Reshaping education in the era of artificial intelligence: insights from Situated Learning related literature

Synthesis — Vargas, Chiappe & Durand (2024) conduct a PRISMA systematic review of 60 peer-reviewed articles spanning three decades to map how Situated Learning has evolved and how AI can reshape education around it. Situated learning — learning through authentic, real-world contexts, communities of practice, legitimate peripheral participation, and contextual knowledge — confronts the traditional school's vertical, one-way passive instruction, its fixation on predefined outcomes, and teachers' limited contextual knowledge. The authors argue AI offers concrete solutions: adaptive systems tuned to learners' evolving needs, intelligent tutoring embedded in authentic scenarios, automation of administrative tasks, and data-driven teacher support. Framed through Education 4.0, AI and situated learning are cast as complementary — AI providing adaptive scaffolding and analytics to scale situated learning's benefits while situated learning grounds AI-driven education in real-world practice and complexity. Human guidance remains essential for ethical grounding.

Vargas, Chiappe & Durand (2024) offer a systematic synthesis of how situated learning — a theory rooted in the early 1990s work of Lave and Wenger — can be reinvigorated by artificial intelligence. Using the PRISMA method, the review analyzed 60 peer-reviewed journal articles from Scopus over three decades (1990s to 2022), extracting and analyzing the concepts associated with situated learning both qualitatively and quantitatively. The paper positions this conversation within Education 4.0, the educational face of the fourth industrial revolution in which IoT, robotics, immersive environments, and especially AI reconfigure the learning experience.

Key Findings

• Situated learning is a dynamic, evolving concept organized around the durable core idea of "learning in context," with emphasis shifting across eras: Workplace Learning in the 1990s, ICT-mediated communities of practice in the 2000s, mobile and gaming immersion in the 2010s, and Education 4.0 Technologies (augmented reality, immersive environments, AI, IoT, educational robotics) in 2020–2023.
• Three obstacles block situated learning. (1) The traditional school system is hegemonic, centralizing knowledge and relying on vertical, one-way passive transmission with little connection to real contexts. (2) The predominant educational approach fixates on predefined outcomes, competencies, and content, excluding learners' and teachers' knowledge, interests, and needs. (3) Teachers often lack contextual knowledge and sensitivity to the cultural diversity and everyday realities of their students.
• Six challenges to implementation: linking art and Creativity into teaching; shaping critical, context-aware individuals with ethical capacity; recognizing students and teachers as active agents; understanding situated learning beyond mere practice; rethinking linear, static teaching toward dynamic horizontal models; and treating context as a facilitator of expression.
• AI offers concrete solutions to these obstacles: adaptive systems tailored to students' evolving needs, Intelligent Tutoring situated in authentic scenarios, automation of administrative tasks, and data-driven teacher support.
• AI enables personalization and student agency, letting learners co-direct their pathways in collaboration with intelligent agents — a shift away from the industrial, mass-production model of schooling.
• Human guidance remains essential: instructors must instill ethical reasoning to question AI biases, so that AI augments rather than replaces the human, ethical dimension of education.
• AI and situated learning are complementary. AI provides adaptive scaffolding and analytics to scale situated learning's advantages; situated learning grounds AI-driven education in real-world practices and complexity — together demanding a reinvention of learning ecosystems.

Educational Significance

The review is significant because it bridges a classical learning theory and contemporary AI research, showing how an Constructivism, context-anchored view of learning can steer AI deployment rather than merely reacting to it. For educators, it reframes the Teaching as designer of flexible, context-connected learning spaces supported — not supplanted — by AI. For instructional designers and policymakers, it argues that effective Adaptive Learning and Personalized Learning depend on grounding AI in authentic contexts and communities of practice, and that scaling such experiences via AI can make engaged, situated learning accessible to far more students. Its insistence on human ethical guidance connects directly to ongoing debates about AI in Education and the responsible integration of AI in Higher Education and schooling.

What this means for practice

• Instructors. Design participatory simulations that combine virtual environments with real experience rather than relying on either alone, so that context stays connected to students' actual settings.
• Instructors. Use AI recommendation systems to surface timely, personalized resources and adaptive support, while keeping human guidance in place for ethical reasoning and for questioning AI bias.
• Faculty developers. Use learning analytics to identify precise teacher support needs and target professional development at the contextual-knowledge and cultural-sensitivity gaps the review documents.
• Administrators. Integrate AI across curricula instead of running isolated projects, and align policy, assessment, and teacher training around contextualization, co-creation, and learner agency.

Limitations

• The review synthesizes 60 articles drawn from Scopus alone, screened from 2,532 records and reduced through a probabilistic subsample (95% confidence, 5% error) to 276 documents; coverage is bounded by that single database and stops at 2022.
• Its evidence is published journal text and term-frequency patterns, so the AI solutions it proposes — adaptive systems, intelligent tutoring, administrative automation, data-driven teacher support — are argued rather than independently tested.
• Restricting inclusion to peer-reviewed journal articles excludes Gray literature and unpublished or negative results, leaving the synthesis open to publication and selection bias.

Citation

Vargas, E. G., Chiappe, A., & Durand, J. (2024). Reshaping education in the era of artificial intelligence: insights from Situated Learning related literature. Journal of Social Studies Education Research, 15(2), 1–28.