The present study examined a universality hypothesis on language acquisition, and three hypotheses on the acquisition order of the verification system and answering system. The universality hypothesis states that children all over the world acquire their languages in the same way. The three hypotheses were as follows. The equivalence hypothesis states that the two systems are basically equivalent and therefore it predicts the simultaneous acquisition of the two systems. The verification primary hypothesis predicts that the verification system is acquired earlier than the answering system. The question simpler hypothesis predicts that the answering system is acquired earlier than the verification system. Sixteen Japanese-speaking 4 year olds and 15 5 year olds were given one verification task and one question answering task. In the verification task, children were individually asked to verify 40 statements consisting of: true affirmatives(e.g.You are a child.), false affirmatives(e.g.You are a baby.), true negatives(e.g.You aren't a baby.), and false negatives(e.g.You aren't a child). In the answering task, children were asked to answer 40 corresponding questions(e.g.Are you a child?, Are you a baby?, etc.). The order of the two tasks was counterbalanced, and the order within a task was randomized. Children's gestural and verbal responses were scored correct if they gave an expected response(e.g.head-nodding or verbal response[sou(right)]to You are a child). All other responses were scored incorrect. All children performed better on the verification task than on the answering task, thus supporting the verification primary hypothesis. They also performed better on affirmatives than on negatives. The difference between the affirmatives and negatives was much greater among the 4 year olds than among the 5 year olds. The 4 year olds found it more difficult to verify false negative statements than true negative statements, whereas the 5 year olds performed equally well on both types of negative statements. These results are shown in TABLE 2, and are contrasted with those(TABLE 3)of English-speaking children. The 4 year olds' data neither fitted the previous verification models(e.g.Carpenter and Just, 1975)nor supported the universality hypothesis on language acquisition. Therefore, a psycholinguistic model to accoount for the 4 year olds' data was proposed. The model in TABLE 4 states that children develop negative knowledge representation upon hearing false affirmative statements and true negative statements, and then compare the statement representation and the knowledge representation. Since the number of comparison is fewer for true negatives than for false negatives, the model could accunt for the better performance on true negatives than on false negatives. As for affirmatives, the new model predicts a similar pattern to the previous models. This proposed model and previous models are discussed in terms of the different linguistic characteristics between the English and the Japanese language.