Date of Award


Document Type

Open Access Dissertation


Communication Sciences and Disorders

First Advisor

Dirk-Bart den Ouden


Verb argument structure (VAS) is pivotal to sentence production and comprehension, since it determines participant roles, as well as their grammatical form and syntactic position in a sentence. Neural correlates of VAS processing have mainly been studied in terms of the number of arguments. Data on the neural and behavioral effects of other VAS characteristics are limited, whereas they would have implications for behavioral and brain stimulation treatments of language disorders.

The present research investigated behavioral and neural effects of three understudied VAS characteristics (number of subcategorization options, number of thematic options and number of number-of-argument options) in single-word-level and sentence-level processing. The results indicate that their effects are highly dependent on processing conditions. A greater complexity in terms of the number of subcategorization and thematic options facilitated single-word processing, possibly due to making verb representations “stronger” and providing them with a greater number of connections in the mental lexicon, but had a detrimental effect in sentence processing, where VAS information needs to be processed to a fuller extent. VAS processing was associated with activation in bilateral (although mainly left-lateralized) frontal, temporal and parietal brain areas, including consistent activation in the left middle temporal gyrus. The third characteristic, the number of number-of-argument options, did not appear to have a robust neural or behavioral effect.

The present research suggests that VAS effects may have a semantic nature, rather than originate from a dedicated VAS module in verb representations, because they were only found for two VAS characteristics that have semantic correlates and because no evidence of automated exhaustive access to purely grammatical VAS information was found in shallower (single-word) processing conditions. This provides a novel account for VAS effects. Still, regardless of the nature of VAS effects, the present research suggests that the number of subcategorization and thematic options of verbs should be taken into account in selection of stimuli for complexity-based behavioral treatments of aphasia. Another clinical implication of this research is that it suggests potential target sites (mainly, left middle temporal gyrus) for brain stimulation treatments of verb and sentence processing.