This Craft of Research

James Foster

A dual passion for language and computer science guides both my scholastic pursuits and the elegance I find in everyday experiences. I want to answer questions such as, how does language shape the way we divide the world into concepts? How is meaning carried by the dark suggestive squiggles that mark the page in front of you? The quality I look for in daily life is the intricate pairing of form and meaning, blending, and self-similarity, like the fractal geometry of nature.

        Broadly, I am interested in computational approaches to natural language and cognition. In particular, I am eager to continue studying metaphor understanding and analogical reasoning, and the component processes of reminding, abstraction, domain mapping, and inference. Although my research interests are diverse, there are many specific questions I intend to investigate computationally and experimentally. For example, how are we reminded of past situations similar to our present circumstances? How do we use analogies to learn and reason about unfamiliar domains? Are the different senses of a polysemous word related by analogy? What is the role of abstraction in domain mapping? Ultimately, my top-level goal is to develop a unified model and theory of cognition using tools from computer science.

        In order to accomplish these goals, I plan to incorporate methods from computer science, psycholinguistics, mathematics, and statistics. My natural curiosity and academic background in these relevant disciplines have prepared me to answer such difficult questions.

        My primary focus at the University of Florida (UF) is computer science. To complement core classes like data structures and operating systems, I have tailored my technical electives to my research interests. For example, this semester I chose to implement my AI term project in Lisp. Dr. Douglas Dankel’s class has given me a rich introduction to the many open problems in modern AI research. And last year I took a combined undergrad/grad level course in computational neuroscience taught by Dr. Arunava Banerjee, in which I discovered just how much we don’t know about the brain.

        A fascination with software system design has helped me to excel in object-oriented programming (OOP) and design patterns electives. These two courses are (in)famous among UF CSE students for their workload and conceptual difficulty. Yet even among the highly motivated students who elect to endure these courses, my knack for abstract reasoning helped me earn the top grade among the 29 who made it through OOP last Spring. The vocabulary and structures of software design seem to lend themselves naturally to the study of human reasoning and learning. I plan to use this training to test how far the mind-machine metaphor holds.

        In addition to experience with imperative and functional programming paradigms, I have enjoyed coding in declarative languages like SQL. My current databases class is an ideal balance of theory and practical experience with modern database systems. An in-depth understanding of relational and ER models will help me design better knowledge representation schemas, such as semantic networks for NLP applications like word sense disambiguation.

        An interest in the mechanics and origins of natural language led me to join the University Scholars Research Program where, under the guidance of my advisor, Dr. Hana Filip, I have continually refined my knowledge of metaphor and analogy, polysemy and ambiguity, and abstraction. In February I will present my review of Gentner, Holyoak, and Kokinov’s book The Analogical Mind at UF’s annual Undergraduate Research Symposium. Independent study has honed my academic and organizational skills that will be crucial in graduate research.

        Last summer, I enrolled in immersive language and art classes in Barcelona, Spain to further develop my fluency in Spanish. The best part of the experience was sitting down to dinner every night with my host family to exchange hearty helpings of language, culture, and perspective. Thinking and communicating in another language has been one of my most rewarding endeavors. I think machine translation can be framed as a problem of representing analogous meaning across languages.

        The language of mathematics intrigues me as much as the language of Spain. A course in discrete structures during my sophomore year left me with a taste of the power of set theory and logic. I will be deepening my understanding of these topics next semester in the math department’s proof-based sets and logic course. I also think concepts from linear and abstract algebra will promote the study of analogical domain mapping.

        My first encounter with the phrase ‘cognitive science’ was in a diagram in the front pages of a cognitive psychology book. I was amazed at the circular figure, which seemed to unite all of my favorite things in two words. Two of my most entertaining textbooks are Psychology and Social Psychology by David Myers. And it is always a treat to watch an episode of Philip Zimbardo’s Discovering Psychology.

        My background in industrial and systems engineering (ISE) includes courses in probability, statistical theory, system simulation, and decision support systems. I completed a professional internship in ISE at Walt Disney World where I supported distribution and textile services. My projects analyzing inventory counts and measuring throughput in Disney’s giant warehouse laundries gave me practical experience in proper study design and raw data collection.

        Aside from my courses, research experience, and professional work, perhaps the most important lesson I’ve learned in college is the utility of writing. Putting inchoate thoughts into well-reasoned words helps me clarify and remember what I learn and observe. My journaling process began as an outgrowth of scribbled notes in the margins of Gödel, Escher, Bach by Douglas Hofstadter, and has become an outlet and a source of creative ideas and observations. Now, I write down everything. If I have a question or a hunch about my studies, I immediately record it in a research notebook for future reference. I also write down many of the linguistic curiosities I encounter in humor, conversation, and advertisement each day.

        My driving motivation is a deep love of learning and teaching. Teaching is how I learn best. My experience as a TA for an introductory programming class at UF helped me show new students how to think more like computer scientists. I was the only TA of four who taught the class in lecture, and I delighted in answering questions during office hours. Articulating a complicated idea forces me to challenge and refine my own understanding of it. For example, I still vividly recall the late nights at Florida Boys State spent explaining and debating the implications of the double slit experiment.

        Professionally, my goal is to become a well-respected professor. Through research and teaching, I want to discover and share new ways of looking at the world, and encourage students to find and pursue their own quests.

        I am ready to join a PhD program that fits my interests and style. I am looking for a community with similar goals, yet diverse interdisciplinary backgrounds. A place where research doesn’t stop at the laboratory door, but rather spills over into the hallways, cafeterias, and lecture halls. A place where I can study not only how we think, but also how we think about thinking. Most of all, a place where the research is rigorous and the people have a good time doing it.





wordpress statistics