EAGER: ALICE Adaptive Learning for Interdisciplinary Collaborative Environments

University Of Georgia Research Foundation Inc, Athens GA

Investigators

Abstract

Interdisciplinary science, technology, engineering, and mathematics (STEM) education has traditionally consisted of a hierarchical and deterministic organization of concepts, such as tree-like structures with starting points and well defined paths. However, this paradigm is inherently inefficient to train students in interdisciplinary fields because: (i) students in these settings usually come from different disciplines, thus having different (often non-overlapping) backgrounds, and (ii) curricula in interdisciplinary fields are comprised by subject matters drawn from different, often traditionally disconnected, areas. This is true particularly in the field of systems biology, where students need to master a biological problem, know the theory of dynamical systems, probability, statistics, and be able to program, just to mention a few subjects. Students who take this class at the senior undergraduate and junior graduate levels usually major in biology, mathematics, computer science, statistics, physics, and engineering. As a result of this multilateral skill requirements and the inherent diversity in an interdisciplinary class, some students in the classroom have expertise in some areas and deficits in others. Students from underrepresented communities are particularly challenged in the process of social integration into a community of learners. Therefore, early identification of weaknesses of this particular set of students is critical to ensure that they do not drop from STEM training programs. This project seeks to development of an open-source Web-based cyber-learning tool that allows a team of instructors spanning several scientific disciplines to curate a constellation of interdisciplinary learning resources for the purpose of creating individualized development plans for students. The personalization of the learning plan or syllabus for each student depends on previous knowledge and individual learning goals. This customization is achieved through an information system called ALICE (Adaptive Learning in Interdisciplinary Collaborative Environments), which connects specific units of knowledge (lexias) though a dynamic path and presents it to the student for the purpose of acquiring a set of competencies. ALICE personalizes education by: (1) creating a knowledge map of course material that is unique for each student (personalized syllabus); (2) suggesting individualized paths across the knowledge map based on student competencies/accomplishments; (3) providing accessible Web-based interfaces for students and instructors for storing and presenting class materials, for assessment, and for recording student paths; (4) establishing social networks for collaborative learning of course material through shared problem-solving tasks and for passing the research learning experience from current students to future ones; (5) informing instructors about progress of students and identifying early students at risk. As such ALICE departs from current systems of adaptive learning in that it is centered on competencies that cross disciplinary barriers, thus dissolving artificial boundaries between subject matters and creating opportunities for true interdisciplinary training. We achieve this adaptivity via a mathematical model that takes into account the user's previous knowledge to minimize cognitive overhead in learning new material.