Modeling artificial intelligence readiness and semiotic override in Indonesian geometry education
TL;DR
Findings show that hardware exposure alone is insufficient for AI-oriented learning transformation, andometry teachers should diagnose symbolic overreliance and strengthen visualization-based scaffolding, while school leaders and policymakers should balance infrastructure procurement with adaptive pedagogy, teacher support, and regional equity.
Abstract
Hardware-centered modernization often equates classroom technology access with preparedness for artificial intelligence (AI)-supported learning, yet this assumption overlooks subject-specific cognitive barriers in geometry. This study introduces semiotic override (SMO) as a diagnostic construct describing students’ tendency to rely on symbolic cues rather than genuine spatial visualization. A cross-sectional survey of 1,208 Indonesian secondary students from Java and the outer islands examined relationships among smartboard usage (SBU), SMO, spatial intuition (SPI), adaptive learning intention (ALI), and AI readiness (AIR). Data were analyzed using covariance-based structural equation modeling (SEM) with Bollen-Stine bootstrap and multi-group invariance testing. SBU did not significantly predict AIR. SMO negatively affected SPI, whereas ALI positively predicted AIR. The negative effect of SMO was stronger among students in the outer islands than in Java. The model explained 63% of AIR and 38% of SPI. These findings show that hardware exposure alone is insufficient for AI-oriented learning transformation. Geometry teachers should diagnose symbolic overreliance and strengthen visualization-based scaffolding, while school leaders and policymakers should balance infrastructure procurement with adaptive pedagogy, teacher support, and regional equity.
