Past TI Summer Projects

Student: Ella Gillespie

Advisor: Dr. Shruti Mahajan (Mary H. '80 and Richard K. Templeton '80 Assistant Professor of Computer Science)

Anxiety management is not an uncommon topic of discussion for college students. The combination of course load, transitioning to adulthood, planning for the future, and related personal trials creates an environment that breeds anxiety. There are a number of tools that we can use in everyday life to relieve this anxiety, including practicing mindfulness, cognitive behavioral therapy, medications, and so forth. In the digital age, mindfulness apps such as Headspace serve as an in-your-pocket journal and meditation guide. While these apps can be helpful on a case-by-case basis with consistent usage, they have two major shortcomings. First, the entire rest the device can be a distraction for the user. Additionally, apps cannot guarantee an appropriate environment for best mindfulness practices. During my summer research, my advisor and I worked towards creating an experience that can fill these gaps. Our controlled environment was designed to act as a sort of mirror: a user sits in a room watching visuals that work to match their levels of anxiety. Over time, the visuals change to a calmer state in order to suggest a change in mentality. Currently, we are testing two types of visuals and are working towards incorporating other sensory stimuli.

Student: Sophia Pegolo

Advisor: Dr. Kristina Striegnitz (Associate Professor of Computer Science)

As a student researcher in Human-Robot Interaction (HRI), my project explored how non-linguistic utterances (NLUs), such as beeps, tones, or synthesized instrument sounds, can be designed to improve communication between humans and robots. While facial expressions and language have been widely studied as expressive tools in HRI, NLUs remain relatively under explored despite their perceived emotional effectiveness in robotic media characters like Wall-E and R2-D2 as well as their strong connection to music, something known to successfully convey emotions. Our work draws inspiration from these expressive models and investigates how sound-based signals can support conveying emotions and intentions in robot communication without relying upon human mimicry.

This summer, I had the opportunity to take an academic deep-dive into this field and design a study based around our on-campus social robot, Valerie. Through a survey, participants were presented with three different character descriptions, a variety of human-robot interaction scenarios, and an array of different instruments and synthesized sounds to play the NLU melodies. These sounds represent different potential “voices” for our characters, and we have the participants assess which sounds fit the best, given the context of character and scenario. Establishing an effective voice is crucial for this future work’s significance. Though not yet conducted, this study hopes to lay the groundwork for future exploration of NLUs as tools to make robot communication more intuitive, emotionally rich, and effective.

Student: Shriya Biswas

Advisor: Dr. Shruti Mahajan (Mary H. '80 and Richard K. Templeton '80 Assistant Professor of Computer Science)

My project combined the disciplines of Cognitive Neuroscience and Human-AI interaction. The research goal is to understand whether humans can identify a human-written text from an AI-generated text and how they evaluate this. What are the decision-making processes, and what factors and biases affect their judgment? In the study, participants will be asked to read several human-written and AI-generated texts, identify whether they think the text is human-written or AI-generated, rate their confidence, and explain their answers briefly. To evaluate which factors have an effect on participants’ responses and confidence, the collected texts vary in terms of narrative perspective (first vs. third person), writing style, factual accuracy, and topic. The results of this study will show how well people can recognize AI content in everyday situations and can be used to guide the development of policies around AI transparency.

This summer, I designed a mixed-methods study, which includes a survey and semi-structured interview. I worked on collecting text samples: AI-generated as well as human-written. AI-generated texts are from ChatGPT, and then human-written texts from various blogs, journals, articles, opinion articles, Reddit, and academic essays. After the survey is concluded, invitations for a 90-minute interview will be sent to a subset of participants who will be selected in a way that ensures variation in detection accuracy. The aim of these interviews is to explore the neural and cognitive mechanisms underlying participants’ judgments of AI versus human-authored texts, focusing on how bias, cultural and ethical considerations, and perceptions of human-like qualities shape their decision-making.

Student: Muhammad Ziayan Omer

Advisor: Dr. Kim Plofker (Associate Professor of Mathematics)

My Summer 2025 research lies at the intersection of mathematics, computer science, and history, revisiting one of the most fundamental problems of premodern trigonometry: the computation of the sine of one degree. Since antiquity, this value had remained a crucial missing piece, as Hellenistic Greek geometry could not progress beyond the sine of three degrees. It was essential for astronomers building accurate observational tables and became the focus of the innovative work by fifteenth-century scholars in Timurid Samarqand (modern-day Uzbekistan). At a time when most mathematics was grounded in geometry, mathematician-astronomers Jamshīd al-Kāshī and Ulugh Beg broke new ground by introducing iterative algebraic methods to address the problem. As part of my research, I reconstructed their ingenious methods and created a Python-based calculator that operates directly in the sexagesimal system they used. By reproducing their algorithms in digital form, I was able to compare their relative efficiency, reduce the heavy burden of manual calculation, and make these procedures accessible to modern researchers.

The project not only supported a recently published scholarly book but also showed how forgotten algorithms can be re-animated to become part of a compelling dialogue about what counts as mathematical progress, about how many paths there are to rigor, and about the plurality of traditions that underlie today’s computational culture. To bring these voices back into view is to expand the story of mathematics itself. Bridging the disciplines of the sciences and humanities is key to broadening our understanding of what counts as scientific progress and where its roots lie.

Student: Ushna Khan

Advisor: Dr. Sarina Kuersteiner (Assistant Professor of History)

During my eight-week research collaboration with Professor Sarina Kuersteiner funded by the Templeton Institute, I delved into the intriguing origins of the word “risk” (Latin resicum), which some scholars propose stems from the Arabic word “rizq,” meaning divine provision, livelihood, sustenance, and more. Through Professor Kuersteiner’s study of resicum, we explored the idea that this term was not merely adopted by Europeans due to economic necessity or legal impediments but rather as a result of cultural exchanges with merchants from the eastern Mediterranean. This research has broadened our understanding of how legal institutions and economic practices were influenced by the cross-cultural interactions of medieval traders.

Student: Emmett Parkerson

Advisor: Dr. Carolyn Rodak (Chair, Department of Civil and Environmental Engineering)

As a student conducting research under the support of the Templeton Institute, my project involves multiple disciplines including environmental science and engineering as well as the social sciences and humanities. I worked with publicly available wastewater release event data displayed on the New York State Department of Environmental Conservation (NYSDEC) website, which served as the basis for my research. Of my three research phases, the first was to clean and organize the NYSDEC website data in a way that made it presentable and easy to follow. The second phase consisted of calculating and evaluating the three metrics of reliability (occurrence), resilience (duration), and vulnerability (volume), known as RRV. Phase three, which I am yet to start, will involve developing new environmental justice indices using the mapping capabilities of ArcGIS spatially to analyze wastewater release data layered with other data from New York State, such as demographics, average income, population, and more. My goal is not only to make the wastewater release data easy to comprehend but also to use it to display noteworthy trends, such as which months show comparatively higher or lower RRV values and how New York State’s overall RRV is trending over the past eight years.

The idea is that if my improvements of New York State wastewater release data (including raw, processed, and GIS data layered with demographics, etc.) could be made available to the public, this increased public awareness would lead to more wastewater system improvement demands, thus providing a greater incentive for the state to invest additional funding in substantial infrastructural improvements, especially regarding any trends with marginalized communities and higher instances of failing sewer systems. Infrastructural improvements regarding our sewer systems and treatment plants begin with identifying the source, and this sewage discharge data can now be used efficiently and applied to the underlying problem: fixing the wastewater infrastructure of New York State.