Lexical Ambiguity

by

Julie E. Boland

• LAST MODIFIED: 22 August 2023
• DOI: 10.1093/obo/9780199772810-0309

Introduction

A lexical ambiguity, or homonym, is a string of sounds (in spoken language) or a string of characters (in written language) that corresponds to more than one word and/or meaning. Homophones sound the same, but may be spelled differently (e.g., I, eye, aye). Homographs are spelled the same, but may be pronounced differently (e.g., tear: as in tear in my eye or tear the paper). Many words are both homophonous and homographic (e.g., bank, bark). The word bat is both semantically ambiguous (it can refer to a flying animal or a piece of sports equipment) and syntactically ambiguous (it can be either a noun or a verb, as in bat her eyes). If the lexical alternatives differ in frequency, the most frequent one is called the dominant meaning and less frequent alternatives are called subordinate meanings. In order to understand a sentence containing an ambiguous word (e.g., The bat flew through the air), one must resolve both the meaning and syntactic form of the word, relying on some combination of context and frequency. In this example, the preceding determiner the rules out a verb meaning of bat, but two noun meanings are possible. In the absence of any richer context, a reader could choose the most frequent meaning or rely on the relative plausibility of the two types of bats flying through the air. Ambiguous words often appear in sentences that either disambiguate the word fully or strongly bias one particular meaning, but even in these cases, the word is considered lexically ambiguous. Researchers usually make a distinction between words with unrelated meanings, as in the examples above, and words with two or more related meanings (e.g., newspaper: as in work at the newspaper or read the newspaper.) The latter type are considered polysemous rather than lexically ambiguous. This distinction has proven useful, but is somewhat fraught because the distinction between related and unrelated meanings is ill-defined. Finally, most research on lexical ambiguity focuses on ambiguities within a language, but interlingual homographs can also be found, in which words with the same spelling, but different meanings, occur across languages. For example, coin occurs in both English and French, but means “corner” in French.

General Overviews

As Rodd 2018 notes, most words have some degree of homonymy or polysemy, so lexical ambiguity resolution is a pervasive component of language processing. Beekhuizen, et al. 2021 takes on the representational challenge of trying to distinguish between homonyms and polysemes within distributional semantics. Falkum and Vicente 2015 reviews the research on polysemy. Research on lexical ambiguity includes studies of ambiguity resolution in sentence context, activation of meanings for words in isolation, and the production of homophones. Lexical ambiguity resolution has been a productive research domain, in part because of what it can reveal about how contextual information is incorporated during lexical access. Consider this sentence: At dusk, a colony of bats flew out of the cave. It seems obvious that the word bats activates the concept of a flying mammal and not a wooden baseball club. In fact, Simpson 1984 outlines a long-standing debate about whether the more common baseball meaning is also activated briefly, and quickly rejected because it does not fit the context. The question arises because activation of the bat concept is triggered by a bottom-up stimulus, the string of letters B, A, T, or the corresponding phonemes, in spoken language. That stimulus is associated with multiple concepts, all of which could be initially activated and then higher-level context used to select the most appropriate meaning. This “multiple access” approach maintains that sentence context cannot prevent the autonomous activation of word candidates, and has been taken as support for the autonomy of lexical access, as discussed in Boland and Cutler 1996. In contrast, the view that that sentence context restricts the activation of word meanings is known as “selective access” and maintains that lexical access interactively combines bottom-up and top-down processing. Somewhat orthogonal to the role of context, is the role of meaning frequency. Simpson and Burgess 1985 demonstrates that the dominant meaning is accessed before the subordinate meaning for homographs in isolation. Within theories of lexical ambiguity resolution, “ordered access” and “re-ordered access” emphasize the role of meaning frequency. Useful references for quantifying the relative meaning frequency of English homographs are provided in Armstrong, et al. 2012. Research on lexical ambiguity resolution has led to heated debates and conflicting results. Tabossi and Sbisà 2001 tries to reconcile some apparent contradictions, focusing on methodological differences among studies. Vitello and Rodd 2015 provides a review of research designed to localize lexical ambiguity resolution in the brain.

• Armstrong, B. C., N. Tokowicz, and D. C. Plaut. 2012. eDom: Norming software and relative meaning frequencies for 544 English homonyms. Behavior Research Methods 44:1015–1027.

  DOI: 10.3758/s13428-012-0199-8

  A relatively recent, relatively large set of norms for English homographs. For a given homograph, participants estimated the frequency of use for each dictionary definition provided. Norms can be downloaded from a link at the end of the abstract.

• Beekhuizen, B., B. C. Armstrong, and S. Stevenson. 2021. Probing lexical ambiguity: Word vectors encode number and relatedness of senses. Cognitive Science 45:e12943.

  DOI: 10.1111/cogs.12943

  Asks whether distributional semantic vectors can distinguish among monosemes, polysemes, and homonyms.

• Boland, J. E., and A. Cutler. 1996. Interaction with autonomy: Multiple output models and the inadequacy of the Great Divide. Cognition 58:309–320.

  DOI: 10.1016/0010-0277(95)00684-2

  Provides historical context on competing theories of lexical and syntactic ambiguity resolution, and how those theories were framed within a larger debate about the modularity of, and within, the language processing system.

• Falkum, I. L., and A. Vicente. 2015. Polysemy: Current perspectives and approaches. Lingua 157:1–16.

  DOI: 10.1016/j.lingua.2015.02.002

  A review of both linguistic and psycholinguistic research on polysemy.

• Rodd, J. 2018. Lexical ambiguity. In Oxford handbook of psycholinguistics. 2d ed. Edited by M. G. Gaskell and S-A Rueschemeyer, 96–117. Oxford: Oxford Univ. Press.

  Provides a thorough introduction to the topic. Describes the ubiquity of ambiguous words, types of ambiguity, how ambiguous words are represented in the mental lexicon, how lexical ambiguities are learned in childhood, how lexical ambiguities are understood in sentences, and the brain mechanisms involved in lexical ambiguity resolution.

• Simpson, G. B. 1984. Lexical ambiguity and its role in models of word recognition. Psychological Bulletin 96.2: 316–340.

  DOI: 10.1037/0033-2909.96.2.316

  Describes three theories of lexical ambiguity resolution: context-dependent (or selective access), ordered-access (i.e., dominant meaning first and/or more strongly), and exhaustive access (or multiple access). Suggests that both meaning frequency and context influence the degree of activation of alternative meanings, and links research on lexical ambiguity resolution to word recognition more broadly.

• Simpson, G. B., and C. Burgess. 1985. Activation and selection processes in the recognition of ambiguous words. Journal of Experimental Psychology: Human Perception and Performance 11:28–39.

  The authors manipulated the SOA (stimulus-onset asynchrony) between an ambiguous word and a word that was unrelated or related to one of the meanings of the ambiguous word, to measure the time-course of activation for the dominant and a subordinate meaning. The task was lexical decision to the related/unrelated word.

• Tabossi, P., and S. Sbisà. 2001. Methodological issues in the study of lexical ambiguity resolution. In On the consequences of meaning selection: Perspectives on resolving lexical ambiguity. Edited by D. S. Gorfein, 11–26. Washington, DC: American Psychological Association.

  DOI: 10.1037/10459-001

  Grapples with the conflicting results that characterize the literature on lexical ambiguity resolution. Focuses on two experiment properties: the experimental paradigms and the types of contexts used.

• Vitello, S., and J. M. Rodd. 2015. Resolving semantic ambiguities in sentences: Cognitive processes and brain mechanisms. Language and Linguistics Compass 9.10: 391–405.

  DOI: 10.1111/lnc3.12160

  Focuses on where lexical ambiguity resolution occurs in the brain.