Processamento da correferência anafórica de pronomes e nomes repetidos em brasileiros aprendizes de francês como L2

Abstract

This work aims to investigate the coreferential processing of learners of a L2 (french) in three proficiency levels (beginner, intermediate, advanced) and compare them with native french speakers. It will be analyzed the extent to which the proficiency influences the processing of the L2 in the specific phenomenon of Repeated Name Penalty (Gordon & Chan 1995, Gordon & Hendrick, 1998; Chambers & Smyth, 1998) and Informational Load Hypothesis (Almor, 1999), besides relating the processing of the L2 Shallow Structure Hypothesis (Felser, 2006) and the Declarative/Procedural Model (Ullman, 2001). Two self-observation reading experiments were applied. In the first experiment, we analyzed the reading times for pronouns and repeated names in retrieved anaphoric in the subject position with learners from the three proficiency levels. The results show that each level of proficiency presents a pattern processing. The second experiment was applied to native speakers of French language and reading times compared with the performance of learners of an advanced level in the first experiment. There was a difference in processing between the two groups, although both suffer repeated name penalty, concerning the processing time; Learners process slower than natives, confirming the Shallow Structure Hypothesis and the Declarative/Procedural Model.