Martin, Randi C2019-05-172019-05-172019-052019-04-18May 2019Chapman, Curtiss Anthony. "How do word frequency and semantic diversity affect selection of representations in word processing?." (2019) Diss., Rice University. <a href="https://hdl.handle.net/1911/105963">https://hdl.handle.net/1911/105963</a>.https://hdl.handle.net/1911/105963The frequency at which words are encountered has long been considered an important factor in language processing, with higher frequency speeding word comprehension and production. That is, many theories claim that word representations are stronger for words encountered more often. However, some recent work has claimed that, rather than word frequency, the factor determining the strength of the representation of words (and accounting for previous effects of word frequency) is the word’s semantic diversity (SemD) – i.e., the variety of contexts in which a word is found (Adelman et al., 2006; Jones et al., 2012). Other recent approaches have suggested that SemD may play a more nuanced role in language processing—helping in certain situations and harming in others. That is, when a specific word meaning must be accessed without sufficient context, greater SemD may lead to more effortful processing, sometimes even minimizing or reversing typical frequency effects (Hoffman et al., 2011, 2013). More effortful processing would occur because one has to distinguish a single meaning from among all of the contextually-associated meanings to which a word may be potentially related. A mechanism that distinguishes a target meaning from among potential meanings would be an important one in language processing generally, but the presence of such a mechanism and its relationship to SemD have not been deeply explored. This dissertation explores the relationship and influence of word frequency and SemD in language processing, including whether SemD explains typical word frequency effects and whether SemD creates a need to distinguish from among competing contextual meanings. Part one explores effects and interactions of word frequency and SemD across a variety of large mega-study databases of language processing tasks. Part two attempts to manipulate the mechanism that resolves competition between word meanings to examine its impact on effects of SemD and language processing more generally. Results indicate that both frequency and SemD are important qualities affecting word processing, but little evidence is found for the role of a meaning selection mechanism that responds to SemD. The impact of these results on current theories of language processing is discussed.application/pdfengCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.Semantic DiversityFrequency effectscognitive controllexical selectionsemantic controlThesis2019-05-17