A P600 component in an artificial grammar without semantics: an

TABULLO, ANGEL; SEVILLA, YAMILA; PASQUALETTI, GUILLERMO; YORIO, ALBERTO; ZANUTTO, SILVANO; WAINSELBOIM, ALEJANDRO

Kobe

Congreso; 29th International Congress of Clinical Neurophysiology; 2010

International Federation of Clinical Neurophysiology

The P600 component is a late positivity found after syntactic anomalies in sentences (Osterhout and Holcomb 1992) So far, it has not been addressed if the P600 is found after violations of an artificial grammar without semantic content, based solely on statistic information. The aims of the present study were: (1) To examine the presence of P600 after structural violations of a semantic-free artificial grammar based on statistical information; (2) to examine if the type of structural violation affects the P600 component. Methods: 21 right-handed Spanish native speakers participated in the study. A semantic-free artificial grammar was used. The lexicon consisted of 17 bisillabic, phonotactically correct pseudowords. The grammar admitted two possible sentence structures. Subjects were trained by mere exposure with a set of 90 correct sentences, and then discriminated between grammatical and ungrammatical new sentences in the test stage. Two types of ungrammatical sentences were presented: (a) conjunction violation (presenting a connector from structure 1 in a sentence with structure 2), (b) category violation (presenting an item from an incorrect category at the end of the sentence). EEG activity was recorded during test stage. 21 right-handed Spanish native speakers participated in the study. A semantic-free artificial grammar was used. The lexicon consisted of 17 bisillabic, phonotactically correct pseudowords. The grammar admitted two possible sentence structures. Subjects were trained by mere exposure with a set of 90 correct sentences, and then discriminated between grammatical and ungrammatical new sentences in the test stage. Two types of ungrammatical sentences were presented: (a) conjunction violation (presenting a connector from structure 1 in a sentence with structure 2), (b) category violation (presenting an item from an incorrect category at the end of the sentence). EEG activity was recorded during test stage. Results: A late centro-parietal positivity was observed within 600 900 ms after grammar violations. The difference between grammatical and ungrammatical sentences was significant (F(1,20) = 7.015; p = 0.016). A further ANOVA, discriminating between violation types, showed a region × laterality × sentence type effect (F(16,304) = 2.647; p = 0.039). Category violations elicited greater positivities than grammatical sentences. No differences were found between conjunction violations. A late centro-parietal positivity was observed within 600 900 ms after grammar violations. The difference between grammatical and ungrammatical sentences was significant (F(1,20) = 7.015; p = 0.016). A further ANOVA, discriminating between violation types, showed a region × laterality × sentence type effect (F(16,304) = 2.647; p = 0.039). Category violations elicited greater positivities than grammatical sentences. No differences were found between conjunction violations. × laterality × sentence type effect (F(16,304) = 2.647; p = 0.039). Category violations elicited greater positivities than grammatical sentences. No differences were found between conjunction violations. Discussion: Results suggests that the P600 can be elicited by expectancy violations based on distributional properties of input. The effect was more clearly observed after category violations. It is possible that this type of violation was more salient, as it was located at the end of the sentence. Results suggests that the P600 can be elicited by expectancy violations based on distributional properties of input. The effect was more clearly observed after category violations. It is possible that this type of violation was more salient, as it was located at the end of the sentence.