Conhecimento linguístico e Teoria da Mente: investigando a interface gramática-pragmática na aquisição do português brasileiro

Abstract

This thesis project focuses on the interface Language and Theory of Mind (ToM), seeking to investigate the cognitive demands involved in the reasoning of False Beliefs (FBs), conducted by children in the acquisition stage of Brazilian Portuguese and the development of the ToM. It is verified to what extent subclasses of mental state verbs are related, such as the factive verb saber (to know) and the epistemic verbs pensar/achar (to think), in the realization of the tasks of 2nd-order FBs by children in this age range. The concept of the ToM is understood as the sociocognitive capacity of an individual to recognize his/her own mental states and those of the others, and thus, to predict actions or behaviors (ASTINGTON & GOPNIK, 1988, 1991; WELLMAN, 1991, 2017). The theoretical perspective assumed is a minimalist conception of language (CHOMSKY, 1995 - present) and, more specifically, of language faculty according to Hauser et al. (2002). The Grammar Pragmatics interface is investigated based on a formal Pragmatics theory of cognitive nature, in line with the Relevance Theory (SPERBER & WILSON, 2001; 2002). The hypothesis that guides this study is that which previous comprehension and orientation questions, inserted in classic FB tests of 2nd-order, minimize the overlapping of linguistic cognitive demands and the ones not specifically linguistic involved in these tests. The starting point of the proposition of de Villiers (2004, 2005 e 2007), to whom the syntax of complementation is a pre-requisite for the development of the ToM, by proposing the existence of other factors directly related to the mastering of this ability. The aim of this study is to propose, in methodological terms, strategies that allow children to organize/retrieve information available in their memory, and distinguish points of view, cognitive abilities necessary to the reasoning of false beliefs. Therefore, two sets of experiments have been devised, divided into 4 stages each , in order to specifically investigate: (i) whether the syntactic structure of the target-question in the classic test of 2nd-order, with one or two epistemic verbs, interferes in the performance of the children who did the task; (ii) whether the kind of epistemic verb (pensar or achar) of the test-question, with only one structural nesting, contributes to the comprehension of FBs of 2nd-order; (iii) whether methodological changes in the application of the classic task, by inserting questions, previous to the test-question, help in the reasoning of FBs of 2nd-order; and (iv) to what extent the factivity correlates to the attribution of false beliefs. The results suggest that the adaptation of the task, along with the insertion of orientation questions, contributes to the conducting of the reasoning of 2nd-order FB, having a greater accuracy rate, especially in the group that performed the experiment with a previous orientation question with the factive verb saber. Besides, based on the better performance of the children in the task done with the verb pensar, when compared to the verb achar, the results indicate that the epistemic verb pensar can be an efficient linguistic clue to a child for understanding different points of view. This research intends to contribute to further the discussions about the relation between Language and the Theory of Mind, inasmuch as it presents a proposition of methodological change to the tasks usually performed as a means of investigating the development of the ToM.