I completed my PhD in UCSD's Cognitive Science Department, advised by Jeff Elman, in March of 2014. My research revolved around the cognitive-computational mechanisms involved in language processing, with particular emphasis on syntax and second language acquisition. Later, my research interests expanded to include the use of technology in educational contexts more generally.
We are experiencing unprecedented global flow of knowledge and ideas, and seeing emergent opportunities for ad hoc educational experiences that are mediated by computer technologies and the Internet. Educators and learners are experimenting with new pedagogical approaches and modes of content delivery, both inside and outside of the classroom, in what has been described as an "online learning revolution".
In this dissertation, I take a step back to examine the factors that are enabling this transformation, and demonstrate how not everyone has been benefiting equally -- in particular, recent developments have been failing to include many communities that are already among the most economically and socially disadvantaged. I present the work we have done to help bridge this gap, through designing, building, and deploying tools to allow learners in offline and bandwidth-constrained contexts to access the same types of novel educational opportunities as their more connected counterparts.
Develop a web-based language learning and teaching environment that combines student modeling, natural language processing, and multimedia computer-mediated communication systems with crowd-sourced content and social web interfaces.
Use performance data to model ESL student knowledge and infer the dependency structure underlying English grammar acquisition, and use these models to develop efficient diagnostic tools for assessing a learner’s knowledge state, predicting future performance, and identifying appropriate study materials.
Accumulate a corpus of text produced by ESL learners, accompanied by structured corrective annotation from native English-speaking tutors, to be used both for qualitative error analysis and for training automated error-detection/correction systems.
Abstract: Recursive structure is viewed as a central property of human language, yet the mechanisms that underlie the acquisition and processing of this structure are subject to intense debate. The artificial grammar learning paradigm has shed light onto syntax acquisition, but has rarely been applied to the more complex, context-free grammars that are needed to represent recursive structure. We adapt the artificial grammar serial reaction time task to study the online acquisition of recursion, and compare human performance to the predictions made by a number of computational language models, chosen to reflect multiple levels and types of syntactic complexity (n-grams, hidden markov models, simple recurrent networks, and Bayesian-induced probabilistic context-free grammars). Evidence is found for a dissociation between explicit and implicit mechanisms of sequence processing, with the SRN more highly correlated with implicit performance, and the PCFG more correlated with explicit awareness of the sequential structure.
2007: Max/MSP Model of Synchrony-Inducing Conjunctive Binding Nodes
|As an undergrad research assistant, I built a proof-of-concept model of Hadley's (2007) theory of synchronous neuronal firing as a byproduct of the activity of conjunctive binding nodes.|