"Compute-To-Learn" Bridges Classroom to Real-World Experiences

Most STEM students are familiar with standard projects conducted over a term: the instructor poses a certain problem and then the students find a solution within a specific set of guidelines and constraints. However, many students have little experience designing their own projects. That classroom experience is unlike the working world where the students need to ask themselves questions few others have asked and create their own unique approach to finding the answers.

The “Compute-to-Learn” program in Michigan Chemistry is designed to help bridge the gap between classroom and real-world experiences. It aims to enhance student learning with hands-on and interactive projects. Importantly,
it gives students another opportunity to experience the scientific process, says Professor Eitan Geva who initiated the program as an honors option to “Physical Chemistry Principles and Applications” classes offered each semester.

“Compute-to-learn” is led by Chemistry Graduate Student Instructors (GSIs) and undergraduate students from previous semesters who serve as peer mentors and is an addendum to the 200-level physical chemistry courses. At the beginning of the term, students are taught the basics of Mathematica, a programming language used by Wolfram-Alpha, and are provided with a series of prompts. Each prompt summarizes a key concept in physical chemistry, such as phase transitions and heat capacity, and poses possible questions for groups to answer. Once each group has selected a prompt and proposed their own ideas, they design a program that models their solution, complete with interactive demonstrations.

One example of a student-developed demonstration is “Evaporation of Water from a Wet T-Shirt,” an interactive module that allows the user to alter the temperature, drying time, wind speed, and location (which affects the pressure) to see how much water is left on the t-shirt and calculate the evaporation rate. Another demonstration is “Simulating Gas Exchange in a Model of Pulmonary Fibrosis,” which displays how the alveolar wall tissue thickness, caused by pulmonary fibrosis, is related to the diffusion rate of oxygen and carbon dioxide.

“This is different from doing homework assignments and taking an exam,” Professor Geva explains. “We’re trying to mimic some aspects of research but still be a part of the class and give students skills many of our classes don’t cover: working on open-ended problems, actually making up their own problems, working on a strategy to finish in a timely manner. You don’t know how to do this in the traditional curriculum.”

Student-led projects, the use of modeling and technology, and peer-led instruction have been proven to be effective tools in chemistry education and retention [See the links below.] and “Compute-to-Learn” cohesively melds all three of these ideas to create a peer-led program of technology-based student-led projects.

Throughout the term, evaluations are given, allowing students to glean perspectives from actual scientists and to be exposed to multiple approaches for problem solving. At the end of the semester, students demonstrate their models to the class and submit them for publication. In fact, many past projects have been published on the Wolfram-Alpha website.

While online publication of their projects is an important ambition for students, the main goal of the program is to give the students the opportunity to think like scientists—asking questions that are important to them and being creative in constructing their solutions, Geva says. Through “Compute-to-Learn,” students contribute to the class and are able to relate the course to their future goals in a tangible way. Professor Geva believes this educational strategy is an essential component of student development. “It’s about expanding the experience. A big part of education is outside of the classroom. I think there should be a more three-dimensional experience. Creativity is not linear.”

Kristina Lenn is a Chemistry Science Communications Fellow.