Syntactic processing differences and the effects of memory-load interference for object relative and subject relative clauses in Turkish

The purpose of the study, which focuses on the relative clauses in Turkish, is twofold: In Experiment 1, it was aimed to fill the gap in Turkish literature with regard to syntactic complexity difference between subject relative clauses and object relative clauses. As a self-paced reading experiment, participants read either subject relative clause of object relative clause to measure the processing cost as a slowing down in reading. The result of Experiment 1 revealed that object relative clauses are harder to process than subject relative clauses suggesting that there is a subject preference in Turkish. Secondly, as a separate but related follow-up study, it was tested whether such complexity interacts with memory load in any way. For this purpose, participants memorized a set of words while reading syntactically complex sentences and syntactically simpler sentences (object and subject relative clauses, respectively). The study also manipulated whether the memory load items and the words in the sentence were similar or not (matched or unmatched conditions). Performance in recall was significantly higher for the unmatched condition than the matched condition. Recall was also higher in simpler sentences than complex sentences. Results of Experiment 2 indicated that syntactic processing is not modular meaning that both linguistic and non-linguistic verbal processings share the same verbal working memory resource. Moreover, online and offline measurements of both studies are compared to show why offline studies are different in nature and why online studies are more reliable as long as sentence processing is concerned.