Participation of High School and Undergraduate Students who are Deaf in Research on American Sign Language Animation
Matt Huenerfauth · 2010 · SIGACCESS Accessibility and Computing, Issue 97 · doi:10.1145/1873532.1873534
Summary
This SIGACCESS newsletter article describes the educational-outreach component of a five-year NSF CAREER Award (#0746556) research project at the Linguistic and Assistive Technologies Laboratory (LATLab) at Queens College, CUNY. The scientific goal of the project is to advance the state of the art in animations of American Sign Language (ASL) by recording native signers with motion-capture gloves, body suits, cameras, and eye-gaze trackers; annotating the resulting corpus for linguistic phenomena; and using machine learning to build models that can drive a virtual signing character. Such technology matters because roughly half a million deaf people in the U.S. use ASL as a primary language, and the majority of deaf U.S. high school graduates read English at or below a fourth-grade level, making text-based web and software content inaccessible. The author describes how the project intentionally integrates deaf high school and undergraduate students as paid research participants each summer. Two to three deaf high school juniors or sophomores from greater New York City deaf and mainstream programs spend three months at the lab, and a visiting Gallaudet undergraduate was hosted through an NSF REU supplement. A deaf native-ASL research assistant — a former science teacher — coordinates recruitment and co-supervises the students, creating an ASL-immersive laboratory environment. The article is written as a practical guide for other SIGACCESS researchers considering similar outreach.
Key findings
The summer program is built around research tasks where native-ASL fluency is an actual expert skill rather than a side attribute: students annotate motion-capture recordings, script signs in 3D animation software, participate in evaluation sessions with community ASL signers, and tour motion-capture hardware used in film and games. A 'tiered mentoring' structure pairs the high school students with the visiting Gallaudet undergraduate, who arrives two weeks early, helps plan activities, and leads weekly annotation-arbitration meetings. End-of-summer evaluation used a modified SURE II instrument; three of four respondents were 'very satisfied,' and self-reported learning scored 5.0/5 on understanding the research process, laboratory techniques, and interpretation of results, with most other items above 4.0. Both high-school juniors who participated in summer 2009 intended to enroll in college the following fall. Students reported learning about the linguistics of their own language and about careers in computing; they also asked for more direct hands-on use of the motion-capture equipment. The author explicitly flags the tiny sample size as a caveat. Contextually, the CEOSE 2004 report cited in the paper documents that students with disabilities pursue science careers at below-average rates.
Relevance
For accessibility practitioners and researchers, this article is most useful as a replicable model for combining scientific research with genuine pipeline-building for disabled students in STEM. Key transferable principles: ensure the students' lived expertise (here, ASL fluency) is load-bearing for the research rather than decorative; match communication modality to participants (ASL as the primary lab language, not just English with interpreters); use tiered mentoring so students see role models 'one step ahead'; and build recruiting partnerships with local deaf high schools and Gallaudet. It also illustrates why participation of native ASL signers is not merely ethical but methodologically necessary for sign-language technology research. Limitations include the tiny sample and the self-reported nature of outcomes; the article predates more recent advances in neural ASL generation and community-led critiques of signing avatars, so its framing of ASL animation as an accessibility win should be read alongside current Deaf-community perspectives.
Tags: American Sign Language · ASL animation · deaf community · STEM education · research outreach · tiered mentoring · signing avatar · motion capture · assistive technology · undergraduate research · broadening participation