David Eddington

Analogy and the dual-route model of morphology

Abstract Prasada and Pinkers (1993) subjects provided past tense forms of nonce verbs. The subjects willingness to provide irregular past tense forms correlated with the verbs phonological similarity to existing irregular English verbs. However, there was no correlation between the number of nonce verbs assigned regular inflection, and the verbs similarity to existing regular verbs. According to the dual-route model, this is expected since irregular items are stored in associative memory, while regular items take an allomorph of -ed by rule. A singleroute connectionist simulation failed to duplicate the subjects behavior on regular verbs. Two instance-based models were applied to the data: Analogical Modeling of Language and the Tilburg Memory Based Learner. Each model employed a similarity algorithm to determine the behavior of all regular and irregular items. Both models successfully mirrored the subjects responses. Therefore, the data are consistent with an instance-based single-route model of morphology.

DIALECT IDENTIFICATION: THE EFFECTS OF REGION OF ORIGIN AND AMOUNT OF EXPERIENCE

This study examines whether two factors, region of origin (i.e., being from either Utah, Western states, or non-Western states) and amount of experience for those not native to Utah (having less than one, more than one but less than five, or over five years living in Utah), influence how well listeners are able to distinguish between Utah and non-Utah speakers and what phonetic characteristics they use to do so. The results suggest that the more similar the listeners dialect is to Utah English, the better his or her ability to identify Utah speakers. Moreover, it was found that listeners from Utah use less stereotypical characteristics of Utah English for identifying Utahns from non-Utahns; those from the Western United States and other locations use more. This study demonstrates that listeners with more experience with Utah English are better able to identify Utah speakers than those with less experience. These findings are also examined in light of stereotypical perceptions of both Utah English and the phonetic characteristics examined in this study.

Dissociation in Italian conjugations: A single-route account

Say and Clahsen (S and C) report an experiment involving assignment of past participle suffixes to nonce words in Italian. Their evidence suggests a dual-route model that assigns the theme vowel of the 1st conjugation, while storing it lexically in other conjugations. However, these nonce words were assigned suffixes by a computer algorithm that determined the past participle of the nonce items on the basis of phonological similarity to existing inflected forms. The outcome of the simulation mirrors that of the subjects closely, suggesting that the dissociations found by S and C are not adequate evidence for a dual-route model of Italian stem formation.

Issues in modeling language processing analogically

Exemplar-based models of language assert that linguistic processing involves analogy to past linguistic experiences stored in the mental lexicon. This study explores how three factors influence the predictions made by exemplar-based simulations of linguistic phenomena. Three questions are posed as they relate to such simulations: (1) Is type frequency or token frequency a better predictor of outcomes? (2) What is the optimal way of aligning the variables in the database so that the most relevant analogs are found? (3) Are there significant differences between representing variables as phonemes versus representing them in terms of distinctive features? Spanish stress assignment and English past tense formation served as the linguistic phenomena on which these issues were tested. The results suggest that type frequency is a better predictor of outcomes, although simulations using token frequency were most successful when only middle frequency words were included. Several methods for aligning variables in the analogical database are discussed. The dual-alignment method has advantages for the English past tense task, but not in predicting Spanish stress. In the Spanish task, strict phonemic representation of words demonstrated no advantage over feature representation. However, phonemic representation produced better results than distinctive features in predicting the English past tense.

T-GLOTTALIZATION IN AMERICAN ENGLISH

In word-final prevocalic position (e.g., right ankle), there are various possible phonetic realizations of /t/ in American English: [t], [r], [p]. The present study focuses on the linguistic and social factors associated with the use of the glottal stop. Data were gathered by having participants repeat sentences they were presented auditorily (e.g., She twisted her right ankle). The particular pronunciation of /t/ in the presented sentences was masked with a tone. Logistic regression analysis identified three significant factors: (1) glottal stops were favored by following front vowels; ( 2 ) younger female speakers were most likely to use glottal stops, which may indicate a change in progress; and ( 3 ) speakers from the Western United States glottalized more than speakers from other parts of the country.

Syllabification of American English: Evidence from a Large- scale Experiment. Part II

Abstract 4990 bi-syllabic English words were syllabified by about 22 native speakers who choose between different slash divisions (e.g. photon: FOW/TAHN, FOWT/AHN). Results of the regression analyses of the items with one medial consonant are discussed. Consistent with previous studies, consonants were drawn to stressed syllables, and more sonorant consonants were more often placed in the coda. A model in which syllables are made to be as word-like as possible is supported; syllables were often created that begin and end in the same phonemes that are legal word-initially and finally, and syllabifications tended to follow morphological boundaries. Orthographic conventions, such as not placing ck or ll syllable-initially were also followed.

Argument constructions and language processing: Evidence from a priming Experiment and pedagogical implications

The notion of argument construction is widely accepted in Cognitive Linguistics circles as a highly explanatory theoretical construct. It has recently been incorporated into the Lexical-Constructional Model (LCM; Ruiz de Mendoza and Mairal 2007, 2008), a theoretical approach to meaning construction that integrates argument constructions into a broader model that incorporates meaning dimensions traditionally dealt with in the domain of pragmatics and discourse analysis. The LCM has an argument level of description, in the form of constructional templates, which are in general equivalent to the argument constructions postulated in Goldberg’s (1995, 2006) Construction Grammar, although there are crucial differences in the descriptive metalanguage and the principles used to explain their linguistic behavior. Since the LCM is geared to have psychological adequacy, like all cognitive (cf. Lakoff 1990) and some functional approaches to language (e.g. Dik 1997), it is essential to find to what extent the notion of argument construction is a real processing mechanism. This paper presents a priming experiment measuring reaction times to sentences that had the same formal configurations but were based on different argument constructions, and to sentences that were formally similar and embodied the same argument construction. Subjects responded significantly faster to target sentences that shared the same construction as the previously viewed sentences. Thus, constructions appear to be psychological entities that play an important role in linguistic processing. If this is correct, then it has potentially important consequences for planning adequate L2 teaching

AMERICAN ENGLISH HAS GOʔ A LOʔ OF GLOTTAL STOPS: SOCIAL DIFFUSION AND LINGUISTIC MOTIVATION

In word-final, prevocalic position (e.g., it is) there are various possible phonetic realizations of /t/ in American English (e.g., [t], [r] , [?] ). The present study examines the linguistic and social factors associated with the use of the glottal stop in American English in 1,101 instances of word-final, prevocalic /t/ from the Santa Barbara Corpus. The glottal stop occurred in 24 % of the cases. Logistic regression analysis was used to identify factors that favor glottaling of /t/. Our findings concur with previous research in that age and region were significant: Westerners in their teens and 20s glottalized more than non-Westerners in the same age groups; speakers who are 30 and older, both Westerners and non-Westerners, glottalize to a much smaller degree. We also found that glottaling is favored by a following stressed syllable; however, gender and following vowel quality were not influencing variables, which contradicts previous experimental findings. If prevocalic glottaling is uncommon word-internally in American English, why is it apparently spreading word-finally? We provide evidence that word-final /t/s are more often followed by word-initial consonants than vowels, which places them in a glottalizing context. Instances with a glottal realization are stored in the mental lexicon and are available as possible pronunciation choices even in prevocalic position.

The phonetic context of American English flapping: quantitative evidence.

The phonetic context in which word-medial flaps occur (in contrast to [th]) in American English is explored. The analysis focuses on stress placement, following phone, and syllabification. In Experiment 1, subjects provided their preference for [th] or [ ] in bisyllabic nonce words. Consistent with previous studies, flaps were preferred before stressless syllables and [th] before stressed syllables, but the following phone also exerted a small degree of influence. Experiments 2 and 3 tested whether [th ] or [ ] are associated with a particular syllable position in bisyllabic words. They demonstrate that [th] is favored in onsets, while [ ] is not consistently placed in either the onset or coda, nor is it generally ambisyllabic. These findings contradict analyses that posit syllable division as a conditioning factor in the appearance of [th] versus [ ]. Experiment 4 examined the pronunciation of 480 multisyllabic words from the TIMIT corpus. VCV was seen to favor [ ], while VCV favored [th]. In addition, flaps tend to be followed by syllabic sonorants and [th] by tense vowels. Because the following phones that influenced [th] and [ ] in Experiment 4 differ from those that were significant in Experiment 1, more research is necessary into the effect that following phones have on the appearance of [th] and [ ].

Where Are the Moun[ɁƏ]ns in Utah?

Linguists have investigated a number of traits that characterize English in Utah. Among these are the cord/card merger (Bowie 2003), the fill/fell merger (Baker and Eddington 2009, DiPaolo and Faber 1990), and the use of propredicate do (DiPaolo 1993). However, one of the most often stigmatized characteristics of Utah English is that Utahns “drop their Ts” in words such as mountain, and kitten, which is pervasive enough that it even receives local press coverage. One challenge for investigating the factors that influence t-dropping, as well as its social meaning, is defining exactly what t-dropping is in phonetic terms; it could involve actual deletion, or may refer to the glottal pronunciation, or it may entail an oral release after the glottal stop. The goal of the present study is to determine the phonetic correlate of this stigmatized pronunciation and to make an initial foray into how age, being from Utah, and gender affect it.