The relative abilities of word frequency, contextual diversity, and semantic distinctiveness to predict accuracy of spoken word recognition in noise were compared using two data sets. Word frequency is the number of times a word appears in a corpus of text. Contextual diversity is the number of different documents in which the word appears in that corpus. Semantic distinctiveness takes into account the number of different semantic contexts in which the word appears. Semantic distinctiveness and contextual diversity were both able to explain variance above and beyond that explained by word frequency, which by itself explained little unique variance.

Purpose-The mental lexicon of words used for spoken word recognition has been modeled as a complex network or graph. Do the characteristics of that graph reflect processes involved in its growth (M. S. Vitevitch, 2008) or simply the phonetic overlap between similar-sounding words?Method-Three pseudolexicons were generated by randomly selecting phonological segments from a fixed set. Each lexicon was then modeled as a graph, linking words differing by one segment. The properties of those graphs were compared with those of a graph based on real English words.Results-The properties of the graphs built from the pseudolexicons matched the properties of the graph based on English words. Each graph consisted of a single large island and a number of smaller islands and hermits. The degree distribution of each graph was better fit by an exponential than by a power function. Each graph showed short path lengths, large clustering coefficients, and positive assortative mixing. Conclusion-The results suggest that there is no need to appeal to processes of growth or language acquisition to explain the formal properties of the network structure of the mental lexicon. These properties emerged because the network was built based on the phonetic overlap of words.Keywords mental lexicon; spoken word recognition; graph theory; complex systems; lexical restructuring Understanding how children add words to their mental lexicon, how they represent the sound patterns of those words in their lexicon for purposes of recognizing spoken words, and how those representations possibly change over time are all critical to understanding the more general problem of language acquisition. The lexical restructuring hypothesis (e.g., Metsala 1997a(e.g., Metsala , 1997bWalley, 1993Walley, , 2005 has been particularly influential in this area. According to this hypothesis, a word's representation in the mental lexicon changes over time from a relatively holistic representation to a much more detailed and segmental representation. Holistic representations suffice when the child knows only a few words. As vocabulary grows, however, and more words are added to the lexicon that are phonologically similar to words already part of the lexicon, their lexical representations evolve to ones with the more detailed, segmental information needed to discriminate similar-sounding words. Lexical restructuring does not occur all at once for the entire lexicon. Rather, it occurs for a particular word as that word acquires phonological neighbors and becomes part of a larger set of similar-sounding words. NIH-PA Author ManuscriptNIH-PA Author Manuscript NIH-PA Author ManuscriptVarious behavioral data have been adduced in favor of the lexical restructuring hypothesis. Metsala (1997a), for instance, compared the performance of children from three different age groups and adults using a word gating paradigm (Grosjean, 1980). She used decreases in the gate duration required to recognize a word as an indicator of segmental representation for that word. High-frequ...

No abstract

We compared the ability of three different contextual models of lexical semantic memory (BEAGLE, Latent Semantic Analysis, and the Topic model) and of a simple associative model (POC) to predict the properties of semantic networks derived from word association norms. None of the semantic models were able to accurately predict all of the network properties. All three contextual models over-predicted clustering in the norms, whereas the associative model under-predicted clustering. Only a hybrid model that assumed that some of the responses were based on a contextual model and others on an associative network (POC) successfully predicted all of the network properties and predicted a word's top five associates as well as or better than the better of the two constituent models. The results suggest that participants switch between a contextual representation and an associative network when generating free associations. We discuss the role that each of these representations may play in lexical semantic memory. Concordant with recent multicomponent theories of semantic memory, the associative network may encode coordinate relations between concepts (e.g., the relation between pea and bean, or between sparrow and robin), and contextual representations may be used to process information about more abstract concepts.

Two experiments measured the effects of coordinate falses, which pair two exemplars from a common category (e.g., "A dog is a cat"), on latencies to true items in a category verification task. Although, relative to anomalous falses, coordinate falses did increase latencies to trues, the size of the increase did not vary with the similarity of the coordinates. Furthermore, latencies were faster to coordinate falses that were more similar than to coordinate falses that were less similar. These results suggest relational similarity is as powerful a predictor of performance in category verification as is attribute similarity. Finally, there was some evidence that the increase in latencies was the same for typical and atypical category exemplars. This last finding, if substantiated, indicates that relational information is generally accessed prior to attribute information.