| Popis: |
The ultimate goal of language comprehension is to obtain meaning. However, this is difficult because many words are semantically ambiguous, mapping onto multiple meanings. Semantic ambiguity resolution has proven a useful tool to investigate language processing in general. However, the majority of research has focused on the initial encounter of an ambiguous word. Less work has investigated the processes occurring after an ambiguous word is encountered, when the initially understood meaning needs to be reinterpreted in light of subsequent inconsistent information. The first part of this thesis investigated the cognitive processes underlying semantic reinterpretation, examining how successful listeners are at this process as well as assessing the time course of suppressing and integrating the contextually inappropriate and appropriate meanings respectively. A semantic relatedness task was employed in which participants listened to ambiguous and unambiguous sentences and decided whether a following visual probe word was related or unrelated to the sentence. The results revealed that listeners are highly effective at reinterpretation but that suppression of the inappropriate meaning is delayed relative to integration of the appropriate meaning. The rest of the thesis examined the neural responses to these sentences by using functional magnetic resonance imaging (fMRI) and Transcranial magnetic stimulation (TMS). The fMRI study demonstrated ambiguity-elevated responses in left inferior frontal gyrus (LIFG) and posterior temporal cortex. These responses were modulated by the frequency of the ambiguous word’s meanings, such that activation was greater for sentences with a higher likelihood of reinterpretation. The final study developed a TMS paradigm to examine whether LIFG is necessary for this process, demonstrating evidence that this region may be important for sentence processing more generally. Together, this thesis has advanced understanding into the cognitive and neural processes underlying semantic reinterpretation that have various implications for models of ambiguity resolution and language comprehension in general. |
