| Abstract: | The present study simultaneously assessed the relative contributions of feedback indicative of comprehension and the apparent age of the listener, either an adult or a doll which resembled a toddler, in a 2 (listeners) × 2 (types of feedback, C = comprehension, NC = noncomprehension) design. Two groups of children, a 3-year-old (N = 13, 7 boys, 6 girls) and a 5-year-old group (N = 12, 6 boys, 6 girls) were asked to tell stories to both the adult and doll in both C and NC conditions. The doll was constructed with an internal speaker such that it could actually carry on a conversation with the children. The conversations were taped, transcribed, and scored for mean length of utterance (MLU), transitional utterance length to each C and NC signal, and the proportion of child questions, exact self-repetitions, repetition and reductions, and rephrases/elaborations. The data analysis revealed that all children appropriately modified the length of their utterances (MLU) in the doll condition but not in the adult condition, indicating that they were sensitive to both the feedback and the nature of their listeners. Older children were more likely than younger children, and girls more likely than boys to adjust the length of their utterances appropriately to each type of feedback, slightly increasing the length of the subsequent utterance to a C signal and decreasing the length to an NC signal. The younger children were also more likely to respond with a simple repetition to NC cues from the adult. |
