Project Overview

Journey is a radical reconception of the textbook for the age of AI. Students view generated pages containing the dynamic information they need in the moment to best facilitate learning, fill-in knowledge gaps, or practice problem solving. The path backwards through the pages are a record of the concepts they've mastered. The journey forward explores a wilderness of new concepts waiting to be tamed. Students can take advantage of the generative AI to query the textbook or they can play with interactive illustrations, tables, and figures. Coding is integrated with rich, immediate feedback. The topic of our first journey is MATLAB.

Key Features

  • Adaptive Learning Pathways: By leveraging Large Language Models and analytics from WebTA, the platform will identify subject areas where students are are engaged and interested or where they are struggling and recommend targeted readings and practice problems. This adaptive learning approach will facilitate a more personalized education, ensuring learners receive the support they need to fill knowledge gaps and excel in their studies.
  • Real-time Code Evaluation and Critique: A core feature of this environment is the integration of WebTA, our Code Critiquer software developed to evaluate student code submissions. WebTA will provide immediate, automated feedback on code quality, efficiency, and adherence to best practices. This rapid feedback loop accelerates the learning process, improves coding skills, and increases self-efficacy in programming.
  • Dynamic Figures: The online textbook will incorporate interactive illustrations, figures, and simulations to enhance understanding and retention. Students will be able to manipulate illustrations, visualize abstract concepts, and immediately observe the impact of their changes. This interactive approach transcends the capabilities of traditional textbooks and is aligned with contemporary research on effective learning strategies.

Research Questions

  1. How effectively does the platform's adaptive learning pathway, driven by WebTA analytics, identify and remediate individual student knowledge gaps in programming concepts compared to traditional linear textbook approaches?
  2. What is the longitudinal impact of immediate, automated code feedback (via WebTA) integrated within a textbook context on students' programming proficiency, problem-solving skills, and self-efficacy in programming compared to delayed or less frequent feedback mechanisms?
  3. How can the generative AI be leveraged to provide diverse, tailored explanations of challenging technical concepts based on individual student understanding and learning preferences, and how does this impact comprehension and knowledge retention?
  4. How do dynamic figures and simulations within the online textbook enhance students' comprehension of fundamental yet abstract engineering principles (e.g., control systems, thermodynamics) compared to static diagrams and textual explanations, and what are the underlying cognitive mechanisms (e.g., mental model formation, working memory) at play?
  5. Can machine learning models be trained on student interaction data within the Journey platform to accurately predict their learning trajectories and identify students who are likely to struggle, enabling timely and targeted interventions?

Research Team

  • Dr. Leo C. Ureel II
  • Dr. Jon Sticklen
  • Dr. Betsy Lehman
  • Jyoti Suhag
  • Benji Sutton
  • Adam Fenjiro