We studied how referents are introduced in oral discourse under various production conditions.
Children aged 7 to 11 and adults told stories from «silent» two-character comic strips to a same-age peer who did not know and could not see them. The comic strips varied as to: 1) the frame presentation mode; 2) the explicitness of the event sequence across frames ; and 3) thematic continuity. We analyzed the ways in which the characters were named when introduced into the discourse for the first time.
The results indicated: a) a progression in the speakers' referential and narrative skills, as shown in previous studies : the characters were usually marked as « new » referents by the age of seven, but it was not until after nine years old that intra-linguistic marking became systematic ; b) a presentation mode effect : showing the entire comic strip on a single page triggered more narration, whereas presentation of the frames one at a time in a booklet led to more descriptions at age seven; and c) an effect of frame sequencing explicitness : not very explicit sequencing led the 11 -year- olds and the adults to express the assumed links between the characters as soon as they were introduced into the discourse.
The production conditions manipulated here turned out to be a good means of assessing the solidity or lability of the acquired skills.