Research Question: Do technology-driven personalized learning experiences using macaronic language improve students’ learning of a foreign language?
Interdisciplinary Approach: This project bridges educational theory and novel techniques from artificial intelligence engineering and machine translation to develop a new web-based foreign language-learning platform.
Potential Implications of Research: The research will produce novel, technology-driven learning experiences that, if successful, will provide new means of teaching foreign languages.
Project Description: Serious obstacles to learning a new language -- such as German -- are mastering the sheer size of the vocabulary and the rules that govern the language (“syntax”). But imagine that you began to learn German by reading an “English” novel that included German syntax. The experience might be something like reading the dialog of the Star Wars character, Yoda. For example, you would encounter sentences such as “I would gladly a bread loaf eat.” Without much effort, you would start to become accustomed to the feel of German word order and usage. As you continued to read the story, the English words would gradually be replaced with German words when the meaning was apparent from the context. Common or useful German words would be introduced first. Thus the story might progress from “ein Loaf of Bread” to “ein Loaf Bread” to “ein Loaf Brot” to “ein Laib Brot” or “ein Brotlaib” (the last two are proper German).
The aim of our project is to design, deploy, and evaluate a new technology for personalized foreign language learning. We will design a website that allows students to choose what they wish to read and write about. Upon accessing the materials, the website will gently introduce the learner to German through macaronic text -- a deliberate mixing of two languages. The macaronic text will be based on real German writing (e.g., novels, newspapers, research papers) that has been incompletely translated to English, serving as a crutch for the native English speaker. To ensure that the macaronic sentences are reasonably natural, the website will use code-switching principles that are based on the phenomenon of bilingual conversants’ switching casually between languages in mid-sentence. When needed, learners will be able to hover over the macaronic phrase and see the English and German versions. Moreover, learners will be engaged periodically as they read. For example, the website may ask the learner to fill in a blank in a sentence or write a few sentences in macaronic text. The website will include a scoring system that will determine where the learner’s answers fall along the beginner-to-expert continuum. Based on new machine translation technology and machine learning technology, the website will automatically adjust the presentation of macaronic text and provide precise, individualized feedback to the learner to gently scaffold future learning. By introducing the learner to a foreign language in this way, we will provide a shallow learning curve that we expect will lead to more motivation and more enjoyment of the learning process.
We will evaluate the new technology in two studies. The first study will examine the relationship between the learners’ website use and improvements in the their foreign language skills over time. Additionally, we will collect and analyze detailed information on each learner’s interactions with the website, such as the type of text the learner chose to read, how much time was spent reading the text, the number of times the learner requested assistance, and the types of errors the learner made during interactive sessions (e.g., fill in the blank or writing activities). The second study will examine how website usage differentially promotes students’ foreign language learning compared to traditional classroom activities. Taken together, this project will provide new machine learning methods that will assess and adapt to the learner. Moreover, it will serve as an exemplar of technology-based personalized foreign language learning.