Phonological Speech Errors Research Articles

Phonetic variability in /s/ and /z/ segment errors: A multitalker acoustic study

Phonological speech errors are a traditional source of evidence for phonological units, such as features and segments, in both linguistic and psycholinguistic theory. Assumptions about the unitary nature of speech errors have been brought into question by research that found evidence for subfeatural errors in speech production [R. Mowrey and I. MacKay, J. Acoust. Soc. Am. 88, 1299–1312 (1990)]. This poster presents data from an ongoing acoustic study of speech errors in /s/ and /z/ productions. Previously reported findings indicate that the errors involve either single or multiple changes in periodicity, amplitude and duration of frication, and vowel rise time [S. Frisch and R. Wright, J. Acoust. Soc. Am. 101, 3178(A) (1997)]. The data are tape recordings of an error-inducing experiment using nonsense tongue twisters. Recordings of six talkers producing four different tongue twisters targeting /s/ and /z/, for examples, ‘‘sit zap zoo sip,’’ were digitized and analyzed. This study examines the phonetic variation in featural and subfeatural errors across talkers and segmental contexts. Results suggest that details of individual talker’s production strategies and the phonetic context influence the distribution of featural and subfeatural errors. [Work supported by NIH Grant No. DC00012.]

The WEAVER model of word-form encoding in speech production

Lexical access in speaking consists of two major steps: lemma retrieval and word-form encoding. In Roelofs (Roelofs, A. 1992a. Cognition 42, 107–142; Roelofs, A. 1993. Cognition 47, 59–87.), I described a model of lemma retrieval. The present paper extends this work by presenting a comprehensive model of the second access step, word-form encoding. The model is called WEAVER (Word-form Encoding by Activation and VERification). Unlike other models of word-form generation, WEAVER is able to provide accounts of response time data, particularly from the picture-word interference paradigm and the implicit priming paradigm. Its key features are (1) retrieval by spreading activation, (2) verification of activated information by a production rule, (3) a rightward incremental construction of phonological representations using a principle of active syllabification, syllables are constructed on the fly rather than stored with lexical items, (4) active competitive selection of syllabic motor programs using a mathematical formalism that generates response times and (5) the association of phonological speech errors with the selection of syllabic motor programs due to the failure of verification.

Acoustic evidence for featural and subfeatural errors in speech production

Phonological speech errors are a traditional source of psycholinguistic evidence for the representations of phonology theory. In an electromyographic (EMG) study of experimentally induced phonological speech errors, Mowrey and MacKay [J. Acoust. Soc. Am. 88, 1299–1312] found that speech errors frequently occur at a subfeatural, gestural level, with no apparent effect on the percept of the word. Mowrey and MacKay’s study considered the activity of a single muscle, and thus was unable to determine whether single gestures acted independently of gestural constellations. This study is a preliminary report from an ongoing acoustic analysis of speech errors. The data are tape recordings of an error-inducing experiment using nonsense tongue twisters. Recordings of six speakers producing four different tongue twisters targeting /s/ and /z/, e.g., sit zap zoo sip, were digitized and analyzed. Some errors involved multiple changes in acoustic properties, including simultaneous changes in periodicity, amplitude of frication, and duration, while others involved a subset of these properties. This evidence suggests that errors can occur at both the single gesture level, affecting noncontrastive acoustic properties, and at the level of the gestural complex or segment, creating a perceptible, linguistically contrastive change. [Work supported by NIH Grant No. DC00012.]

‘‘SLIP-ing’’ in phonologically similar neighborhoods

Phonological speech errors are the class of speech errors in which one or more phonemes are added, deleted, substituted, or reversed. Previous work [Dell (1990); Stemberger and MacWhinney (1986)] has found that low-frequency words are more likely to be involved in phonological speech errors than high-frequency ones. If word frequency, influences performance in speech production, do other characteristics of words, such as neighborhood density or neighborhood frequency, influence speech production? The influence of phonologically similar neighborhoods on a type of whole word speech error (malapropisms) has previously been demonstrated [Vitevitch (1996)]. The present research examined the influence of phonologically similar neighborhoods on phonological speech errors. Using the SLIPS technique [Motley and Baars (1976)], phonological speech errors were elicited from participants with real word pairs that varied in their neighborhood characteristics. The results show that low-frequency words ‘‘slipped’’ more often than high-frequency ones, replicating past results from error corpora and from other studies using this methodology. Moreover, words with sparse phonologically similar neighborhoods tended to ‘‘slip’’ more than words with dense phonologically similar neighborhoods. The implications for these findings on models of speech production, speech perception, and lexical representation are discussed.

The unfolding of suprasegmental representations: a cross-linguistic perspective

An analysis of phonological speech errors in Arabic, English and German is carried out with a view to probing into the organization of segments within syllables and words. Arabic slips are shown to be less structure-sensitive than English and German ones. Being absent from underlying representation, suprasegmental structures are assumed to unfold gradually in real time. The erection of hierarchical representations is claimed to be slower in Arabic than in English and German because the nonconcatenative morphology of Arabic prevents an early assignment of consonants to structural slots. In contrast, English and German words allow the early build-up of hierarchical structures because sufficient phonological information is available from the beginning of the derivation.

Consonant Harmony as a Compensatory Mechanism in Fluent Aphasic Speech

This study addresses how fluent aphasics construct complete phonological representations, given the premise that their phonological speech errors result from faulty information about stored lexical representations. We explored whether consonant harmony, a common rule-governed process of feature copying, operates as a compensatory device for completing phonological representations in fluent aphasia. This was examined in a corpus of phonemic paraphasias (n = 543) produced by 8 fluent aphasics during picture naming. Consonant substitutions due to a single feature change (n = 143) were analyzed for the properties of consonant harmony predicted by the phonological principles embodied in a Universal Markedness version of Underspecification Theory (e.g., Chomsky and Halle, 1968). Results indicated that harmony constrained the feature substitution errors involving the feature class of voice (e.g., calendar→ /gæl▪viŋ/), but not place of articulation (e.g., igloo→/ıdlu/); substitutions due to an error in manner were rare. These findings were used to argue that for English-speaking fluent aphasics a consonant harmony rule for the feature voice is incorporated into a compensatory output mechanism that is used to complete faulty lexical-phonological representations.

Structure and content in language production: A theory of frame constraints in phonological speech errors

Theories of language production propose that utterances are constructed by a mechanism that separates linguistic content from linguistic structure, Linguistic content is retrieved from the mental lexicon, and is then inserted into slots in linguistic structures or frames. Support for this kind of model at the phonological level comes from patterns of phonological speech errors. W present an alternative account of these patterns using a connectionist or parallel distributed proceesing (PDP) model that learns to produce sequences of phonological features. The model's errors exhibit some of the properties of human speech errors, specifically, properties that have been attributed to the action of phonological rules, frames, or other structural generalizations.

Structure and Content in Language Production: A Theory of Frame Constraints in Phonological Speech Errors

Theories of language production propose that utterances are constructed by a mechanism that separates linguistic content from linguistic structure, Linguistic content is retrieved from the mental lexicon, and is then inserted into slots in linguistic structures or frames. Support for this kind of model at the phonological level comes from patterns of phonological speech errors. W present an alternative account of these patterns using a connectionist or parallel distributed proceesing (PDP) model that learns to produce sequences of phonological features. The model's errors exhibit some of the properties of human speech errors, specifically, properties that have been attributed to the action of phonological rules, frames, or other structural generalizations.

Effects of Frequency and Vocabulary Type on Phonological Speech Errors

Abstract Phonological speech errors only rarely occur in common function words. This paper explores whether this effect is due to these words' status as function (closed-class) morphemes, or simply their high frequency of usage. Experiments that elicit phonological speech errors revealed strong requency effects (the segments of frequent words tend not to be misordered) and no effect of vocabulary type-the segments of function morphemes (e.g. by) were as likely to slip as those of homophonic content items (e.g. buy). A spreading-activation model of lexical retrieval in production was adapted to account for the effects of frequency on phonological errors.

The retrieval of phonological forms in production: Tests of predictions from a connectionist model

This article reviews a model of retrieval processes in language production that accounts for phonological speech-error data ( Dell, 1986, Psychological Review, 93, 283–321). A distinction is drawn between empirical phenomena that are built into the model and phenomena that can be considered predictions from it. Data from experiments creating phonological speech errors are presented as tests of those predictions.