Mini-Projects

Last updated: November 23, 2009

Contact: Prof. Matthew Kam - mattkam AT cs DOT cmu DOT edu

Under the auspices of the Human Development Lab and Video Games Research Lab at Carnegie Mellon University's Human-Computer Interaction Institute, here are some brief descriptions of mini research projects that undergraduate researchers or Masters-level students can work on when getting their feet wet with research for their first time. Some projects are more research oriented, whereas others will appeal more to hackers. While we have tried our best to keep this list up-to-date with our latest ideas, there are obviously more ideas that we lack the time to write up, especially those that are more scholarly and theoretical in nature. We encourage you to identify a few mini-projects or themes that catch your interest, and to chat with us further. We have had contributors representing a diverse range of disciplinary backgrounds, including, but not restricted to: Cognitive Science, Computer Science, Design, Economics/Business, Education, Fine Arts, Information Systems, Linguistics and Social Entrepreneurship.

  1. MILLEE: Cellphone e-learning games that aim to put language literacy within the reach of children in the developing world, especially those who cannot attend school regularly due to child labor. Our design methodology is informed by best practices in commercial language learning packages and traditional village games that children in the developing world play. Since the project began in 2004, we have conducted over 10 rounds of field studies in India, spanning over 12 months on the ground. We are beginning a controlled experiment with 800 rural children in 40 villages in India, with early replication underway in rural Africa, China and elsewhere.

    1. Identify a language learning competency in English that you would like to work on. Work with us to design and storyboard an educational game for the cellphone that targets this learning objective. Implement this design in J2ME.

    2. Write a J2ME class library that processes log entries to recreate gameplay screen sequences as a video feed. The goal is to allow our cellphone games to be demo'ed over the web (by posting the video online), analyzed by game studies researchers, etc. Next, create a composite video feed by combining this feed with a feed of video recordings on how the user is interacting with the cellphone.

    3. Write an educational data mining toolkit that performs log data analysis for our cellphone games. One feature of this toolkit is to generate graphs that succinctly illustrate important points from this log data. Part of the mini-project includes thinking about what are some of the best graph visualizations for learners, teachers, parents, researchers, policy-makers and other stakeholders.

    4. Write a J2ME class library that performs simple processing of speech input on cellphones. This library is intended to be used by programmers in developing speech-enabled English language learning applications for the cellphone.

    5. Analyze some video recordings of good Indian, English instructors interacting with rural students. Identify some conversational elements that inform the design of a culturally appropriate voice-user interface for English learning on cellphones. Implement it in J2ME.

    6. Think about gender-inspired design. Specifically, think about how our cellphone games can be redesigned so as to: (i) promote collaborative learning between rural mothers and daughters, (ii) encourage rural mothers to monitor their daughters' learning progress, or (iii) encourage rural mothers to provide more support for their daughters' education. Come up with prospective designs and implement one of them.

    7. Currently women and children in developing regions are often aware of their unfavorable situations but unaware of the options they have to come out of it. Design a simulation-style game in which the user could build their household full of female/male children, different ages, occupation/livelihood, needs/desires, etc. Then the user would be able to play out the lives of the different people in the household and see the different outcomes according to the way they prioritize their lives. E.g. Daughter of age 9 is given a priority to work as a tailor rather then go to school. Her life plays out as continuing to be a tailor throughout her life. But the user can then change the priorities and see the different options: affordable private schools, micro-financing, etc - options that could help them pull themselves out of poverty if they knew the option existed and how to capitalize upon them.

  2. Playpower: $10 Computer for Educational Games and Game Programming. An 8-bit "education computer" is currently sold in dozens of developing countries for as little as $10. It uses an existing TV as a screen, and comes with a variety of software, including BASIC programming. Unfortunately, most of the learning games currently included on this computer are limited in quality. Playpower.org, an organization based at CMU and supported by the MacArthur Foundation, is developing new, higher-quality 8-bit learning games for this platform. Students in this project can contribute to the design of open-source 8-bit games and learning software for the $10 computer.

    1. Develop code for a virtual keyboard to support games for teaching touch-typing.

    2. Storyboard learning games that generate or support real-world social interactions (in the context of games played on family televisions).

    3. Storyboard learning games based on characters or themes from classic Indian literature.

    4. Develop cc65 compiler libraries that enable Playpower game programmers to develop games in C.

    5. Develop a code template that facilitates the production of choose-your-own adventure stories.

    6. Port a classic learning game, such as Number Munchers, from the 6502-based Apple II computer to the 6502-based $10 computer.

    7. Develop software framework to support the teaching of BASIC programming on the $10 computer.

  3. Low-Cost Player for Classroom Video: Manufacturers are now producing a $100 device that can store a thousand hours of video, for playback on a television. As the cost of storage continues to fall, this device is expected to cost as little as $20 in the near future. The low cost, ease-of-use, and durability of the device means that it may contribute to efforts to rapidly scale educational access in developing contexts. However, a device of this kind may also be useful to all teachers who make use of educational videos in the classroom.

    1. Developing interactive video UI designs that enhance the ability of teachers to manage large quantities of video.

    2. Design interactive video content for enhancing peer learning, group discussions, and social interaction around video content.

    3. Developing 10 hours of video-based curriculum, potentially consisting of video from Sesame Street, National Geographic, and public domain video archives. This curriculum can be developed for field testing in Pittsburgh classrooms and/or rural Indian classrooms.

We have deliberately numbered the above list, which makes it easy for interested students to refer to specific mini-project ideas when chatting with us.