The neurophysiology of grammatical constraints : ERP studies on the influence of prosody and pragmatics on the processing of syntax and morphology in Swedish
Author
Summary, in English
This dissertation investigates the interaction of information from word order, morphology, lexical semantics, and prosody in the on-line processing of Swedish sentences using Event-Related Potentials.
Study I examines how the form, meaning, and syntactic position of objects influence their interpretation in ‘Object Shift’ contexts. In Swedish, object pronouns can either follow or precede sentence adverbs. Following a sentence adverb, pronouns are associated with sentence stress and may express inaccessible referents. Preceding the sentence adverb, they are clitics, and can only have accessible referents. Nouns cannot be clitics in Swedish and thus always follow sentence adverbs. In study I, sentence adverbs illicitly following nouns yielded a P600, showing difficulty in structural integration. After indefinite nouns, sentence adverbs also produced an early negativity, hinting at a pragmatic complication.
Studies II and III investigate the interaction of verb semantics, word order, and intonation in the interpretation of assertive force. Swedish assertive main clauses differ from subordinate clauses due to their post-verbal sentence adverbs and the presence of a left-edge boundary tone. After neutral verbs like ‘say,’ embedded main clauses are often used instead of subordinate clauses in order to express assertions. This is not possible with verbs like ‘hope,’ which question the truth of their complement. In the experiments, sentence adverbs in embedded clauses increased structural integration cost, as seen in a P600. The effect was greatest in embedded main clauses, where embedding verbs like ‘hope’ further increased the P600, while left-edge boundary tones reduced the effect.
Study IV examines Swedish word accents and morphology processing. Incorrect declension suffixes gave rise to N400 and P600 effects, showing influence on semantic processing as well as difficulty in integrating illicit suffixes into the word structure. Previous research has shown that Swedish word accents are conditioned by suffixes. In Study IV, Accent 2-inducing suffixes combined with stems associated with Accent 1 yielded a P600 effect, showing difficulty in processing the word structure.
Study I examines how the form, meaning, and syntactic position of objects influence their interpretation in ‘Object Shift’ contexts. In Swedish, object pronouns can either follow or precede sentence adverbs. Following a sentence adverb, pronouns are associated with sentence stress and may express inaccessible referents. Preceding the sentence adverb, they are clitics, and can only have accessible referents. Nouns cannot be clitics in Swedish and thus always follow sentence adverbs. In study I, sentence adverbs illicitly following nouns yielded a P600, showing difficulty in structural integration. After indefinite nouns, sentence adverbs also produced an early negativity, hinting at a pragmatic complication.
Studies II and III investigate the interaction of verb semantics, word order, and intonation in the interpretation of assertive force. Swedish assertive main clauses differ from subordinate clauses due to their post-verbal sentence adverbs and the presence of a left-edge boundary tone. After neutral verbs like ‘say,’ embedded main clauses are often used instead of subordinate clauses in order to express assertions. This is not possible with verbs like ‘hope,’ which question the truth of their complement. In the experiments, sentence adverbs in embedded clauses increased structural integration cost, as seen in a P600. The effect was greatest in embedded main clauses, where embedding verbs like ‘hope’ further increased the P600, while left-edge boundary tones reduced the effect.
Study IV examines Swedish word accents and morphology processing. Incorrect declension suffixes gave rise to N400 and P600 effects, showing influence on semantic processing as well as difficulty in integrating illicit suffixes into the word structure. Previous research has shown that Swedish word accents are conditioned by suffixes. In Study IV, Accent 2-inducing suffixes combined with stems associated with Accent 1 yielded a P600 effect, showing difficulty in processing the word structure.
Department/s
Publishing year
2009
Language
English
Document type
Dissertation
Publisher
Lund University
Topic
- Languages and Literature
Keywords
- linguistics
- event-related potentials
- ERP
- pragmatics
- language processing
- prosody
- syntax
Status
Published
Supervisor
ISBN/ISSN/Other
- ISBN: 978-91-628-7881-8
Defence date
16 October 2009
Defence time
13:15
Defence place
Pangea (hörsal 229), Geocentrum II, Sölvegatan 12, Lund
Opponent
- Kai Alter (professor)