Fatemeh Nemati

Given the prevalence of semantically ambiguous words in natural languages, correctly interpreting their meaning is an essential part of the successful comprehension of the desired message of an utterance or a sentence. Ambiguous word processing has attracted considerable attention in the previous studies, as many researchers investigating the influences of various characteristics of word representation such as orthography, semantics, and phonology on words processing. Theories of word processing agree that in addition to the target word, semantically related words are activated. However, relatively little is known about how co-activating semantic neighbors affect ambiguous word processing. This thesis sought to fill this gap by studying the influence of network connectivity and semantic set size as measures of semantic neighborhood density on the processing of English ambiguous words by Iranian EFL learners. Following the administration of the Oxford Placement Test (OPT) to measure the participants’ proficiency levels, the vocabulary size test (VST) was used to measure the English vocabulary knowledge of the participants. Then, a go or no-go semantic categorization task of English words was performed by 54 Iranian advanced learners in order to collect their response times and accuracy rates. Linear mixed-effects regression models and generalized linear mixed-effects regression models were employed for data analysis. Participants responded to unambiguous words more quickly than ambiguous words. However, there was also evidence of a small polysemy advantage in comparison with homonymy.