MEDIA, Pa. — During the past academic year, computer science major Renusree Bandaru was an active student—spending her time serving as president of the women in engineering club, participating in the multicultural club and, of course, staying on top of her classes. But the end of the school year didn’t slow down Bandaru, who participated in Penn State’s Multi-Campus Research Experience for Undergraduates (MC REU) program this past summer.
The MC REU program allows undergraduate students in the College of Engineering to conduct research with mentors from across the University — one from the student’s home campus and one based at University Park.
For Bandaru, who attended Penn State Brandywine and is currently in the process of transitioning to University Park, this multi-campus project was the perfect fit. Brandywine Assistant Professor of Information Sciences and Technology Martin Yeh served as her home campus mentor, while industrial and manufacturing engineering doctoral student and graduate assistant Hannah Nolte served as her University Park mentor.
“My research project focused on trying to better understand an individual’s learning stage by detecting the difference in electrical signals in the brain using machine learning techniques,” Bandaru said. “One of the main goals of this research was to identify a model that can be used to analyze the EEG data efficiently.”
Bandaru explained that learning has been described by cognitive scientists as having three different stages — declarative learning, transition learning and procedural learning. Her research uses various types of machine learning technology to interpret electroencephalography (EEG) data that allows scientists to identify a person’s learning stage. Identifying learning stages can be correlated to an individual’s mastery level and help scientists better understand how people learn.
“The first stage of cognitive learning is called declarative learning, where one starts to learn,” she said. “The third stage is called procedural learning, where one is considered an expert. And the second stage is a transition from declarative to procedural.”
Bandaru spent approximately 40 hours per week working on the research project, which took 10 weeks to complete. Project deliverables included a poster presentation at the virtual MC REU Research Exhibition and a scholarly research paper.
“This innovative project seeks to explore whether an individual’s learning stage can be discovered from human brainwave patterns,” Yeh said. “Machine learning techniques are used when data is too complex for humans to understand. Brainwave signals are an example of such a situation. Leveraging machine learning can have a transformative effect on how we analyze human brainwaves in learning and in other situations.”
The team’s research project provides opportunities for an evidence-based prediction of learning outcomes, especially in remote-learning settings where instructors cannot observe learners in person. It also could help educational technologists create more effective computer-based learning systems and pinpoint when to teach new materials and when to revisit learned materials.
“Renusree’s biggest assets as an undergraduate researcher are her willingness to explore options of a problem and to ask questions,” Yeh said. “Her dedication and resilience allowed her to take on more complex topics and learn from them.”
“Completing a research project like this as an undergraduate student is valuable in so many ways,” Bandaru said. “It can help students identify a career path, network with professionals in their field of study, and stand out when it comes time to secure an internship or job.”