Neuroscience Letters 443 (2008) 77–81 Contents lists available at ScienceDirect Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet The impact of semantic-free second-language training on ERPs during case processing Jutta L. Mueller ∗ , Stefan Girgsdies, Angela D. Friederici Max-Planck Institute for Human Cognitive and Brain Sciences, Stephanstr. 1a, 04103 Leipzig, Germany article info Article history: Received 18 March 2008 Received in revised form 18 June 2008 Accepted 18 July 2008 Keywords: Language learning ERP Second language Japanese N400 P600 abstract Some language-related ERP responses are only observed in high proficiency L2 speakers. It is unknown, however, how these ERP patterns are influenced by language training. We tested the effect of semantic- free training on ERPs related to syntactic processing in auditory sentence comprehension in German participants learning a miniature version of Japanese. When presented with correct sentences and sen- tences containing a case violation, the learners showed an N400-like negativity and a P600-like positivity resembling the ERP pattern reported for Japanese natives. They contrasted with a previously tested group of learners who, though they had been provided with full language training including semantic infor- mation, had only shown the P600. The results suggest that the absence of semantic input facilitates the development of some aspects of native-like language processing operations in L2 learners. © 2008 Elsevier Ireland Ltd. All rights reserved. Brain imaging and electrophysiological studies show that syntactic processing in non-native languages is often reflected by non-native patterns of brain activity [25,10,24]. Only when very high levels of proficiency are achieved, a convergence of native and non-native brain responses is observed [21]. One approach that can be used to investigate language learning at a stage of high proficiency is the use of artificial or semi-artificial miniature languages which are char- acterized by a restricted set of items and rules [7,18]. Within such a language, participants can achieve a high skill-level in a short period of time within a controlled experimental setting. In contrast to studies with real second language learners, such paradigms make it possible to control the amount and type of input, although at the expense of external validity. In the present study, we exploited this methodology to investigate how type of foreign language train- ing impacts on the development of neurocognitive mechanisms for language processing. The degree of proficiency achieved in a second language learned after puberty varies considerably and is certainly lower than that of the proficiency achieved during first language acquisition [1]. This suggests that there may be more or less efficient strategies to deal with second language input which influence the syntac- tic processing mechanisms available to learners. The current study investigates whether the provision of semantic information during language training helps or hinders the development of native-like ∗ Corresponding author. E-mail address: muellerj@cbs.mpg.de (J.L. Mueller). event-related potentials (ERPs) during syntactic processing. For this study, we focused on the acquisition of the syntactic feature of case marking during auditory sentence comprehension in Japanese. The language sensitive ERPs that are relevant in the present con- text are the N400 and the P600. The N400 is a negative-going ERP peaking at about 400 ms that is known to increase in amplitude in response to a semantically unexpected word [12]. While it is mostly seen as an indicator for lexical processing during semantic integra- tion [11] there is some evidence to suggest that it can also reflect earlier stages of lexical processing, such as lexical access. This has become evident in lexical priming studies [6,4,13] and in learning studies using non-words which precluded the processing of any semantic meaning [22,3]. Recent studies suggest that, in addition to lexical-semantic processes, the N400 can also reflect the pro- cessing of thematic hierarchies in which the relationships holding between sentential arguments that carry thematic roles, such as ‘actor’ or ‘undergoer’ [8,9] are established. The P600, a positive deflection at around 600 ms, is often seen as indicating syntactic processing, such as dealing with syntactic complexity, reanalysis or repair [7]. Although both ERP effects, the N400 and the P600 vary to a certain degree with respect to topography, amplitude and latency depending on modality and task, the consistency in their eliciting conditions suggests they are reliable indicators for syntactic and semantic processing in language tasks. In a previous study, it was shown that processing of case mark- ers in a miniature version of Japanese, termed ‘Mini-Nihongo’, yielded different ERP patterns in Japanese native speakers and in trained non-native listeners [15]. The trained non-native listeners 0304-3940/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2008.07.054