Affective Scaffolding in Creative Learning: The Vivience Framework for Empathic Digital Humans as Collaborative Partners

Abstract

  As Generative AI (GenAI) systems increasingly transition from passive computational tools to active dialogic partners within educational ecosystems, the affective and communicative dimensions of human–AI interaction demand rigorous theoretical interrogation. Prevailing scholarship predominantly prioritises system efficiency and informational accuracy, frequently marginalising the affective investments and socio-emotional scaffolding requisite for creative learning. Within these vulnerable contexts, ostensibly "neutral" AI interfaces architected for transactional efficiency rather than relational pragmatics are frequently interpreted by learners as affectively indifferent or implicitly evaluative. This phenomenon, conceptualised herein as the "Neutrality Trap," is theorised to engender a discursive void that stifles the psychological safety necessary for creative risk-taking. To address this lacuna, this paper advances the "Vivience Framework for Affective Scaffolding," an integrated conceptual architecture for Empathic Digital Humans (EDHs). Drawing from foundational interactional modelling outlined in Singh (2025), this framework operationalises AI empathy not as an anthropomorphic simulation of sentience, but as a design-dependent communicative scaffold that supports learning regulation. By synthesising Vygotskian scaffolding (Vygotsky, 1978), Self-Determination Theory (Ryan & Deci, 2000), and Control-Value Theory (Pekrun, 2006), the paper articulates the theoretical mechanisms by which AER is hypothesised to mitigate "Affective Load,” the extraneous cognitive burden induced by anxiety and perceived judgment. Finally, the paper posits the "Vivience Protocol" as a rigorous methodological reference design for future empirical inquiry, advocating a paradigmatic shift from monologic, transactional AI to dialogic, relationally attuned AI partners in education.