Skip to main content
U.S. flag

An official website of the United States government

Open Social Learner Model (OSLM) Closed

Open Social Learner Model artwork with example screenshot

Combining the power of open social learner modeling and adaptive navigation for a more effective personalized learning experience.

The Challenge

Adaptive learning systems, such as intelligent tutors, typically include models of the learners’ knowledge and skills. In some systems, the learners can review these models; these are called “open learner models.” These have proven effective at supporting learning, but could that approach be improved upon by integrating social learning?

The Solution

Develop and test a prototype learning system with an intelligent content interface that leverages open social learner modeling and adaptive navigation. Enable social learning through student peer groups to facilitate the identification and usage of learning content and enhance motivation by showing comparisons of each learner’s performance compared to his or her peers’ progress.

About the Project

The purpose of this Open Social Learner Model (OSLM) project was to develop architectural solutions and authoring tools for open social learner modeling that leverages:

  • Open Student Modeling (OSM), an approach to technology-based learning making student models available to learners.
  • Open Social Student Modeling (OSSM), a complement to cognitive aspects of OSM combined with social aspects allowing students to explore models of peer students and/or an aggregated class model.

Modeling approaches can be used to determine a learner’s state and represent it as an individualized learner model. When individualized learner models are used with adaptive instructional systems, the most relevant content items and learning pathways can be provided to the learner.

The OSLM project focused on the development of an open social learner modeling interface for diverse learning content made available on a portal. This new interface, referred to as the Mastery Grid System, incorporates enhanced algorithms for personalized guidance using knowledge-based and social approaches. The system is grounded in self-regulated learning and learning motivation theories, as well as social comparison.

The Mastery Grids System provides flexible user-centered navigation across different content levels (e.g. topic, each content type) and different content types, and tracks user feedback on recommendations dynamically. The system can be used to report on both a student’s progress level (based on activities) and knowledge level (based on estimated user knowledge). The data on user activities are tracked, and a student’s knowledge level is updated to a centralized user modeling server. A student’s changing state of knowledge is visualized in the system to support self-regulated learning and to guide students toward the most appropriate practice topics and content.

The social aspect of the interface is grounded in progress comparisons. The interface generates visualizations that display the progress of a student’s knowledge by topic and compares personal

progress with the progress of the class. A student’s learning progress and knowledge is displayed in colored grids from four perspectives that show the student’s progress:

  1. “Me” or my progress
  2. “Me vs group” or comparison grid
  3. Performance of the students in the group (“Group” or group grid)
  4. Performance of all the students in the class

Mastery Grids not only enable a student to be aware of the strength and weakness among his/her own knowledge modules, but also encourages students to catch up with other students (observing average class progress or advanced students’ progress), and follow the potential good learning path of the peers.

This project also leveraged the Mastery Grids System to research the effectiveness of OSM and OSSM. One such research activity was a large-scale classroom study that explored the impact of the social dimension of OSSM. Students in a database management course accessed non-required learning materials (examples and problems) via the Mastery Grids interface using either OSM or OSSM approaches. The results revealed that OSSM-enhanced learning, especially for students with lower prior knowledge, compared with OSM, had a much higher impact on engaging and retaining students. The use of OSSM also motivated students to perform significantly more work with non-mandatory learning content, and with more efficiency.


Project Details

Period of Performance



Learning + Technology Research Group at Aalto University
Personalized Adaptive Web Systems Lab, University of Pittsburgh