An erp study of the processing of subject and objectrelative clauses in Turkish

The first aim of this study is to determine which type of relative clauses – subject relative clauses or object relative clauses – are easier to process. The second aim, assuming that there will be a difference between subject relative clauses (SRC) and object relative clauses (ORC) in terms of the processing difficulty, is to reveal on which words or word groups in the sentence the difficulty presents itself. In an Event-Related Brain Potentials (ERP) experiment, participants were given a total of 240 sentences, where 120 had SRCs in them (60 of which were in the grammatical function-subject, "[The senator who applauded the PM] scolded the minister"), and the other 60 were in the grammatical function-object,"The minister scolded [the senator who applauded the PM"]); the remaining 120 sentences had ORCs (again, 60 of which were in the grammatical function-subject, "[The senator that the PM applauded] scolded the minister," and the other 60 were in the grammatical function-object, "The minister scolded [the senator that the PM applauded])." The following factors were used in the statistical analysis: Grammatical Function (GF) (subject, object) x Clause Type (SRC, ORC) x A(nterior)-P(osterior) distribution (frontal, parietal) x LAT(eralization) (left, right). Behavioral findings revealed that participants perform better when SRCs are in the subject function and ORCs are in the object function. In ERP findings, the results of the experiment revealed that subject relative clauses are easier to process than object relative clauses in the subject function. Therefore, object relative clauses revealed Left Anterior Negativity. The second result is the cognitive difficulty observed on the head noun of the relative clauses. In this study, relative clauses were investigated for the first time by using the ERP technique. Key words: Language Processing, Event-Related Brain Potentials, Relative Clauses, LAN