Vortrag: Optimizing input complexity for the acquisition of indefinite and definite articles and personal pronouns: An artificial language learning study on gender-like category induction
During language acquisition, German grammatical gender poses difficulty due to a lack of reliable noun cues from which gender can be inferred and because of syncretism within the paradigm. These factors affect the acquisition of indefinite articles, definite articles, and personal pronouns associated with nouns. In this talk, I will give a brief overview of an experiment on combining multiple gender cues to disambiguate the paradigm’s overlaps: By adding inflected adjectives, further morphemes were included. In an artificial language learning experiment, one participant group was presented sentences including semi-artificial color adjectives, while another group was exposed to the same sentences without artificial adjective suffixes. There were no significant differences between the two groups concerning sentence production with trained material and concerning grammaticality judgment tasks with trained and novel material. Nonetheless, participants of the inflected condition performed significantly higher compared to the other group at generalizing their knowledge to new material during a sentence production task with untrained items. The study concludes that the higher input complexity caused stronger paradigm knowledge.