From Cognitive Debt to Cognitive Amplification: The Amplifier-Crutch Model of Equitable Educational AI

Kosmyna et al. introduced the concept of cognitive debt to describe diminished neural engagement, recall, and ownership of thought during AI-assisted essay writing. Their multimodal study—combining EEG, textual analysis, and interviews—was methodologically innovative, but its interpretation risks overstating harm and under-contextualizing benefits. This paper critiques and extends Kosmyna et al.'s framing by arguing that cognitive debt is conditional rather than deterministic. It introduces the amplifier–crutch model, a conceptual framework for analyzing when AI-tools weaken learning and when they strengthen it, providing academics, policymakers, and educators with a framework that moves beyond binary narratives. Drawing on research across cognitive offloading, metacognition, and educational integration of AI-tools, the paper synthesizes evidence from neuroscience, educational psychology, and global policy studies. The framework integrates these literatures and generates 14 testable hypotheses with operational definitions. The analysis reveals that AI-tools function as a crutch when learners defer uncritically to outputs, fostering shallow processing and dependency. Conversely, it functions as an amplifier when pedagogy, strategy, and context scaffold reflective engagement. Key boundary conditions include learner expertise levels, task complexity thresholds, and cultural-contextual factors. By treating cognitive debt as a lens rather than verdict, this framework provides a generalizable model for research, practice, and policy. The amplifier–crutch model shifts focus from whether such tools help or harm to understanding the conditions that determine outcomes, enabling evidence-based approaches to educational AI that design for amplification rather than fear inevitable decline.