Weber, K. 1 , Christiansen, M. 2 , Petersson, K. M. 1, 3 , Indefrey, P. 4 & Hagoort, P. 1, 3
1 Radboud University Nijmegen, Donders Institute for Brain, Cognition and Behaviour
2 Cornell University, Department of Psychology
3 Max Planck Institute for Psycholinguistics
4 Heinrich-Heine University Düsseldorf, Department of Linguistics
How do we learn new syntactic structures and vocabulary without explicit instruction? We constructed a miniature language, “Alienese”, with a new vocabulary and grammar. Alienese was made up of three transitive (verb-object-subject, object-subject-verb, subject-verb-object) and one intransitive word-order (subject-verb). Verb-object-subject and object-subject-verb were novel word-orders to the participants.
In the beginning, participants simply read Alienese sentences. The meaning of the vocabulary and structures could be extrapolated from accompanying photographs depicting the sentences. One of the novel transitive word-orders was more frequent than the others: 60% vs 20 %.
The next day involved a syntactic priming paradigm. Participants first read a prime sentence, followed by a corresponding picture describing the relevant action. Next, a second sentence was presented with the same (primed) or different (unprimed) structure, followed by two pictures with reversed roles. Participants had to choose the picture that matched the sentence, which required that they had learned the three transitive word-orders.
After this session they could correctly translate 45% (range: 13 to 96 %) of the 46 Alienese verbs into their native language. Moreover, participants were significantly better at choosing the correct picture after syntactic structure repetition (primed: 89%, unprimed: 81%). Between the novel structures this effect was stronger for the infrequent word order (improvement after syntactic repetition: 14%) than for the frequent one (improvement: 4%).
Thus, the infrequent word-order was primed more strongly indicating that syntactic priming might be a mechanism for learning novel and especially infrequent sentence structures supporting the implicit learning theory of syntactic priming.
In the fMRI experiment, we will investigate the learning process within the first hour of reading Alienese sentences. The hypothesis is that already at this early stage syntactic repetition will lead to repetition effects in the brain, indicating a role of syntactic priming in language learning right from the start.