- Equity and Engagement in STEM Undergraduate Research
- Comparative Analysis of Steroid Mediated Neuroprotection
- Understanding minor lipid biosynthesis and function in Escherichia coli
- Creating an Experiment Builder for the Cloud (Option A)
- Measuring Statistical Learning with EEG (Option B)
- Modelling Pollinator Interactions and Management
- How do Prey Respond to Indicators of Relative Predation Risk?
Equity and Engagement in STEM Undergraduate Research
Professor David Bradley, Physics & Astronomy
This project will collect and examine demographic data and student/faculty survey data to determine current levels of equity and engagement in science, technology, engineering, and mathematics (STEM) undergraduate research programs such as URSI and DIR. The project will also explore possible improvements to increase this equity and engagement. Methods will include both qualitative and quantitative approaches, and will be well-grounded in science education research. Students interested in applying should be interested in science and issues related to diversity, inclusion, and equity.
Comparative Analysis of Steroid Mediated Neuroprotection
Professor Kelli Duncan, Biology
Following injury to the brain, a subset of steroid hormones (androgens, estrogens, and progestins) act neuroprotectively. This hormone-mediated promotion of neuronal survival and repair occurs via alterations of second messengers, regulation of apoptotic and anti-inflammatory genes, and upregulation of neurotropic factors. However, if you look across vertebrates, some vertebrates are more adversely affected by neuronal damage than others. Specifically, work in the zebra finch brain has shown that zebra finches have a faster and more robust steroidal response to injury than other more commonly studied rodent models. The goal of this study is to examine both the expression and effect of steroid hormones and their receptors following brain trauma in different vertebrate model organisms (Zebra finches, Japanese quail, and Mice). Additionally, we also plan on examining the activation of downstream steroid-induced targets following TBI to determine if different organisms activate different targets following TBI (Traumatic Brain Injury). We propose that studies of these species will help in our understanding of the underlying and potentially common mechanisms that regulate the steroid mediated response to neuronal damage. This project will be conducted as a collaboration between neuroscientists from the Biology and Psychological Science departments, as well as URSI fellows; a DIR student would be strongly welcome.
Understanding minor lipid biosynthesis and function in Escherichia coli
Professor Teresa Garrett, Chemistry
The project would investigate the phenotypic changes that occur to Escherichia coli when the lipid composition is altered. The DIR student would construct novel bacterial strains, grow cells and investigate a number of phenotypes using fluorimetry, mass spectrometry, and colorimetric assays. We hold weekly lab meetings where students present data and ask questions of each other.
Creating an Experiment Builder for the Cloud (Option A)
Professor Josh de Leeuw, Cognitive Science
Researchers in cognitive science and related disciplines are increasingly using the internet as a means of running behavioral experiments. Online experiments can reach larger and more diverse samples than laboratory experiments, and are one way to solve the pervasive issue of underpowered experiments in the behavioral and cognitive sciences. This project will continue development of a graphical experiment builder that allows researchers to create experiments with jsPsych (www.jspsych.org) without any background in programming. We will develop both frontend and backend components for the experiment builder, focusing on tools that facilitate open sharing of experiment materials. We will also create tutorials and sample experiments. If progress permits, we will conduct a usability study of the software.
Measuring Statistical Learning with EEG (Option B)
Professor Josh de Leeuw, Cognitive Science
Statistical learning is the ability to discover regularities in sensory input without explicit guidance, an ability which is central to many cognitive phenomena. Despite the importance of statistical learning, there is no widely accepted theory of the mechanism(s) involved. One kind of evidence that is lacking, for a variety of good reasons, is information about how this learning unfolds over time for an individual learning a particular pattern. One possible way to measure this learning is by recording the brainwaves of people while they learn. In this DIR project, we will use electroencephalography (EEG) to record changes in brainwaves as people learn. Then we will explore machine learning techniques for analyzing the data and discovering what patterns change as people learn.
Modelling Pollinator Interactions and Management
Professor Benjamin Morin, Mathematics & Statistics
The DIR project will involve a combination of interviews, ecological research, and mathematical modelling. We will investigate the role of bee pollination on local crops (e.g., apple orchards), the role of pest management on the agricultural systems, and form a few simple mathematical models that can possibly lead to insights on the impact of such controls. In addition to learning a little math and using software such as Excel to solve these models we will make some trips to vassar farms, a local orchard, and possibly get to interview and discuss strategies with local farmers and extension agents in order to motivate our modelling investigation.
How do Prey Respond to Indicators of Relative Predation Risk?
Professor Justin Touchon, Biology
Tadpoles of the Neotropical treefrog Dendropsophus ebraccatus have a remarkably flexible development wherein they grow different tail shapes and colors if raised with different predators. A tadpole raised in an environment with dragonfly larvae will grow a large red tail, whereas a tadpole with fish will grow a small clear tail. However, we know relatively little about how much prey animals (such as tadpoles) can develop nuanced responses to variation in predation risk. Using my research colony of D. ebraccatus here at Vassar, my DIR student will execute from start to finish an experiment where tadpoles are raised with varying concentrations of the chemical cues which indicate fish or dragonfly predators, thereby allowing us to test how well tadpoles can fine tune their response to risk. This project is a direct follow-up to a senior thesis I am currently advising which has produced extremely intriguing results, and the DIR student will be involved with the entire experiment, from breeding the frogs to collecting their eggs and rearing the tadpoles with varying predator cues, to photographing the tadpoles and measuring their morphology at the end of the experiment.