Mass counts: ERP correlates of non-adjacent dependency learning under different exposure conditions

Citron, Francesca M M, Oberecker, Regine, Friederici, Angela D and Mueller, Jutta L (2011) Mass counts: ERP correlates of non-adjacent dependency learning under different exposure conditions. Neuroscience Letters, 487 (3). pp. 282-286. ISSN 0304-3940

Full text not available from this repository.


Miniature language learning can serve to model real language learning as high proficiency can be reached after very little exposure. In a previous study by Mueller et al. [18] German participants acquired non-adjacent syntactic dependencies by mere exposure to correct Italian sentences, but their ERP pattern differed from the one shown by native speakers. The present study follows up on that experiment using a similar design and material and is focused on two important issues: the influence of acoustic cues in the material and the impact of the learning procedure. With respect to the latter we compared alternating learning and test phases to a continuous learning and test phase. In addition, a splicing procedure eliminated prosodic cues in order to ensure that non-adjacent dependencies were learned instead of adjacent ones. Results for the continuous phase design showed a native-like biphasic ERP pattern, an N400 followed by a left-focused positivity. In the alternating design behavioural accuracy was lower and only an N400 was found. The results suggest an advantage of continuous learning phases for adult learners, possibly due to the absence of ungrammatical items present in the test phases in the alternating learning procedure. Furthermore, the replication of the earlier study with prosodically controlled material adds evidence to the general finding that syntactic non-adjacent dependencies can be learned from mere exposure to correct examples

Item Type: Article
Schools and Departments: School of Psychology > Psychology
Depositing User: Francesca Citron
Date Deposited: 21 Feb 2013 15:34
Last Modified: 21 Feb 2013 15:34
📧 Request an update