This award supports the continuation of the REU site at Ohio Wesleyan University focusing on interdisciplinary scientific computation. This site has an RET component integrated into the site. Each undergraduate participant will work on a ten-week research project under the guidance of an Ohio Wesleyan University faculty member in one of the following disciplines: astronomy, computer science, mathematics/statistics, and physics. High school science teachers will work on similar projects for a six week period. The research projects will focus on one or more aspects of scientific computation, such as: numerical (coding in a "traditional" programming language), symbolic (using a computer algebra program such as Mathematica®), and data visualization. An important goal of the site is to make participants aware of the ability of computational techniques to cross disciplinary boundaries, i.e. to be effective at solving problems of interest in many different fields of scientific inquiry. Weekly meetings, in which individual research groups will give progress reports and all participants will be able to discuss the projects, will foster a sense of community among the participants and point out the strengths of computational techniques to tackle problems in varying disciplines. A parallel computing cluster, owned by the university, will be available for participants to use on their projects. A series of weekly, half-hour, faculty-led seminars will introduce participants to widely used computational tools, such as: symbolic computing with Mathematica®, methods of data fitting, numerical techniques for solving differential equations, and more. This award is co-funded by the Division of Physics, the Division of Mathematical Sciences, and the Office of Cyberinfrastructure.
Outcomes: The project’s main objective is to give undergraduates interested in the mathematical and physical sciences experience with computational techniques as applied to cutting-edge research problems. During the summers of 2010-2014, 27 students from colleges and universities distributed around the country visited the Ohio Wesleyan University (OWU) campus to study research problems under the guidance of OWU faculty members in order to learn about computational methods (numerical, symbolic, and graphical) and how those methods can be used to solve a wide class of problems. Students, most for the first time, were exposed to an intensive, ten-week research experience that gave them an accurate representation of the nature of scientific research and also, through workshops and presentations by faculty and other students in the program, showed them how many techniques can make headway on diverse problems across several scientific disciplines. Students also receive guidance on careers in science, preparation for graduate school, and tips for giving scientific presentations. Seven faculty research mentors from the departments of Physics/Astronomy and Mathematics/Computer Science guided research students on the following projects: Stellar Surface Imaging via Light Curve Inversion. (Astrophysics) Search for Exotic Shapes in the "Wild West" of the Nuclear Landscape. (Nuclear Physics) Sampling Distribution of Regression Statistics with Data Subjected to Type II Censoring (Statistics) Artificial Intelligence for Modern Board Games. (Computer Science) Braid Group Representations and the Temperly-Lieb Algebra. (Mathematics) Synchronization in Scale Free Networks. (Biophysics) Dynamics of Josephson Junction Arrays. (Condensed Matter Physics) Students presented their results in oral or poster format at a final symposium on the last day of each summer program. The Ohio College Summer Research Symposium is held each July on the OWU campus. Summer research students from other Ohio colleges and universities (Denison University, Kenyon College, Oberlin College, Ohio Northern University, Wittenberg University, and the College of Wooster), are invited to join the OWU/REU research students at the symposium to share the results of their research. These presentations have been recorded and can be viewed online at reu.owu.edu. Click on the link "Symposia" on the left side of the page. Many REU students also presented their results at regional and national conferences such as: The Young Mathematicians’ Conference, SACNAS National Conference, Meeting of the American Physical Society Nuclear Physics Division, Meeting of the American Astronomical Society, and the meeting of the Society for Neuroscience. Most REU participants pursue graduate studies in physics, astronomy, mathematics, or computer science and are interested in a career in scientific research and teaching at the university level. Others obtain employment in their chosen professions. Society clearly benefits from the addition of talented and well-trained individuals such as these to the scientific workforce. Example Project Description: Understanding how and why networks of neurons synchronize is important to better understanding several brain pathologies, including Parkinson’s disease and epilepsy. There are two primary factors that determine the level of neuronal synchrony. The first is the structure of network connectivity. Recent findings have shown that many brain networks are connected with a scale-free structure, which feature a small population of highly-connected "hub" cells. The second factor is a cellular property known as excitability type, which characterizes how an individual neuron responds to perturbation. There are two different excitability types, Type I and Type II, and networks of Type II neurons are known to synchronize better than networks of Type I cells. However, no one had ever before investigated what happens when these two types of cells are mixed together and coupled with a scale-free structure. In particular, does the network synchronize better when the hub cells are Type I or Type II? Our preliminary results show that relatively few Type II neurons can hijack the network to high synchronization levels if they are hub cells, but if they are not hub cells you need a much higher percentage of Type II cells in order to appreciably impact network synchrony. (See accompanying figure.) These findings could guide our understanding of how epileptic seizures are generated.
