Karoline Lohse, Andrea Hildebrandt, Frauke Hildebrandt

• Adult-child interactions can support children's development and are established as predictors of program quality in early childhood settings. However, the linguistic components that constitute positive interactions have not yet been studied in detail. This study investigates the effects of hypotheses proposed by adults on children's responses in a dyadic picture-book viewing situation. In 2 experiments, adults' use of hypotheses (e.g., "Maybe this is a dwarf's door") was tested against the use of instructive statements ("This is a dwarf's door") and in combination with open questions ("What do you think, why is the door so small?"). In Experiment 1, hypotheses differed from instructions only by the modal marker "maybe". Children's responses to hypotheses were longer and contained more self-generated explanations as compared to responses to instructions. The use of hypotheses also seemed to encourage children to attach more importance to their own explanations. In Experiment 2, combining hypotheses with open-ended why questionsAdult-child interactions can support children's development and are established as predictors of program quality in early childhood settings. However, the linguistic components that constitute positive interactions have not yet been studied in detail. This study investigates the effects of hypotheses proposed by adults on children's responses in a dyadic picture-book viewing situation. In 2 experiments, adults' use of hypotheses (e.g., "Maybe this is a dwarf's door") was tested against the use of instructive statements ("This is a dwarf's door") and in combination with open questions ("What do you think, why is the door so small?"). In Experiment 1, hypotheses differed from instructions only by the modal marker "maybe". Children's responses to hypotheses were longer and contained more self-generated explanations as compared to responses to instructions. The use of hypotheses also seemed to encourage children to attach more importance to their own explanations. In Experiment 2, combining hypotheses with open-ended why questions elicited longer responses but no more self-generated explanations in children than openended questions alone. Results indicate that subtle differences in adults' utterances can directly influence children's reasoning and children's contributions to dialogues.…