摘要：Résumé Dans l'expérience 1, nous montrons que des enfants sont capables d'apprendre une Grammaire Artificielle (GA) dans des conditions et au sens usuels (cf. Reber, 1993). Cette capacité d'apprentissage ne se développe pas durant la période d'âge (9 à 11 ans) examinée. Pour l'expérience 2, nous réutilisons les GA et mots d'apprentissage de l'expérience 1. Mais nous construisons un nouvel ensemble de mots-tests destiné à mettre en échec le système de règles prêté aux enfants à l'issue de l'expérience 1. Le résultat est celui prédit: les enfants de 10 et 11 ans jugent grammaticaux les mots qui ne le sont pas, et inversement.
其他摘要：Summary: Artificial grammar learning by 9- to 11-year-old children. In experiment 1, we demonstrate that children can learn an artificial grammar in the standard conditions (see, for example, Reber, 1993). This implicit learning capacity does not develop in the examined age-period (9 to 11). Children's verbal explanations come under the three classical models of artificial grammar learning : abstraction of rules, comparison with training exemplars, and memorization of fragments. However, a complementary analysis shows that children's adherence to some grammatical rules plays a major role. In experiment 2, we use the same artificial grammar and learning items as in experiment 1. But, having discovered a system of imperfectly valid rules used by the children, we constructed a new set of test items. This new set is designed for checkmating the rule system of experiment 1 if the (new) children continue to use it. The results conform to the prediction : the 10- and 11-year-old children judged as grammatical the test items which were not, and conversely judged grammatical items as non-grammatical. Two other factors are taken into consideration in these experiments : the similarity and association (as measured by associative strength) between the learning items and the test items. The results suggest, if these factors are not confounded with grammaticality, that they have little impact on the judgment of grammaticality. Furthermore, in the present research they appear rather as a by-product of children's adherence to the rules. In the conclusion, we emphasize the questionable status — which is exacerbated by experiment 2 — of the notion of an « imperfectly valid» rule. We also underline that, in artijicial grammar learning, the grammar and the learning items do not alone determine the test performance.
