EDU-MATRIX: A Society-Centric Generative Cognitive Digital Twin Architecture for Secondary Education

Wenjing Zhai; Jianbin Zhang; Tao Liu

cs.MAFeb 21, 2026

EDU-MATRIX: A Society-Centric Generative Cognitive Digital Twin Architecture for Secondary Education

Authors: Wenjing Zhai, Jianbin Zhang, Tao Liu

Organizations: The High School Affiliated to Beijing Normal University, Beijing, China · Department of Electronic and Communication Engineering, North China Electric Power University, Hebei, China

Abstract

Existing multi-agent simulations often suffer from the "Agent-Centric Paradox": rules are hard-coded into individual agents, making complex social dynamics rigid and difficult to align with educational values. This paper presents EDU-MATRIX, a society-centric generative cognitive digital twin architecture that shifts the paradigm from simulating "people" to simulating a "social space with a gravitational field." We introduce three architectural contributions: (1) An Environment Context Injection Engine (ECIE), which acts as a "social microkernel," dynamically injecting institutional rules (Gravity) into agents based on their spatial-temporal coordinates; (2) A Modular Logic Evolution Protocol (MLEP), where knowledge exists as "fluid" capsules that agents synthesize to generate new paradigms, ensuring high dialogue consistency (94.1%); and (3) Endogenous Alignment via Role-Topology, where safety constraints emerge from the agent's position in the social graph rather than external filters. Deployed as a digital twin of a secondary school with 2,400 agents, the system demonstrates how "social gravity" (rules) and "cognitive fluids" (knowledge) interact to produce emergent, value-aligned behaviors (Social Clustering Coefficient: 0.72).

Figures & tables

Figure 1. The Paradigm Shift Unlike traditional models (Left) where rules are locked inside agents, EDU-MATRIX (Right) uses a "Society-Centric" paradigm where the Central Microkernel acts as a gravitational source, dynamically injecting rules into clean cognitive agents.

Figure 2. System Architecture The Social Microkernel (Center) orchestrates the interaction between the agent population and the plugin ecosystem, enabling the dynamic injection of environmental rules.

Figure 3. Cognitive-Oriented Social Topology The graph reveals high-density "Micro-Knowledge Communities." Agents are "held in place" by their social connections, ensuring value consistency.

Figure 4. Modular Logic Evolution Protocol A four-stage cycle that transforms ephemeral interactions into solidified "Logic Assets" (Long-Term Collective Memory).

Figure 5. The "Neural Handshake" Workflow A hybrid interaction mechanism where symbolic constraints (Gravity) intercept unsafe neural outputs (Fluid), visualizing the conflict for human-in-the-loop resolution.

Figure 6. Logic Synthesis Reactor This module processes "Knowledge Capsules"—modular logic fragments—by fusing disparate inputs, such as "Centenary School History Topology" and "Polymorphic Architecture Protocols," into new, synthesized logic assets.

Figure 7. EDU-MATRIX Deployment Hub This interface monitors the global synchronization of matrix nodes, including the KAI Digital Facilitator, the Digital School History Museum, and the Frontier Technology Lab.

Figure 8. Neural Topology Map This map visualizes the active logic nodes and skill links surrounding the Matrix Core. The system maintained a global resonance synchronization rate of 98.4\% throughout the simulation. — Figure 8. Neural Topology Map This map visualizes the active logic nodes and skill links surrounding the Matrix Core. The system maintained a global resonance synchronization rate of 98.4% throughout the simulation.

Figure 9. KAI Digital Facilitator Interface The interface demonstrates the "Neural Handshake" in action, where symbolic institutional rules intersect with neural LLM generation to resolve cognitive conflicts or ethical boundary crossings.

Evaluation Dimension	Indicator Value	Conclusion
Social Clustering Coefficient	0.72	Realistic formation of student cliques.
Global Resonance Sync	98.4%	High alignment with institutional values.
Value Injection Efficacy	+42%	Increase in "social contribution" discussions.
Dialogue Consistency	94.1%	Long-term persona stability.

Table 1. Key Performance Indicators

Digital Twins

Digital twins are virtual models synchronized with real systems to support monitoring, prediction, simulation, and decision-making.

See all 253 papers

AgentSchool: An LLM-Powered Multi-Agent Simulation for Education
May 28, 2026Yulei Ye, Wenhao Li, Zhong Wen +23Multi-Agent LLM Systems Human Behavior Simulation
Shanghai Institute of AI for Education, East China Normal University · School of Computer Science and Technology, Tongji University · School of Computer Science and Technology, East China Normal University +3
EduMirror: Modeling Educational Social Dynamics with Value-driven Multi-agent Simulation
Jun 6, 2026Jingzhe Lin, Hengbin Yu, Yongdan Zeng +1Human Behavior Simulation AI in Education
School of Artificial Intelligence, Beijing Normal University, Beijing, China · Beijing Key Laboratory of Artificial Intelligence for Education, Beijing, China · Engineering Research Center of Intelligent Technology and Educational Application, Ministry of Education, Beijing, China +2
Social Simulations: from Agent-Based Modeling to Digital Twins
Jul 15, 2026Erica Cau, Andrea Failla, Valentina Pansanella +1Human Behavior Simulation Agent-Based Modeling
Department of Computer Science, University of Pisa · ISTI-CNR

May 28, 2026J/K move · Enter open · S save

AgentSchool: An LLM-Powered Multi-Agent Simulation for Education

Yulei Ye, Wenhao Li, Zhong Wen, Yunshu Huang+22

Shanghai Institute of AI for Education, East China Normal University · School of Computer Science and Technology, Tongji University · School of Computer Science and Technology, East China Normal University+3

Multi-Agent LLM Systems Human Behavior Simulation AI in Education User Modeling Large Language Model-Based Social Simulation Teacher-Student Learning

Open paper AskPDF arXiv

Despite the rapid deployment of LLMs into classrooms, validating educational AI remains uniquely intractable: interventions act on developing learners whose cognitive and social trajectories are irreversibly shaped, while real-world trials are slow, ethically constrained, and institutionally locked. LLM-based educational simulators have emerged as a potential remedy, but many still collapse learning into persona-conditioned role-play and, when optimized only to reproduce existing classrooms, can structurally penalize the institutional novelty that pedagogical reform requires. In this work, we introduce AgentSchool, an LLM-driven multi-agent simulator that models learning as state transition rather than prompted behavior. AgentSchool couples cognitively growable student agents -- equipped with weighted subject knowledge graphs, thinking-workflow pools, and explicit misconceptions -- with adaptive teacher agents that plan, scaffold, and reflect along the Zone of Proximal Development, embedded in a configurable scenery generator that situates instruction within both formal and informal learning fields, and a multi-scale simulator that decouples interaction scale, temporal granularity, and simulation duration. Experiments show that structured student agents produce more differentiated mastery and misconception traces than a baseline simulator, while teacher-agent comparisons show backbone-dependent patterns consistent with ZPD-informed adaptation. Further, AgentSchool generates plausible traces of peripheral participation, clique formation, aggressor-induced cohesion, and opinion-leader emergence consistent with classroom social theories. Beyond its role as an educational research instrument, AgentSchool frames education as a socially meaningful testbed for long-horizon memory, multi-agent coordination, and future institutional reasoning under organizational pressure.

Abstract

Digital Twins

Explore similar work

AgentSchool: An LLM-Powered Multi-Agent Simulation for Education

EduMirror: Modeling Educational Social Dynamics with Value-driven Multi-agent Simulation

Social Simulations: from Agent-Based Modeling to Digital Twins