International School for Advanced Studies (S.I.S.S.A.),
Extracting regularities from the environment is crucial in the course of development. In the process of language acquisition both the lexicon and the underlying grammatical rules have to be learned. In a series of experiments we use eye-tracking to study rule learning in monolingual and bilingual infants. In Study 1 we found that 7- and 12 month-old infants could extract and generalize simple patterns that had a well-defined structure (words with repeated syllables, e.g. “kiki”) but fail to extract patterns that had no clear structure (different syllables, e.g. “kila”).
Study 2A asked whether 7- and 12-month-olds are able to extract two rules simultaneously when confronted with more complex grammars that both had an inherent structure (words with adjacent repetitions, e.g “zozomo” or distant repetitions e.g. “zomozo”). Both age groups generalized the rule in the case of words with adjacent repetitions, but not with distant repetitions. This suggests that when faced with two rules that differ in complexity infants extract the more salient one (adjacent repetitions). In Study 2B we investigate whether being exposed to two languages from birth would lead to an advantage in learning two ‘artificial’ rules. Bilingual children have to learn two different syntactic systems belonging to the two languages with halved exposure times. The results suggest that bilingual 12-month-olds succeed to extract both adjacent and distant repetitions.
Study 3 investigates the limitations of rule learning by asking how flexible are 7-month-old monolinguals and bilinguals in learning a second rule after already being exposed to a different rule system. While both monolinguals and bilinguals showed fast learning over the trials of the first phase, only bilinguals learned in the second phase. Seemingly, 7-month-old bilinguals are more efficient in inhibiting a previously learnt rule, and in switching attention to a new one as compared to monolinguals.