive learning systems can be used to operationalize situated cognition as well (Ouyang et al., 2023), as AI can tailor responses to steer learners in developing the nuanced skills and emotional responses needed in a profession. For example, in a business course focusing on mergers and acquisitions, students can interact with AI-powered virtual stakeholders representing different companies, each with unique corporate cultures, financial positions, and negotiation styles. The AI system can simulate complex business scenarios where students must navigate cultural nuances, interpret financial data, and apply negotiation strategies in real-time. Through natural language processing, virtual agents can provide immediate feedback on students’ approaches, helping them understand how their decisions align with professional practices and industry standards. Finding experts to coach novices can be challenging, but AI systems can be trained to think like experts, emulating prominent business leaders like Bill Gates, giving learners multiple opportunities to master the art of negotiating a merger. Furthermore, these AI systems can be trained to mimic the characteristics of various communities of practice—for instance, distinguishing between the operational styles of educational business sectors versus information technology sectors—allowing business students to experience different professional cultures and make informed decisions about which community they would like to participate in, all within a low-stakes environment. These immersive environments support cognitive apprenticeship by modeling expert thinking, coaching students through complex decisions, and gradually fading in their provision of support as learners develop expertise.

For educators with access to only generative AI, similar learning objectives can be achieved through strategic prompt engineering. For example, students could engage with an AI system and prompt it to act as an experienced business leader: “You are the CEO of a successful small technology company. I am a potential buyer representing a larger corporation interested in acquiring your company. Please act as though I am a novice business negotiator, providing constructive feedback on my negotiation approach, helping me understand cultural considerations, and guiding me to consider multiple stakeholder perspectives”. The AI can be further prompted to adjust its responses based on the student’s skill level, simulate different negotiation styles, and demonstrate how approaches might vary across different business cultures and communities of practice. Teachers can design prompts that create scaffolded learning experiences, where students progress from basic negotiations to more complex scenarios involving multiple stakeholders and cultural considerations. Students could then be asked to submit a copy of their interaction with the chosen generative AI system along with a reflection on their learning experience.

These examples illustrate how AI can enhance situated learning by supporting its core components: authenticity is achieved through realistic scenarios and professional contexts; social interaction is facilitated through dynamic exchanges with AI agents and peers; and cultural embeddedness is maintained through carefully designed situations that reflect real professional communities of practice. These AI-enhanced approaches enable cognitive apprenticeship at scale, allowing students to participate in authentic professional practices while receiving expert guidance and feedback. Building on situated cognition’s emphasis on how learning is embedded in social and cultural contexts, we now turn to distributed cognition theory, which examines how knowledge and cognitive processes are distributed across individuals, tools, and environments in complex systems.

2.3. Distributed Cognition Theory

Distributed cognition theory (DCog) emerged from the need to understand how information processing and decision-making occurred in a more holistic context beyond the individual mind (Perry, 2003). DCog started with the work of Edwin Hutchins (Hutchins, 1995) in his book Cognition in the Wild, where his analysis of ship navigation revealed the cultural nature of cognition and the importance of focusing on cognition as a system. DCog posits that cognitive processes, that is, how we process information, are not just internal in our minds but are distributed as we interact with our environment, tools, artifacts, culture, etc. The things and people we interact with are referred to as agents and our cognition is shaped by these agents. This differs from traditional cognition theories that focus on “the individual mind as a context-free engine for intelligence” (Hazlehurst, 2015). This theoretical framework (Inputs) recognizes that cognition is not confined to individual minds but is distributed across entire systems of human and non-human agents.

The pedagogical techniques (Methods) employed in distributed cognition focus on analyzing and supporting the flow of information across entire systems rather than individual learning. These methods emphasize system-level analysis of cognitive processes, carefully mapping how information flows between human and non-human agents, and examining how tools and technology mediate cognitive tasks. A key methodological focus is tracking how information transforms as it moves through different parts of the system, recognizing that each transformation contributes to the collective cognitive process (Hutchins, 1995; Hazlehurst, 2015).

For distributed cognition to be effective, several contextual factors (Conditions) must be present. Learners need access to appropriate tools and technologies that support cognitive distribution. Clear communication channels must exist between all system components, whether human or technological. Additionally, there must be a clear understanding of how different agents contribute to cognitive processes and the ability to coordinate effectively between human and technological resources. Importantly, there must be recognition of how information transforms as it moves through the system, ensuring that these transformations enhance rather than hinder cognitive processes (Perry, 2003; Hazlehurst, 2015).

The expected results (Outcomes) of distributed cognition include enhanced system-wide problem-solving capabilities and improved coordination between human and technological agents. Learners develop a better understanding of how cognitive load can be effectively distributed across available resources and tools. They also develop more heightened metacognitive awareness about distributed thinking processes, enabling them to more effectively leverage both human and technological resources in cognitive tasks (Perry, 2003).

Using AI to Operationalize Distributed Cognition Theory

The way learners interact with AI will shape how they process information and influence how tasks are performed and how decisions are made, whether that is for writing, calculations, or practicing in a professional field such as healthcare, engineering, social work, etc. It is therefore important for educators to start breaking apart tasks, what James Lang refers to as unbundling in a podcast interview with Derek Bruff (Rethinking Teaching in an Age of AI with James M. Lang and Michelle D. Miller—Intentional Teaching, 2023). Unbundling is a process where complex learning activities are systematically analyzed and separated into their component parts. This process helps to identify which components require core human expertise that should be developed independently, and which components can be enhanced through AI collaboration, while ensuring essential skill development is not compromised.

Asatiani et al. (2020) explore how implementing automation, if not accomplished well, can result in deskilling. Deskilling in this context refers to the loss or decline in core proficiency and expertise due to an increased reliance, or overreliance, on technology. They recommend utilizing a strategic DCog approach to avoid this, which can be applied to AI use in education: 1. Understanding Human–Machine Interaction involves analyzing how humans and machines can work together effectively in the given context, recognizing the strengths and weaknesses of both, and designing workflows that leverage strengths while mitigating weaknesses. 2. Collaborative Design encourages the creation of systems that support human decision-making rather than replace it, which means involving educators and learners in the process of designing AI tools. 3. Continuous Learning and Adaptation fosters a culture of ongoing learning and adaptation by providing training and support for educators to develop their skills in using AI tools effectively. 4. Feedback Mechanisms are implemented to understand the impact of automation on human performance. By regularly assessing how AI tools affect teaching practices and learning outcomes, educators can ensure that AI integration and assistance serve as a complement to human skills rather than a replacement.

To operationalize this framework effectively, educators can implement a two-phase approach that systematically builds competencies while maintaining core skills; for example, let us consider clinical reasoning for medical student:

Phase 1: Establishing Core Competencies. Students first develop fundamental skills without AI assistance. In medical diagnosis, for instance, students may learn to gather patient history systematically; perform physical examinations; interpret basic lab results; and generate initial differential diagnoses. These skills are developed independently to ensure strong foundational competencies.

Phase 2: Integrating AI as a Cognitive Partner. Once core competencies are established, AI is introduced as an additional cognitive agent. The same medical student now performs initial assessments independently, then generates preliminary differential diagnoses, then consults AI for additional perspectives or rare conditions, critically evaluates AI suggestions using their core knowledge (this step is very important seeing that AI can be biased and can hallucinate), and then documents how their thinking process integrates AI insights. Educators can consult their disciplinary journals to see examples of previous AI integration in their discipline, for example, in medical education (Paranjape et al., 2019).

Orellana et al. (2022) address the term “reliance” and investigate the dynamics of operators’ reliance on automation over time. They studied operators and found that reliance on automation through AI changed with time based on multiple factors, one being the perceived reliability of the technology. This insight is particularly relevant when implementing the two-phase approach, where educators must ensure that students remain mindful of when and how much they depend on AI to complete tasks.

For educators working primarily with generative AI, this distributed cognitive process can be supported through carefully designed prompts. For example, in medical education, the following prompt can be used: “You are an AI clinical decision support system. I will present my initial diagnostic reasoning. Please analyze my thinking process; identify potential cognitive biases; suggest alternative diagnoses I might have missed; and help me think through the diagnostic process without providing a final diagnosis”.

One effective strategy is to incorporate reflective practices into the learning process, encouraging students to critically assess their use of AI and its impact on their learning outcomes. Moreover, AI can serve as a cognitive tool, extending students’ thinking abilities by providing democratized access to information and resources, reflecting DCog. For example, generative AI writing assistants can help students organize thoughts, generate ideas, and refine their writing while learning the processes. Additionally, educators can design assignments that require students to engage with AI tools in a way that complements their own cognitive efforts, rather than substituting them. Promoting digital literacy and critical thinking skills will empower students to discern when AI is beneficial and when it may lead to complacency. By establishing a balanced framework for AI use, educators can cultivate a learning environment that leverages the strengths of AI while nurturing students’ independent problem-solving abilities.

As educators think about DCog in the age of AI, below are five guiding questions based on the work of Perry (2003), Hazlehurst (2015), and Orellana et al. (2022).

• How do people adapt and use tools in their work and learning to support their cognitive abilities?
• How does the context of AI influence cognitive processing?
• How can we strategically determine how AI assistance and interaction shapes the minds of our learners?
• What is appropriate reliance on AI and what is inappropriate for each discipline/context?
• How do we safeguard against inappropriate reliance on AI in a way that could negatively affect the achievement of competency and the ability to perform the core skills needed for a profession?

These questions provide a comprehensive framework for educators to determine how to incorporate AI to harness the positive benefits of DCog while also exploring the impact of AI on cognitive processes and learning outcomes. Thus, the questions act as a decision-making tool but may also guide the Scholarship of Teaching and Learning around AI.

2.4. Summary

In summary, Dewey’s experiential learning emphasizes learning through experience, advocating for active participation and reflection, and focusing on realism, interaction, and continuity. AI can support this by creating adaptive pathways in immersive personalized learning environments that allow students to explore real-world scenarios and discover through interactive manipulation and real-time feedback. Situated cognition posits that knowledge is inherently tied to the context in which it is learned, emphasizing authentic practice and cultural embeddedness. AI can facilitate learning by providing contextually relevant resources and experiences tailored to individual learners. Distributed cognition extends this idea by highlighting how cognitive processes are not confined to individuals but are distributed across people, tools, and environments in complex systems. AI can enhance this distribution by acting as a cognitive partner while maintaining clear boundaries for appropriate reliance on technology. Each principle provides unique insights for creating meaningful and authentic learning experiences in different ways, and AI enables both educators and students to optimize these experiences at scale.

Table 1. Using AI to operationalize learning principles.

Learning Principle	Inputs	Methods	Conditions	Outcomes	AI Operationalization	Example Activity
Dewey’s Experiential Learning	Realism, interaction, and continuity	Project-based learning with reflective practice	Structured freedom within authentic environment	Development of transferable skills and deeper understanding	AI-driven immersive environments (AR/VR) or generative AI prompts for adaptive pathways	Math students use AI simulation to explore geometric shapes, discovering their mathematical properties through interactive manipulation and real-time feedback.
Situated Cognition	Social and cultural context of learning	Cognitive apprenticeship and authentic practice	Access to communities of practice and expert guidances	Integration into professional communities and expert-like thinking	AI virtual stakeholders and adaptive scenarios that mirror professional contexts	Journalism students practice interviewing with AI-generated scenarios, learning culturally responsive reporting through diverse virtual stakeholder interactions.
Distributed Cognition	System-wide information processing	Two-phase approach: core competency building followed by AI integration	Strategic distribution of cognitive information processing across human and technological agents	Enhanced metacognitive awareness and effective resource utilization	AI as a cognitive partner with clear boundaries for appropriate reliance	Dental students combine AI analysis with clinical judgment for diagnosis, developing human–AI partnership skills in treatment planning.

provides a structured framework showing how each principle’s inputs, methods, conditions, and outcomes can be operationalized through AI integration while maintaining the essential human elements of teaching and learning.

3. The Role of the Teacher

Teacher–student interactions remain of vital importance in education, and while AI may alter what that relationship looks like, the role of the teacher is still crucial. Teachers are pivotal in orchestrating AI-driven experiences to foster authentic problem-solving, critical thinking, and collaborative learning. Their expertise lies in designing activities that seamlessly blend AI technologies with human-centered pedagogies, nurturing cognitive skills, emotional intelligence, empathy, and interpersonal abilities. This section will explore how AI enhances the teacher’s role by blending AI with human-centered pedagogies, addressing equity and access, leveraging the transformative power of AI, shifting to collaborative teaching and learning, utilizing AI tools and data analytics, and maintaining ethical considerations and lifelong learning.

AI does not disrupt education or render educators redundant; rather, it has emerged as a potent edTech tool that brings established learning principles into practical application. John Dewey and other constructivist thinkers, along with the theories of situated cognition and distributed cognition as mentioned previously, envisioned education centered on experiential learning, social interaction, contextual knowledge, and the distribution of cognitive processes across people and tools—core principles that AI can now actualize on a broad scale. AI enables educators to create dynamic, student-centered learning experiences that prioritize critical thinking and real-world applications. Through this capacity to engage students more deeply and effectively, AI enhances, rather than replaces, the educator’s role, supporting teachers in guiding and shaping learners. In this way, AI bridges foundational educational philosophies with contemporary technology to address the evolving needs of today’s complex learning landscape. Like all edTech tools, however, AI’s value in education will depend on how purposefully teachers align its use with clear objectives that fulfill desired outcomes. The difference is that teachers must define these objectives, outcomes, and instructional methods in ways distinct from traditional approaches, using AI to foster more personalized, context-rich, and skill-focused learning experiences.

Teachers play a key role in reducing the equity and access gaps (Eden et al., 2024). Unequal access to emerging technologies and digital literacy also risks widening equity gaps. Teachers’ guidance and facilitation are essential in ensuring that technologies are used equitably and effectively (Ziegler et al., 2021). Their human touch, empathy, and ability to foster a nurturing learning environment cannot be replicated by technology alone. Teachers must ensure inclusive and accessible integration, providing training, support, and multiple modalities (e.g., online, hybrid, self-paced) to accommodate diverse needs (Chu, 2019; Bianchini et al., 2015). By thoughtfully leveraging their pedagogical expertise and humanity, teachers can harness the transformative power of AI to empower students from all backgrounds, unlocking their full potential. Furthermore, AI can enhance student motivation by providing personalized experiences that demonstrate the usefulness of information, increase chances of success, and capture interest. Teachers’ guidance and support can create a nurturing environment that fosters motivation and self-regulated learning (Chu, 2019).

The AI era demands a shift from authoritative to collaborative teaching and learning. Historically, teacher–student interactions have been mostly non-interactive knowledge transfers of the standardized curriculum, with feedback limited to specific assessment moments. In a future where AI tools become commonplace, teachers can largely act as facilitators in the learning process. Teachers can guide students through the learning process, helping them navigate the AI tools and resources, and ensuring that the learning activities are aligned with the students’ learning goals and real-world contexts (Gentile et al., 2023).

AI tools such as chatbots, diagnostic and prescriptive analytics, and advancements in NLP can further bolster interactions by automating responses to frequently asked questions and supporting teachers in understanding students’ individual needs better. AI can serve as a facilitator, not a substitute, enriching the teaching process and allowing for more personalized and effective teaching. Teachers can curate the learning experiences for their students. They can select the appropriate AI tools and design learning activities that will provide students with authentic, real-world learning experiences. Rather than viewing AI as an adversary, educators should embrace it as a partner in the teacher–student relationship that enhances the learning experience (Gentile et al., 2023).

AI tools can also generate extensive data about students’ learning processes. Teachers can play a critical role in interpreting these data, understanding what they mean for each student’s learning progress, and making informed decisions about how to support each student’s learning. Using AI and data analytics to operationalize situated cognition can enhance the learning environment through real-time data collection and monitoring that allows for just-in-time, personalized intervention, but raises ethical concerns regarding data privacy and security. Teachers need to understand these systems deeply to manage potential biases and errors (Gentile et al., 2023).

Transparency in AI use opens the door for conversations. Teachers should share how and when they use AI with their students, as this will demonstrate acceptable use and allow students to see that they have agency in how cognition is distributed with AI. This is probably one of the few times where most educators are learning alongside their students. Although this can be uncomfortable, it helps to take the journey together, as illustrates true lifelong learning for our learners.

Lastly, teachers continue to play a vital role as mentors and champions. They provide students with feedback, encouragement, and help them to see connections between their learning and future careers. Context will always be important for learning, and teachers are responsible for ensuring that context exists, which is core to learning principles such as situated cognition. By creating meaningful, high-quality, interactive, and continuous AI-based experiences, teachers can ensure that AI serves as a powerful tool to enhance education.

4. Integrating AI into the Curriculum: The Four-Step AI Response Continuum Framework

In exploring how AI can be operationalized and scaled within educational settings, we have discussed various learning principles, challenges, and considerations. The reality is that many educators are experiencing burnout and technology fatigue following the COVID-19 pandemic (Hodges & Ocak, 2023; Geraci et al., 2023; Westphal et al., 2022; Pressley, 2021; Agyapong et al., 2023), while others are eager to embrace AI’s potential. This diversity in readiness and approaches to AI adoption requires a flexible framework that meets educators where they are at.

The Four-Step AI Response Continuum Framework was developed by Julaine Fowlin as a direct response to faculty needs observed through extensive faculty development workshops and consultations. While formal research validation is still needed, the framework has been iteratively refined based on educator feedback and practical implementation experiences.

The framework’s four steps—ignore, address, redesign, and redefine (see Figure 1)—provide educators with a pathway to view their AI learning and integration as a journey, empowering them to determine where to start based on their current context and readiness level. This framework provides a structured approach to AI integration that acknowledges the practical realities of implementing new technologies in educational settings, while respecting the varying levels of comfort and readiness among educators.

Overview of Steps

• Ignore: The first step recognizes that some educators may prefer to overlook AI. While most are aware that AI is here to stay, it is important to acknowledge that some may still be at this stage. This step is a gentle reminder of why moving to the next phase is crucial.
• Address: In this phase, we focus on how we want students to engage with AI. This involves clarifying what constitutes authorized and unauthorized usage, along with the reasoning behind these distinctions. This step requires minimal changes and emphasizes communication. Importantly, it shifts the focus away from AI detection and towards guiding learners with clear expectations and appropriate guardrails, while also sharing the rationale behind these decisions.
• Redesign: The third step is where the real integration of AI begins. Educators are encouraged to examine their existing assessments and learning activities to identify opportunities for AI inclusion that can enhance learning. This step also involves rethinking workflows, allowing educators to determine where AI can improve their efficiency and growth—such as editing syllabi or creating rubrics.
• Redefine: Finally, this phase invites educators to see AI not merely as a tool but as a subject worthy of study. This involves developing new areas of content emphasizing AI literacy, ethics, and applications across various disciplines. By redefining the curriculum, educators can ensure that students acquire the knowledge and skills necessary to thrive in an AI-driven workforce. Ignoring AI can be dangerous because it can lead to misunderstandings about its appropriate use. Students might misinterpret its presence as tacit approval for using it in ways that have not been discussed or agreed upon. This could result in students relying on AI for tasks that should be performed independently or in ways that could compromise academic integrity and/or the development of key competencies needed in their profession.

As McMurtrie (2023) states, “The one thing that academics can’t afford to do, teaching and tech experts say, is ignore what’s happening. Sooner or later, the technology will catch up with them, whether they encounter a student at the end of the semester who may have used it inappropriately, or realize that it’s shaping their discipline and their students’ futures in unstoppable ways.” (p. 34).

Thus, establishing clear guidelines on AI usage ensures everyone understands its role and avoids its potential misuse, and we caution educators not to ignore AI and to move on to addressing it. Addressing AI, however, should move beyond AI detection. According to Gallant (2024), the arrival of artificial intelligence (AI) in 2022 stirred significant concern across higher education institutions. Many feared that it would lead to a surge in academic dishonesty. However, the root causes of such misconduct—extrinsic motivation, low self-efficacy, perceived assignment meaninglessness, and ample cheating opportunities—have persisted over time. Now, with AI in the picture, it is evident that tackling these fundamental issues can help to mitigate the risk of AI-assisted cheating, just as it would for any other form of dishonest behavior. To combat academic dishonesty effectively, Gallant (2024) asserts that institutions must prioritize strategies that reduce the temptation to cheat, regardless of technological advancements. Cultivating intrinsic motivation among students encourages genuine interest in learning, while enhancing self-efficacy empowers them to tackle challenging tasks confidently. Additionally, making coursework more meaningful by linking it to real-world applications boosts student engagement. A focus on process rather than product is helpful, as this allows educators to scaffold student learning, allowing students to submit drafts, and creating a learning environment where formative assessments are valued. Such an approach reduces the motivation to engage in academic dishonesty as students are more likely to see a stepwise approach to success, rather than a one-big-step approach filled with high stakes and high pressure summative assessments where they feel they only have one chance to illustrate mastery (Bowen & Watson, 2024; Lang, 2013). We discourage the use of AI detection, as the reliability is too mixed and can put educators and institutions in a moral dilemma, and false accusations can affect the student mental health challenges that many institutions are grappling with. If AI detection is used, we have to take into account false positives—cases where the tool incorrectly flags content as AI-generated when it was written by a human (Bowen & Watson, 2024). Leveraging AI thoughtfully—such as designing assessments to minimize cheating opportunities—can be beneficial. Faculty could even establish an AI education task force, backed by funding and training, to explore these avenues. By embracing AI as an educational tool and reinforcing support systems, we can uphold academic integrity and turn potential threats into growth opportunities (Gallant, 2024).

Addressing AI at the course level can be as simple as a syllabus statement and having class discussions about when and how AI should be used, if at all, along with sharing the reasons behind your decisions and the consequences for going against usage policy. Bryant’s University Center for Teaching Excellence offers some very good examples of sample syllabus statements from “some use” to “no use” (https://cte.bryant.edu/sample-syllabus-statements-regarding-ai-and-chat-gpt%ef%bf%bc/, accessed on 23 January 2025). Lance Eaton also curated a list of Syllabus Statements and Policies into an Open Educational Repository: https://docs.google.com/document/d/1RMVwzjc1o0Mi8Blw_-JUTcXv02b2WRH86vw7mi16W3U/edit?tab=t.0, accessed on 23 January 2025.

In the redesign phase, educators can start by considering the principles discussed from experiential learning, situated cognition, and distributed cognition. Drawing on James Lang’s metaphor of unbundling, educators can evaluate the conditions under which AI promotes or limits learner growth. The task for educators in this redesign phase is to unbundle existing assignments and determine what can be coupled with AI and what should remain without AI assistance. Bowen and Watson (Bowen & Watson, 2024) have some very good examples of prompts that could be used to redesign an assignment to include AI, for example, instructing students to ask AI to write a new business proposal in a given context, asking business majors to evaluate the results, and then making a list of errors or improvements that could be made to improve the proposal. So, let us conduct a meta redesign. That is, we will use AI as educators to help us with suggestions for redesigning student assignments with AI: a win-win.

We asked Scite.ai’s research assistant to extend the example above with the business proposal to clearly illustrate the learning principles we discussed, and below is the beautiful outcome (see Box 1) which also illustrates the distributed cognition among us and the AI. It took our human effort to determine the best prompt and to truly evaluate the example. In the example below we left the citations Scite.ai included to keep the authenticity of the output. We did not check the citations, as in this context they are illustrative.

Box 1. Redesigning a Business Proposal Assessment.

Prompt: I am writing a journal article and want to give examples from Dewey’s experiential learning, situated cognition and distributed cognition related to redesigning an assessment with this as the intro help please: “Drawing on James Lang’s metaphor of unbundling when we think about redesigning with AI and the principles of Dcog educators can now ask under what conditions will the use of AI promote growth for learners and under what conditions will it limit growth. So the task for educators in the redesign phase is to unbundle or break apart existing assignments and then determine what can be coupled with AI and what should remain without AI assistance” can you extend this example: “instructing students to ask AI to write a new business proposal in a given context and then asking business majors to evaluate the results. Make a list of errors or improvements that could be made to improve the proposal.”

Output: In the context of redesigning assessments with AI, an illustrative example can be drawn from a business course where students are tasked with creating a new business proposal. This example effectively integrates the principles of experiential learning, situated cognition, and distributed cognition, while also leveraging AI tools to enhance the learning experience.

Original Assessment:

The original assessment required students to individually draft a business proposal based on a hypothetical scenario provided by the instructor. While this approach allowed for creativity, it limited collaboration and real-world application, as students primarily relied on their own research and writing skills.

Redesigned Assessment:

In the redesigned assessment, educators unbundle the original assignment into several components that utilize AI tools while enhancing experiential learning, situated cognition, and distributed cognition.

1. Experiential Learning Component:

Students begin by using an AI tool to generate a draft business proposal based on a specific context, such as launching a sustainable product line. The AI can provide a structured outline, suggest market analysis data, and even generate initial content based on the parameters set by the students. This hands-on interaction with AI allows students to engage with the technology actively, fostering a deeper understanding of business proposal components and the factors that contribute to a successful proposal (Kim, 2023). Following this, students are tasked with refining the AI-generated proposal, incorporating their insights and real-world considerations, thus enhancing their experiential learning through practical application.

2. Situated Cognition Component:

To ground the learning experience in real-world contexts, students are encouraged to collaborate with local businesses or entrepreneurs who have experience in sustainable practices. They can conduct interviews or surveys to gather insights on market needs and consumer preferences. AI tools can assist in analyzing this data, identifying trends, and integrating these findings into their proposals. This situated approach allows students to contextualize their learning, applying theoretical knowledge to practical scenarios and understanding the complexities of the business environment (Arvin, 2023).

3. Distributed Cognition Component:

The redesigned assessment emphasizes collaboration among students by utilizing AI-driven collaborative platforms. Students work in groups to evaluate the AI-generated proposals, identifying errors or areas for improvement. Each group member can contribute their expertise, using AI tools to analyze the proposals critically. For instance, they may use AI to assess the financial projections or marketing strategies suggested in the proposals. This collaborative effort not only enhances their cognitive engagement but also allows them to distribute the cognitive load, leveraging each other’s strengths to produce a more refined final proposal (Barua et al., 2022).

Evaluation and Reflection

After completing the project, students engage in a reflective exercise where they list the errors or improvements that could be made to enhance the AI-generated proposal. This evaluation process encourages critical thinking and self-assessment, as students must articulate their reasoning for each suggested change. They can also discuss how the integration of AI influenced their understanding of business proposal development and the potential benefits and limitations of relying on AI in this context (Selwyn, 2022).

Conclusion

Through this example, the redesigned assessment not only integrates AI to enhance the learning experience but also aligns with the principles of experiential learning, situated cognition, and distributed cognition. By unbundling the original assignment, educators can create a more dynamic and engaging learning environment that promotes growth and development among learners. This approach encourages students to actively participate in their education, collaborate with peers and industry professionals, and apply their knowledge in meaningful ways, ultimately preparing them for real-world challenges in the business landscape.

References: [These references are illustrative, copied and pasted directly from the AI output. Always verify all AI-generated citations for accuracy].

Arvin, N. (2023). Teacher experiences with ai-based educational tools. aitechbesosci, 1(2), 26–32. https://doi.org/10.61838/kman.aitech.1.2.5

Barua, P., Vicnesh, J., Gururajan, R., Oh, S., Palmer, E., Azizan, M., … & Acharya, U. (2022). Artificial intelligence enabled personalised assistive tools to enhance education of children with neurodevelopmental disorders—a review. International Journal of Environmental Research and Public Health, 19(3), 1192. https://doi.org/10.3390/ijerph19031192

Beard, C. (2018). Dewey in the world of experiential education. New Directions for Adult and Continuing Education, 2018(158), 27–37. https://doi.org/10.1002/ace.20276

Benecke, D. and Bezuidenhout, R. (2011). Experiential learning in public relations education in south africa. Journal of Communication Management, 15(1), 55–69. https://doi.org/10.1108/13632541111105259

Iredale, A. (2012). Down the rabbit-hole. Higher Education Skills and Work-Based Learning, 2(1), 54–62. https://doi.org/10.1108/20423891211197749

Kim, S. (2023). Change in attitude toward artificial intelligence through experiential learning in artificial intelligence education. International Journal on Advanced Science Engineering and Information Technology, 13(5), 1953–1959. https://doi.org/10.18517/ijaseit.13.5.19039

Modran, H. (2024). Using the theoretical-experiential binomial for educating ai-literate students. https://doi.org/10.20944/preprints202402.1712.v1

Ord, J. and Leather, M. (2011). The substance beneath the labels of experiential learning: the importance of john dewey for outdoor educators. Journal of Outdoor and Environmental Education, 15(2), 13–23. https://doi.org/10.1007/bf03400924

Roberts, J. (2018). From the editor: the possibilities and limitations of experiential learning research in higher education. Journal of Experiential Education, 41(1), 3–7. https://doi.org/10.1177/1053825917751457

Selwyn, N. (2022). The future of ai and education: some cautionary notes. European Journal of Education, 57(4), 620–631. https://doi.org/10.1111/ejed.12532

Wilson, J., Brain, R., Brown, E., Gaind, L., Radan, K., & Redmond, J. (2016). Interdisciplinary study abroad as experiential learning. Comparative and International Education, 45(2). https://doi.org/10.5206/cie-eci.v45i2.9291

The redefining step can be as simple as inviting guest speakers from the AI field and hosting panels with experts who can provide diverse perspectives and insights. Regular check-ins with students allow for ongoing dialogue and feedback, while providing resources such as readings and videos empower students to educate themselves further on AI ethics and responsible use. These strategies ensure that students develop a comprehensive understanding of AI etiquette and are equipped to navigate its use thoughtfully and responsibly in the classroom and beyond. A good example of redefine is seen in the example from the Medical University of South Carolina (MUSC) in their post on MUSC AI Initiatives Education Timeline: “Almost 900 students in 10 different academic programs are required to complete IP 711: IP Foundations and TeamSTEPPS, which, as of Fall 2024, includes a module on AI that introduces students to AI concepts and definitions, application to healthcare, impact on healthcare teams and teamwork, and ethical and security considerations” https://education.musc.edu/education-innovation/blog/2024/october/musc-ai-initiatives, accessed on 23 January 2025. Additionally, a great resource for introducing students to AI as we redesign and redefine our curricula is the free student guide to navigating college in the era of artificial intelligence developed by Elon University in partnership with The American Association of Colleges and Universities (AAC&U) https://www.aacu.org/publication/ai-u-v1, accessed on 23 January 2025.

5. Conclusions

The synthesis of age-old learning principles and the presented framework within this article ultimately address AI tool integration to create learning experiences that are intentional, purposeful, relevant, socially constructed, and incremental in the attainment of mastery. While the novelty and function of AI tools may be of interest to educators, it remains important to design their integration in alignment with and with a strong focus on the curriculum and learning objectives. Such a purposeful integration of AI tools must ensure that learning experiences maintain relevance to both the learning objectives and the learner.

The social construction of learning experiences in incorporating AI tools is vital. Both situated and distributed cognition speak to elements of social construction that must be considered in educational program design. In the case of situated cognition, the principle of socialized learning is direct. Meanwhile, in distributed cognition, socialization is a component of the environmental and cultural considerations of learning design. Social interaction and the reflection of society should be part of the realism that is central to Dewey’s experiential learning. The combination of these principles ideally yields AI-enhanced learning experiences that include collaboration and activity, motivating learners to progress.

The learner’s heightened awareness of their own cognition through the use of AI tools empowers their capacity for self-direction. Critical thinking and problem-solving skills are enhanced and developed in realistic, robust AI-based learning experiences.

These principles inform the importance of integrating AI tools into educational programs within an educator-centered framework. This article presents such a framework addressing the stages of AI tool engagement for educators.

The selection of appropriate AI tools and systems requires the careful consideration of several factors: alignment with pedagogical goals, integration with existing learning management systems, data privacy compliance, accessibility features, and institutional resource constraints. This is particularly important for public universities where budget limitations may impact technology adoption decisions, as evidenced by the increased strain on institutional resources and educator capacity following the COVID-19 pandemic (Hodges & Ocak, 2023; Geraci et al., 2023; Westphal et al., 2022; Pressley, 2021; Agyapong et al., 2023). Building awareness and trust in AI systems among students and staff is key. Effective communication through emails, websites, social media, and direct engagement can boost adoption and confidence in these systems. Training staff on how to work with AI and addressing their job security concerns are also important steps. Ethical considerations are essential when implementing AI in education. Universities and developers must handle data responsibly and follow privacy laws. It is important to acknowledge the wide-ranging impact of AI on teaching, ethics, society, and the economy. While AI offers great opportunities, it should enhance, not replace, the human aspects of education, such as emotional support and personalized teaching. By viewing AI as a supportive tool, universities can maintain the holistic nature of education and ensure it benefits everyone involved (Sharma et al., 2022).

By seamlessly integrating emerging technologies with their pedagogical expertise, teachers serve as agents of the transformation of students’ realities, unlocking their full potential and preparing them for success in an ever-evolving world.