Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

HTN-Based Tutors: A New Intelligent Tutoring Framework Based on Hierarchical Task Networks

Momin N. Siddiqui, Adit Gupta, Jennifer M. Reddig, Christopher J. MacLellan

TL;DR

The paper tackles fixed granularity and rigid cognitive models in intelligent tutoring systems by introducing HTN-based tutors, which encode expert knowledge as Hierarchical Task Networks to enable adaptive scaffolding and hierarchical skill organization. The framework integrates a short-term state representation, an expert model with Tasks, Operators, Methods, and Axioms, and a model-tracing mechanism that selects task decompositions and backtracks based on student actions. It enables dynamic granularity, strategy recognition, and reuse of skills across tutors, potentially reducing cognitive load and improving performance. The work provides a conceptual and illustrative demonstration (logarithmic-expression tutoring) and outlines a research agenda including controlled experiments and platform integration to evaluate effectiveness.

Abstract

Intelligent tutors have shown success in delivering a personalized and adaptive learning experience. However, there exist challenges regarding the granularity of knowledge in existing frameworks and the resulting instructions they can provide. To address these issues, we propose HTN-based tutors, a new intelligent tutoring framework that represents expert models using Hierarchical Task Networks (HTNs). Like other tutoring frameworks, it allows flexible encoding of different problem-solving strategies while providing the additional benefit of a hierarchical knowledge organization. We leverage the latter to create tutors that can adapt the granularity of their scaffolding. This organization also aligns well with the compositional nature of skills.

HTN-Based Tutors: A New Intelligent Tutoring Framework Based on Hierarchical Task Networks

TL;DR

The paper tackles fixed granularity and rigid cognitive models in intelligent tutoring systems by introducing HTN-based tutors, which encode expert knowledge as Hierarchical Task Networks to enable adaptive scaffolding and hierarchical skill organization. The framework integrates a short-term state representation, an expert model with Tasks, Operators, Methods, and Axioms, and a model-tracing mechanism that selects task decompositions and backtracks based on student actions. It enables dynamic granularity, strategy recognition, and reuse of skills across tutors, potentially reducing cognitive load and improving performance. The work provides a conceptual and illustrative demonstration (logarithmic-expression tutoring) and outlines a research agenda including controlled experiments and platform integration to evaluate effectiveness.

Abstract

Intelligent tutors have shown success in delivering a personalized and adaptive learning experience. However, there exist challenges regarding the granularity of knowledge in existing frameworks and the resulting instructions they can provide. To address these issues, we propose HTN-based tutors, a new intelligent tutoring framework that represents expert models using Hierarchical Task Networks (HTNs). Like other tutoring frameworks, it allows flexible encoding of different problem-solving strategies while providing the additional benefit of a hierarchical knowledge organization. We leverage the latter to create tutors that can adapt the granularity of their scaffolding. This organization also aligns well with the compositional nature of skills.
Paper Structure (19 sections, 2 figures)

This paper contains 19 sections, 2 figures.

Table of Contents

  1. Introduction
  2. Background
  3. HTN-Based Tutors
  4. State
  5. Expert Model
  6. Tasks
  7. Operator
  8. Method
  9. Axiom
  10. Model Tracing with HTNs
  11. Example
  12. Discussion
  13. Adaptive Scaffolding
  14. Granularity
  15. Strategy Recognition
  16. ...and 4 more sections

Figures (2)

  • Figure 1: A representation of fraction addition problem-solving knowledge in rule (left) and HTN formats (right), showing methods (ellipse) and operators (rectangle) for head tasks, with lettered callouts indicating equivalent steps in both frameworks.
  • Figure 2: An HTN-based tutor implementation for solving logarithmic expressions is shown. On the left, various paths for model tracing correspond to different skill levels: green for high-skill, yellow for moderate, and red for low-skill students.