Effects of language modality on word segmentation: An experimental study of phonological factors in a sign language*

This paper analyzes the word-segmentation strategies used in signed and spoken languages. The claim is that there is a strong “modality effect” in the word segmentation strategies used in these two types of languages. The experimental results show that both signers and non-signers use place of articulation and movement more heavily than handshape to make word segmentation judgments; however, signers are more sensitive to handshape than nonsigners are for making such judgments. This work shows that there are strategies for segmenting visual language input that are different from those used in segmenting auditory language input, regardless of language exposure. From the experimental evidence presented here and from the work on word segmentation in spoken language, one can conclude that viewers use a word-sized unit itself to segment strings into words, which is argued to be due in large part to the visual/gestural nature of sign languages. In contrast, listeners depend most heavily on the syllable in their word segmentation strategies, which is argued to be due the auditory/vocal nature of spoken languages. This work can at least partially explain the variation in the wellformedness constraints on found in signed and spoken languages, which capitalize on this modality effect.