Adaptive Learning using Finite State Machine Logic

Abstract

We demonstrate the feasibility of Finite State Machine (FSM) logic to design adaptively scaffolded activities, presenting early work integrating adaptive learning into a learning tool in widespread use globally. We describe how integrating FSM logic with existing assessment architecture enables us to extend from direct measurement to scaffolding and feedback interventions on a spectrum of timescales from 1-second to several hours. Four prototypes are shared, demonstrating how this FSM logic affords design across differing learning contexts. Implications for design of efficiency and empowerment at scale, potential for content co-creation, and transformation of learning are discussed.