Meaning Representation Within and Across Languages

Open Access
Tokowicz, Natasha
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
August 09, 2000
Committee Members:
  • J Toby Mordkoff, Committee Member
  • James Lantolf, Committee Member
  • Judith Fran Kroll, Committee Chair
  • Karen Sue Quigley, Committee Member
  • context availability
  • lexical decision
  • ambiguity
  • translation
  • concreteness
  • bilingualism
Most models of bilingual memory assume that concepts are similar across the bilingual's two languages. However, there are cases in which conceptual overlap may not be perfect across translation equivalents. The results of Experiment 1 show that translation production generally is slower and more error-prone the more ambiguous words are. Furthermore, ambiguity was shown to interact with conceptual salience (the degree to which a word refers to a perceptible object or ‘concreteness’, and the ease with which one can think of a context for a word or ‘context availabilty’) and cognate status (lexical overlap of translation equivalents). The effects of multiple meanings were reduced for cognates and enhanced for words high in conceptual salience, whereas the effects of multiple translations were relatively constant across word types. Furthermore, ambiguity was shown to influence performance on a within-language task, lexical decision, such that ambiguous words were responded to more quickly than unambiguous words. Conceptual salience also influenced lexical decision latencies such that words high in conceptual salience were responded to more quickly than words low in conceptual salience. Thus, ambiguity and conceptual salience do not interact in the within-language lexical decision task, and the two factors have direct and independent influences on word recognition. The results of the third experiment showed that alternate conceptual candidates become activated during the translation process, similar to results for within-language picture naming (Peterson & Savoy, 1998). The results are discussed in relation to current models of monolingual and bilingual language representation and processing.