Best practices for conducting evaluations of sign language animation

Abstract

Automatic synthesis of linguistically accurate and natural-looking American Sign Language (ASL) animations would make it easier to add ASL content to websites and media, thereby increasing information accessibility for many people who are deaf. Based on several years of studies, we identify best practices for conducting experimental evaluations of sign language animations with feedback from deaf and hard-of-hearing users. First, we describe our techniques for identifying and screening participants, and for controlling the experimental environment. Finally, we discuss rigorous methodological research on how experiment design affects study outcomes when evaluating sign language animations. Our discussion focuses on stimuli design, effect of using videos as an upper baseline, using videos for presenting comprehension questions, and eye-tracking as an alternative to recording question-responses.