Dissertation abstract:

This dissertation is concerned with the role of prosody and, specifically, linguistic rhythm for the syntactic processing of written text. My aim is to put forward, provide evidence for, and defend the following claims:

• While processing written sentences, readers make use of their phonological knowledge and generate a mental prosodic-phonological representation of the printed text.
• The mental prosodic representation is constructed in accordance with a syntactic description of the written string. Constraints at the interface of syntax and phonology provide for the compatibility of the syntactic analysis and the (mental) prosodic rendition of the sentence.
• The implicit prosodic structure readers impose on the written string entails phonological phrasing and accentuation, but also lower level prosodic features such as linguistic rhythm which emerges from the pattern of stressed and unstressed syllables.
• Phonological well-formedness conditions accompany and influence the process of syntactic parsing in reading from the very beginning, i.e. already at the level of recognizing lexical categories. At points of underspecified syntactic structure, syntactic parsing decisions may be made on the basis of phonological constraints alone.
• In reading, the implicit local lexical-prosodic information may be more readily available to the processing mechanism than higher-level discourse structural representations and consequently may have more immediate influence on sentence processing.
• The process of sentence comprehension in reading is conditioned by factors that are geared towards sentence production.
• The interplay of syntactic and phonological processes in reading can be explained with recourse to a performance-compatible competence grammar.

The evidence from three reading experiments supports these points and suggests a model of grammatical competence in which constraints from various domains (syntax, semantics, pragmatics, discourse structure, and phonology) interact in providing the possible structural, i.e. grammatical descriptions.

Following the introduction (chapter 1), chapter 2 presents an experimental study on the effects of linguistic rhythm on processing non-canonical structures in oral reading, examining the interplay of syntax, focus structure, prosody, and working memory. The experiment attests that a written word sequence that forces readers to deviate from the optimal rhythmic alternation of stressed and unstressed syllables may critically hamper the processing of complex, non-canonical sentences. At points of high memory demand, the cognitive load associated with a stress clash configuration may lead to the (temporary) overwriting of memory traces that are necessary for the processing of syntactic long distance dependencies. I argue that this overwriting becomes possible because the clash information, which is implicitly coded in the written string, is processed immediately while access to stored information in working memory may be tardy.
This chapter is published as Kentner, G. (2015). Stress clash hampers processing of noncanonical structures in reading. In: Vogel, R. and van de Vijver, R. (eds.): Rhythm in cognition and grammar. Berlin: de Gruyter.)

Chapter 3 affords more data on effects of linguistic rhythm on written sentence comprehension. Two reading experiments examine the influence of stress-based linguistic rhythm on the resolution of local lexical-syntactic ambiguities. Both speech production data from unprepared oral reading as well as eye-tracking results from silent reading demonstrate that readers favor syntactic analyses that allow for a prosodic representation in which stressed and unstressed syllables alternate rhythmically. The findings contribute evidence confirming immediate and guiding effects of linguistic rhythm on the earliest stages of syntactic parsing in reading.
(This chapter is published as Kentner, G. (2012). Linguistic rhythm guides parsing decisions in written sentence comprehension. Cognition 123 (1), pp. 1-20.)

In chapter 4, the performance data from the experiment reported in chapter 3 are modeled as an incremental constraint satisfaction process in the framework of an Optimality Theoretic parsing account. Solely making use of constraints derived from competence grammar, the model is capable of capturing the data and advocates the simultaneous application of syntactic, prosodic and syntax-phonology interface constraints in incremental processing. The model predicts that, in the case of syntactic indetermination, weak prosodic constraints may decide about syntactic ambiguity resolution. The performance-compatible OT grammar integrates the processes of syntactic parsing and prosodification in reading, hence dissolving the strict separation of language production and comprehension. At the same time the OT model endorses a bidirectional relationship between syntax and phonology in grammar.
(This chapter will appear as Kentner, G. (2017). Rhythmic parsing. The Linguistic Review 34. [Online ahead of print])

This page has been accessed at least times since 25th October 2012.
last updated 22 November 2016