De rol van het werkwoord in het Nederlands tijdens on-line gesproken zinsverwerking en gesproken zinsproduktie
12 / 2001 - 01 / 2007
Studies to on-line processing of verbs are rare. Experiments that study on-line sentence processing have concentrated mainly on 'movement' and on English. Since there is no verb movement in this language, not many researchers were interested in the on-line processing of them. In Dutch, however, there is verb movement. The basic position of a Dutch verb is supposedly at the end of a clause (as can be seen in "De onderzoeker die in de regen naar huis fietst."). However, in matrix sentences, the verb appears second in the sentence ("De onderzoeker fietst naar huis."). It is assumed that the verb has moved from its basic position to the second position in the sentence. For both object- and wh-movement in English a priming effect has been shown of the meaning of the moved constituent at its 'basic' position. In other words, even when a constituent has moved, a bit of the meaning remains in place, probably to keep the original structure of the sentence in order. There are, however, other explanations for this priming effect. It has been claimed that the relation between the verb and its arguments are the reason for the reactivation patterns, rather than movement. This is one of the reasons to test Dutch sentence processing; languages with a basic word order SVO, such as English, cannot be used to tease these theories apart. Another reason to investigate the processing of verbs in Dutch is that young children and language-impaired speakers (both children and adults) have severe problems with the realisation of the proper verb form in the proper position in the sentence. In order to understand these language impairments better, it is crucial that the psycholinguistic mechanisms of 'normal' language systems are thoroughly researched. This project focuses on the position of the verb in Dutch: how do speakers process information on the position of the verb, what is the role of the verb in relation to its arguments, when does the brain realise that a verb is not in its proper position? By studying these issues, we aim to obtain insight in the psycholinguistic processes that are involved in processing and production of verbs in healthy and language-impaired speakers. The method we use to explore on-line sentence processing is cross-modal lexical priming. This method tabs directly into the psycholinguistic mechanisms used to process language, without using anything other than natural language or requiring metalinguistic awareness. Subjects listen to normal sentences, while strings of letters appear on a monitor. The subject is required to make lexical decisions (word/non-word) on these letter strings and their reaction times are measured. The 'words', either related or unrelated in meaning to the target verb, can be manipulated to the millisecond as to when to appear in the sentence. In this way, it is possible to look for priming effects practically wherever we want, but essentially at the site of the moved constituent.