| Children learning their first language and adults learning a second language by emersion are able to acquire grammar without studying the rules. Previous research in artificial grammar learning has shown that performance after unintentional (implicit) learning can exceed performance after intentional (explicit) learning. The present project aims at clarifying the efficacy of unintentional learning of artificial grammars. The main hypothesis is that unintentional learning is more efficient than intentional learning, when attention is directed at a task other than learning the rules, for which knowledge of the rules is, nevertheless, functional. To test this hypothesis a number of overt tasks will be designed, varying in degree to which knowledge of the grammar is useful in performing the overt task. Since it is assumed that communication is the main function of natural language, it is hypothesized that tasks requiring semantic deciphering of grammatical patterns will most strongly induce implicit learning of artificial grammars. A second issue in the literature is whether implicit learning is more efficient for either simple or complex rule systems. This will be investigated by varying the complexity of the artificial grammars used. The research questions will be answered by conducting laboratory experiments. |
