Gloss Type
Articles by Chun and Plass (1995, 1996) as well as Davis and Lyman-Hager (1997) were used to help determine the type of annotation/gloss to be used. These articles provided details on the differing interacting factors that correspond with the media types to be annotated. As a result, where appropriate, student glossing options for a French word or expression consisted of: a written definition (French only), pronunciation or narration (by a native speaker), a video clip and/or a picture/graphic. Clicking the ěPlay this pageî or the ěPlay until endî option resulted in the entire text being ěread toî the child by a native speaker. Chun and Plass (1995) quote research from Hulstijns (1992) and Kellogg and Howe (1971) which proposed avoiding English translations. This warning, coupled with the fact that the students had received a limited amount of formal English instruction up to the time of the treatment (see Implications for Design), resulted in only French definitions being featured.
Multimedia Elements
Audio (pronunciation and Play this page), video clips, written definitions and still pictures (graphics) for use in the treatment were created and digitized by the author of this study on a Power Macintosh 65/250 computer. A 15 year old female from the French Islands of St. Pierre and Miquelon recorded the pronunciations and play this page through the aid of two pieces of shareware, Sound App and Sound Machine.
Three video clips, without sound, were created on an inexpensive, digital camera using QuickTime (see Glossary) software. A 13 year old French Immersion student from the community performed the actions for the videos ěse moucherî (to blow one's nose), ěéternuerî (to sneeze) and ělíessuieî (líessuie referred to wiping the nose with the sleeve). Each video had a length of approximately 6 to 8 seconds. The actions and performance on the video clips were clear and unambiguous. Written definitions were drawn from Le Petit Robert, dictionnaire de la langue française. For the still-pictures (graphics), a variety of Clip-Art images were used. It had also been intended to provide animated gifs for certain words and expressions but technical problems prevented their use.
The media, once created, were linked to the text with the multimedia
authoring software called GALT. (Glossing Authentic Language Texts) Mary-Ann
Lyman-Hager, one of the creators, explains its function,
ěThis software will allow words, phrases, or groups of words on a ëpageí
to become hyperlinked to digital media or simple text glosses. The selection
of a difficult word or phrase activates one or more of the seven explanatory
categories that can be annotated with text, picture, digital video, digital
audio, and a relationship tree. Additionally, GALT provides an audio option
so that the entire passage may be "read" to students.î (Lyman-Hager, 1997b,
p2)
The Footnote Controls were modified for this study to offer only the 4 hyperlinked choices previously mentioned: definition, pronunciation, video and graphic. The Audio Controls allowed the user the options: Play until end, Play this page, Pause and Continue. (see Appendix 6) Appendix 7 provides a sample of the definition option, Appendix 8 shows a still image of the video cue for ěessuieî and a graphic sample is provided in Appendix 9. To facilitate the completion of the first objective of this study: ěto track the multimedia choices made by 8 and 9 year old L2 readers according to modality preferenceî, the software recorded individual media cues selected, showed the order in which they were used, what media were overlooked and the time spent in the session and with each cue.