The Role of Context on Processing of Turkish Subject and Object Relative Clauses

Abstract

Sentence processing is one of the major topics that is studied in psycholinguistics. Relative clauses due to their complex structures have offered a rich test bed to researchers studying in the psycholinguistics (Traxler, Morris and Seely, 2002; Chien-Jer and Bever, 2006). The findings obtained from psycholinguistic studies indicate that the processing of SRCs is much easier than the processing of the object relative clauses in some languages (i.e., Chinese: Lin and Bever, 2006; Dutch: Frazier, 1987; German: Holmes and O’Regan, 1981; Hebrew: Arnon, 2005). In other words, there is an asymmetry between the relative clause types in terms of processing. However, it is thought that when relative clauses are given in context, the relevant asymmetry may be eliminated (Yang, Mo ve Louwerse, 2012). This idea is based on Carl and Steedman’s Referential Support Theory (1985). This study is an attempt to illustrate whether discourse has an effect on the processing of Turkish relative clauses, and whether the processing asymmetry between Turkish subject and object relative clauses disappears when the context is added. In order to test the effects of the context, the study employed an eye-tracking method. Participants were chosen with convenience sampling. During the eye-tracking experiment, sixty-four (64) sentences were presented to the fifty-three (53) native Turkish speakers who are undergraduate. Of these sentences sixteen (16) were isolated sentences while the remaining sixteen (16) were given in context. Meanwhile, thirty-two (32) filler sentences including 16 isolated sentences and 16 sentences in context were added to the materials. During the experiment the participants read all the sentences, and where and how long they fixate were recorded. In addition, following each item, a comprehension question was asked to the participants. The resulting accuracy of the questions was thought to display the discourse effect on the processing of Turkish subject and object relative clauses. The data obtained were analysed by using Friedman Test. The findings of the study indicate that discourse had an effect on the processing of Turkish subject and object relative clauses in a negative way. To clarify, the results of the total fixation time, the regressive eye movements and the comprehension questions showed that the asymmetry between the isolated relative clauses had been eradicated because SRCs became more difficult to process within context. Thus, based on these findings, it can be stated that on the processing of Turkish relative clauses the presence of discourse does not have an effective role.