Referential Communication: Effects of Listener Presence on the Performance of Young Speakers*
Nancy E. Kossan, Ellen M. Markman · Merrill-palmer Quarterly · 1981
In the standard referential communication situation, children often fail to pro vide enough information for a listener to find a target from among distractors. Though separated by an opaque barrier, the child and the listener are seated in close proximity. The listener's proximity might prevent these children from re alizing that only verbal information is useful. This hypothesis was tested, with first-graders, by comparing the standard situation to one where the listener is removed and the children told to convey their messages over citizen's band radio. Another group of children was asked to describe single isolated refer ents. Three measures of communicative effectiveness were used: (a) adults' suc cess at selecting the referent, (b) adults' confidence, and (c) amount of discrimi native information in the message. The messages of children talking to a listener behind a screen were longer, but no more informative than were those of children describing single referents. In contrast, children were able to com municate more effectively when the listener was removed. The content of the children's messages varied in ways appropriate for the medium, i.e., for a me dium by which people communicate over long distances when they cannot be physically present. The children's inability to detect that the listener does not share information with the speaker when the listener is behind the screen, compared with their ability to detect the lack of shared information in the citi zen's band radio communication situation, suggested that the standard proce dure used to assess referential communication skills can lead to an underesti mation of children's communicative competence.