Conversations between mothers and children from three different cultural groups were analyzed to determine culturally preferred narrative elicitation patterns. The three groups included Japanese-speaking mother-child pairs living in Japan, Japanese-speaking, mother-child pairs living in the United States, and English-speaking Canadian mother-child pairs. Comparisons of mothers from these groups found that: (1) both Japanese-speaking groups provided less evaluation of their children's discourse than the English-speaking group; (2) both Japanese-speaking groups gave more verbal acknowledgement than did the English-speaking group; and (3) Japanese mothers in the United States requested more description from their children than Japanese mothers living in Japan. At 5 years of age, Japanese-speaking children, whether living in the United States or Japan, produced about 1.2 utterances per turn, whereas English-speaking children produced about 2.1 utterances per turn. Thus, whereas English-speaking mothers allow their children to take long monological turns, and even encourage this behavior, Japanese mothers simultaneously pay considerable attention to their children's narratives and facilitate frequent exchanges. Implications of these findings are further considered in the light of improving cross-cultural understanding. (MDM)