Models Of Lexical Ambiguity Resolution Research Articles

Лексическая неоднозначность в экспериментальных исследованиях: метаанализ

The article analyses experimental studies of lexical ambiguity (polysemy and homonymy). In total, 42 papers published between 1981 and 2017 were selected for analysis. Selected works were analyzed from the point of view of a) interaction of the main factors of lexical ambiguity resolution (meaning dominance and context), b) modularity/interactivity of language systems. The choice of these issues was caused by the fact that, despite a significant number of papers addressing them, a common theoretical model of lexical ambiguity resolution is still missing. The analysis concluded that none of the existing models could be chosen as the only true one. I believe that it is more promising to combine all models into one; in this case, we should not talk about “models” but about “algorithms” or “strategies”. This merge is compatible with the results of studies of hemispheric asymmetry, which showed that the right and left hemispheres differently process ambiguous words. This difference is projected on the difference between the two main points of view on language processing, interactive and modular. Also, as a result of the analysis, the properties of meaning dominance and context as the main factors influencing the perception of ambiguity are summarized, and the ways and possibilities of their interaction are determined. These factors can participate in the process of resolving ambiguity autonomously or together, with different functions, speed, and levels of perception.

Tracking competition and cognitive control during language comprehension with multi-voxel pattern analysis

To successfully comprehend a sentence that contains a homonym, readers must select the ambiguous word’s context-appropriate meaning. The outcome of this process is influenced both by top-down contextual support and bottom-up, word-specific characteristics. We examined how these factors jointly affect the neural signatures of lexical ambiguity resolution. We measured the similarity between multi-voxel patterns evoked by the same homonym in two distinct linguistic contexts: once after subjects read sentences that biased interpretation toward each homonym’s most frequent, dominant meaning, and again after interpretation was biased toward a weaker, subordinate meaning. We predicted that, following a subordinate-biasing context, the dominant yet inappropriate meaning would nevertheless compete for activation, manifesting in increased similarity between the neural patterns evoked by the two word meanings. In left anterior temporal lobe (ATL), degree of within-word pattern similarity was positively predicted by the association strength of each homonym’s dominant meaning. Further, within-word pattern similarity in left ATL was negatively predicted by item-specific responses in a region of left ventrolateral prefrontal cortex (VLPFC) sensitive to semantic conflict. These findings have implications for psycholinguistic models of lexical ambiguity resolution, and for the role of left VLPFC function during this process. Moreover, these findings demonstrate the utility of item-level, similarity-based analyses of fMRI data for our understanding of competition between co-activated word meanings during language comprehension.

Exhaustive semantic activation for reading ambiguous verbs in Chinese sentences

Studies of lexical ambiguity resolution in sentential contexts have not sufficiently considered the relatedness among an ambiguous word’s meanings as a predicting factor for semantic activation. To better understand the relation between lexical access and discourse processing and the effect of semantic relatedness on lexical ambiguity resolution, a cross-modal lexical priming experiment focusing on Mandarin ambiguous verbs of varying degrees of semantic relatedness was conducted. The results indicated that both meanings of an ambiguous verb were activated regardless of contextual biases and the degrees of semantic relatedness between the associated meanings. Taken together with previous research, the present study suggests that the meanings of an ambiguous word (i.e., homophonic homographs, which share both phonological and orthographic representations) are co-activated exhaustively if they are syntactically licensed by the context. These results thus support the exhaustive semantic activation model of lexical ambiguity resolution and the syntax-first theory of sentence processing.

On Chinese EFL Learners’ Homonym Processing in Relation to Their Organization of L2 Mental Lexicon

Based on the models of lexical ambiguity resolution and the organization of L2 mental lexicon proposed by previous researchers in recent years, this paper aimed at investigating the correlation between the Chinese EFL learners’ homonym processing in relation to their organization of L2 mental lexicon through cross-modal sentence-priming lexical decision tasks and word association tasks completed by both English major and non-English major students. The results showed that there is a positive correlation between the learners’ organization of L2 mental lexicon and their performance on English homonym processing. Inferential statistic analysis indicates that learners whose L2 mental lexicon is more syntagmatically related can process the English homonym more effectively. They can use contextual cues (semantic and syntactic information) to deactivate inappropriate meanings while low proficiency learners may be slower and less effective in using the sentence context cues needed for disambiguation. Therefore, it has been tentatively indicated that effective spreading activation of a word during reading and listening is important and helpful for learners’ comprehension, and pedagogically suggested that consciously categorizing syntagmatically related words in L2 mental lexicon during teaching and learning new English words can help reconstruct lexical networks so as to correctly and quickly retrieve the words from the mental lexicon in a certain context and better understand the meaning of the whole context in reading and listening.

Different temporal pattern of information processing about facial emotion conveyed by single neurons between the monkey amygdala and ventrolateral prefrontal cortex

To successfully comprehend a sentence that contains a homonym, readers must select the ambiguous word’s context-appropriate meaning. The outcome of this process is influenced both by top-down contextual support and bottom-up, word-specific characteristics. We examined how these factors jointly affect the neural signatures of lexical ambiguity resolution. We measured the similarity between multi-voxel patterns evoked by the same homonym in two distinct linguistic contexts: once after subjects read sentences that biased interpretation toward each homonym’s most frequent, dominant meaning, and again after interpretation was biased toward a weaker, subordinate meaning. We predicted that, following a subordinate-biasing context, the dominant yet inappropriate meaning would nevertheless compete for activation, manifesting in increased similarity between the neural patterns evoked by the two word meanings. In left anterior temporal lobe (ATL), degree of within-word pattern similarity was positively predicted by the association strength of each homonym’s dominant meaning. Further, within-word pattern similarity in left ATL was negatively predicted by item-specific responses in a region of left ventrolateral prefrontal cortex (VLPFC) sensitive to semantic conflict. These findings have implications for psycholinguistic models of lexical ambiguity resolution, and for the role of left VLPFC function during this process. Moreover, these findings demonstrate the utility of item-level, similarity-based analyses of fMRI data for our understanding of competition between co-activated word meanings during language comprehension.

Using puns to study contextual influences on lexical ambiguity resolution: Evidence from eye movements

Participants' eye movements were monitored while they read sentences containing biased homographs in either a single-meaning context condition that instantiated the subordinate meaning of the homograph without ruling out the dominant meaning (e.g., "The man with a toothache had a crown made by the best dentist in town") or a dual-meaning pun context condition that supported both the subordinate and dominant meanings (e.g., "The king with a toothache had a crown made by the best dentist in town"). In both of these conditions, the homographs were followed by disambiguating material that supported the subordinate meaning and ruled out the dominant meaning. Fixation times on the homograph were longer in the single-meaning condition than in the dual-meaning condition, whereas the reverse pattern was demonstrated for fixation times on the disambiguating region; these effects were observed as early as first-fixation duration. The findings strongly support the reordered access model of lexical ambiguity resolution.

Sentential and discourse topic effects on lexical ambiguity processing: an eye movement examination.

Models of lexical ambiguity resolution posit a role for context, but this construct has remained relatively undefined in the literature. The present study isolated two different forms of contextual constraint and examined how these sources of information might differentiate between a selective access and a reordered access model of ambiguity processing. Eye movements were monitored as participants read passages that contained either a balanced or a biased ambiguous word. The sentence containing the ambiguous word was held constant and instantiated either the subordinate meaning (Experiment 1) or the dominant meaning (Experiment 2) through the use of local context. These sentences were embedded in passages in which the topic was consistent, inconsistent, or neutral with respect to the meaning biased by the critical sentence. Experiment 1 provided evidence suggesting that the subordinate meaning of an ambiguous word was not selectively accessed even when sentence and discourse topic information biased that meaning. The data from Experiment 2 provided evidence that even the dominant meaning was not selectively accessed. These contextual sources of information were evaluated in terms of the roles they play in models of lexical ambiguity resolution.

The influence of global discourse on lexical ambiguity resolution

The influence of global discourse on the resolution of lexical ambiguity was examined in a series of naming experiments. Two-sentence passages were constructed to bias either the dominant or the subordinate meaning of a homonym that was embedded in a locally ambiguous sentence. The results provided evidence for the immediate (0-msec interstimulus interval) resolution of lexical ambiguity and were subsequently replicated in Experiment 2, in which an 80-msec stimulus onset asynchrony exposure duration was employed for the homonyms. Strong dominant and subordinate biased discourse contexts activated only the contextually appropriate sense of a homonym. In Experiment 3, each sentence of the discourse was presented in isolation. The pattern of activation obtained in Experiments 1 and 2 was found to be contingent on the integration of the two sentences to construct an overall global discourse representation of the text. The results support a context-sensitive model of lexical ambiguity resolution.

Contextual Strength and the Subordinate Bias Effect: Comment on Martin, Vu, Kellas, and Metcalf

Martin, Vu, Kellas, and Metcalf (this issue) claim to have demonstrated that the subordinate bias effect (when preceding context instantiates the subordinate meaning of an ambiguous word that has a highly dominant meaning, reading time on that word is lengthened) can be eliminated by strong context. They argue that this provides evidence critical to discriminating between competing models of lexical ambiguity resolution: the reordered access model (in which access of meanings for an ambiguous word is exhaustive but in which the order of access is influenced by prior disambiguating context) and the context-sensitive model (in which access is selective in the presence of prior disambiguating information). We argue that there are methodological problems with their demonstration, but even if there were not, it is unclear that the subordinate bias effect is appropriate for discriminating between competing models of lexical ambiguity resolution (the reordered access model and the context-sensitive model). The effect is an empirical finding and not a fundamental tenet of the reordered access model.

Contextual Strength Modulates the Subordinate Bias Effect: Reply to Rayner, Binder, and Duffy

Martin, Vu, Kellas, and Metcalf (this issue) report research indicating that strongly biased contexts can eliminate a subordinate bias effect, whereas in weakly biased context, a subordinate bias effect is evident. Martin et al. further show that in weakly biased contexts the subordinate bias effect increased in a continuous fashion as the frequency difference between dominant and subordinate meanings of ambiguous words increased. The results supported a context-sensitive model of lexical ambiguity resolution. Rayner, Binder, and Duffy (this issue) challenge the research on theoretical and methodological grounds. Here, we address the major issues raised by Rayner et al. We refer the reader to Kellas and Vu (in press) for additional comments.

Strength of discourse context as a determinant of the subordinate bias effect.

Two experiments examined the influence of strength of discourse bias on lexical ambiguity resolution. Short passages were constructed to bias polarized ambiguous words (homonymous) strongly or weakly toward the dominant or subordinate meanings. Using a self-paced reading task in Experiment 1, it was demonstrated that in strongly biased discourse, reading times for homonyms in dominant discourse did not differ from those in subordinate discourse. However, when the discourse was weakly biased, homonyms were read faster in dominant discourse than in subordinate discourse. Experiment 2 combined the reading paradigm with a naming task in order to provide an assessment of specific word-meaning activation. Reading times on ambiguous words replicated the results of Experiment 1. In addition, naming latencies for probe words revealed that only the contextually appropriate sense of a homonym was activated in strongly biased discourse. In contrast, both contextually appropriate and inappropriate senses were activated following a weakly biased subordinate discourse, whereas only the dominant sense was activated following weakly biased dominant discourse. The results demonstrate (1) an immediate influence of prior discourse information on lexical processing; and (2) that the strength of discourse constraints can play a governing role in lexical ambiguity resolution. The results were interpreted within the framework of a context-sensitive model of lexical ambiguity resolution.

A Parallel Distributed Processing Model of Lexical Ambiguity Resolution:

A Parallel Distributed Processing Model of Lexical Ambiguity Resolution:

A Connectionist Model of Lexical Ambiguity Resolution in Sentence Comprehension

A Connectionist Model of Lexical Ambiguity Resolution in Sentence Comprehension

University of Alberta norms of relative meaning frequency for 566 homographs

For many models of lexical ambiguity resolution, relative frequency of the different meanings of homographs (words with more than one meaning) is crucial. Although several homograph association norms have been published in the past, none has involved a large number of subjects responding to a large number of homographs, and most homograph norming studies are now at least a decade old. In Experiment 1, associations to 566 homographs were collected from an average of 192 subjects per homograph. Frequency of occurrence for the three most common meanings is reported, along with the corresponding associates, and a measure of the overall ambiguity of each homograph. Homographs whose meanings differed in part of speech were more ambiguous overall than homographs whose different meanings belonged to a single grammatical class. Homographs whose pronunciation depended on meaning (heterophones) were no more ambiguous than nonheterophones, and word frequency was unrelated to overall ambiguity. Estimates of homograph balance across different norming studies were compared, and homographs with two meanings of approximately equal relative meaning frequency (balanced homographs) and homographs with one clearly dominant meaning (polarized homographs) were identified. In Experiment 2, reliability of meaning categorizations was measured for a subset of the homographs in the first experiment. Meaning categorizations were shown to be highly reliable across raters.

The Effect of Meaning Frequency on Processing Lexically Ambiguous Words: Evidence from Eye Fixations

Subjects read sentences containing lexically ambiguous words while their eye movements were monitored Biased ambiguous words (those that have one highly dominant sense) were used in sentences containing a prior context that instantiated their subordinate sense Control words were matched in frequency both to the dominant and to the subordinate meaning of the ambiguous word (high- and low-frequency controls) Subjects fixated longer on both the ambiguous word and the low-frequency control than on the high-frequency control When the target was ambiguous, however, the duration of posttarget fixations was longer and the likelihood of making a regression to the target was greater than when the target was an unambiguous control The results are discussed in relation to current models of lexical ambiguity resolution

Lexical ambiguity and eye fixations in reading: A test of competing models of lexical ambiguity resolution

Eye movement data were used to test two models of lexical ambiguity resolution. When ambiguous target words are preceded by disambiguating context, the integration model holds that the disambiguating material has no effect on the initial access phase of the meaning assignment process. In contrast, the reordered access model holds that disambiguating material can affect the initial access phase of the process by increasing the availability of one of the potential meanings. Subjects read sentences containing ambiguous words that had one very dominant and another more subordinate meaning. The words were preceded either by (1) evidence that the subordinate meaning of the word should be instantiated, (2) evidence that the dominant meaning should not be instantiated, or (3) neutral material. The ambiguous words were followed by material that conclusively disambiguated them toward their subordinate meanings. Gaze durations on the ambiguous words were longer when the words were preceded by evidence against their dominant meanings. Processing times for the disambiguating material were longer when the words were preceded by neutral material. The results were interpreted as support for the reordered access model.