Language Concealment of Platform Media and the Cognitive Reconstruction of Critical Media Literacy

Abstract

In the digital era characterized by the deep integration of Generative Artificial Intelligence (GenAI) and algorithmic recommendation systems, traditional media literacy education is facing a severe paradigmatic crisis. The instrumentalist approach, which relies on ‘fact-checking’ and ‘information skills’, fails to effectively address the cognitive narrowing and emotional polarization driven by algorithms. Grounded in the ontological foundation of the ‘Progressive Weakening-Compensation Principle’ (Wang, 2009) and drawing upon the three-layer cognitive structure of ‘Soma-Emotion-Meaning’ alongside the Regulate-Imagine-Discover (RID) model (Zhang, 2026), this paper offers an ontological diagnosis of the cognitive erosion mechanisms perpetuated by platform media. The analysis reveals that algorithmic recommendation systems hijack the ‘somatic markers’ at the emotional layer, thereby stripping subjects of their epistemic agency. This process triggers the ‘Language Concealment Effect’—detaching linguistic symbols from their embodied experiences and authentic emotional anchors, reducing them to hollow clickbaits. To counter this crisis, this paper proposes a cognitive reconstruction of critical media literacy, advocating a shift from mere ‘information critique’ to ‘ontological reflection’. By restoring embodied perception at the ‘soma-emotion’ layers, we can rebuild the cognitive stabilization scaffolding, thereby reclaiming individuals’ meaning-making capacities
and cognitive autonomy in the algorithmic age.