The dark side of incremental learning: A model of cumulative semantic interference during lexical access in speech production

Naming a picture of a dog primes the subsequent naming of a picture of a dog (repetition priming) and interferes with the subsequent naming of a picture of a cat (semantic interference). Behavioral studies suggest that these effects derive from persistent changes in the way that words are activated and selected for production, and some have claimed that the findings are only understandable by positing a competitive mechanism for lexical selection. We present a simple model of lexical retrieval in speech production that applies error-driven learning to its lexical activation network. This model naturally produces repetition priming and semantic interference effects. It predicts the major findings from several published experiments, demonstrating that these effects may arise from incremental learning. Furthermore, analysis of the model suggests that competition during lexical selection is not necessary for semantic interference if the learning process is itself competitive.

from Cognition

The dark side of incremental learning: A model of cumulative semantic interference during lexical access in speech production

Naming a picture of a dog primes the subsequent naming of a picture of a dog (repetition priming) and interferes with the subsequent naming of a picture of a cat (semantic interference). Behavioral studies suggest that these effects derive from persistent changes in the way that words are activated and selected for production, and some have claimed that the findings are only understandable by positing a competitive mechanism for lexical selection. We present a simple model of lexical retrieval in speech production that applies error-driven learning to its lexical activation network. This model naturally produces repetition priming and semantic interference effects. It predicts the major findings from several published experiments, demonstrating that these effects may arise from incremental learning. Furthermore, analysis of the model suggests that competition during lexical selection is not necessary for semantic interference if the learning process is itself competitive.

from Cognition

How the brain repairs stuttering

Stuttering is a neurodevelopmental disorder associated with left inferior frontal structural anomalies. While children often recover, stuttering may also spontaneously disappear much later after years of dysfluency. These rare cases of unassisted recovery in adulthood provide a model of optimal brain repair outside the classical windows of developmental plasticity. Here we explore what distinguishes this type of recovery from less optimal repair modes, i.e. therapy-induced assisted recovery and attempted compensation in subjects who are still affected. We show that persistent stuttering is associated with mobilization of brain regions contralateral to the structural anomalies for compensation attempt. In contrast, the only neural landmark of optimal repair is activation of the left BA 47/12 in the orbitofrontal cortex, adjacent to a region where a white matter anomaly is observed in persistent stutterers, but normalized in recovered subjects. These findings show that late repair of neurodevelopmental stuttering follows the principles of contralateral and perianomalous reorganization.

from Brain

Information structural constraints on children’s early language production: The acquisition of the focus particle auch (‘also’) in German-learning 12- to 36-month-olds

This article presents new findings for the acquisition of the focus particle auch (‘also’) in German-learning children. In a longitudinal study with 11 children between 1;00 and 3;00 years of age complemented by two experiments with children aged 2;4 and 2;8, the authors investigated children’s production of the accented and unaccented auch. The results confirm earlier findings of a temporal delay between the first occurrences of both auch-variants. Based on the empirical findings, an account for this asymmetry is proposed that relates it to a more general developmental tendency that is characterized by a growing linguistic explicitness in embedding a given utterance in its discourse context. It is suggested that the observed delay is caused by the type of relation between the particle and its related constituent: in contrast to the accented auch the unaccented auch is anaphorically related to the sentence topic. It is proposed that the initial omission reflects a general tendency in early child language to drop topic material.

from First Language

The prelexical development in children implanted by 16 months compared with normal hearing children

Objective
Few studies exist which document the early speech development of German-speaking children or German-speaking children who are deaf and using cochlear implants. The current study aims to: (1) document the pre-canonical and canonical speech development of German-speaking children who are deaf and receive cochlear implants by the age of 16 months and (2) compare these children’s results with those of children with normal hearing.

Design
This longitudinal study included 5 German-speaking children with normal hearing and 5 with sensorineural deafness. All children from the deaf group received hearing amplification before cochlear implantation, received their first implant by 16 months of age, and became bilateral implant users by 31 months of age. The pre-canonical and canonical vocalisations of each child were recorded on video- and audiotapes in a semi-standardised playing situation every 4 weeks over a span of 1 year. In the cochlear implant group, the recording started 4–5 days postoperatively (first implant); in the normal hearing group it began between the ages of 4 and 5 months. The video and audio recordings were analysed using EUDICO Linguistic Annotator version 2.4 (Nijmegen, The Netherlands) and International Phonetic Alphabet transcription.

Results
Both groups showed individual patterns of babbling acquisition, though the groups’ patterns of acquisition were similar when analysed for consonant manner and place. Some children started with plosives and others, with nasals, but all acquired fricatives and laterals next. Onset of canonical babbling for children in the cochlear implant group began 0–4 months after first fitting of the first device, while children from the normal hearing group demonstrated an onset of canonical babbling between 4 and 9 months of age.

Conclusion
Our results show that deaf children who receive cochlear implants at an early age are capable of reaching the canonical babbling milestone in a shorter time than children with normal hearing typically do and that their consonant phoneme acquisition follows a similar sequence to normal hearing peers’. These results are consistent with the literature indicating that early identification and intervention are important for allowing children with cochlear implants the opportunity to catch up to hearing peers.

from the International Journal of Pediatric Otorhinolaryngology

Speech and language outcomes in children with auditory neuropathy/dys-synchrony managed with either cochlear implants or hearing aids

The aim of this study was to assess the receptive language and speech production abilities of a group of school-aged children with auditory neuropathy/dys-synchrony-type hearing loss. Ten children who had received a cochlear implant in one or both ears participated. Findings for this group were compared with those for a matched cohort of implanted children with other forms of sensorineural hearing loss and with those for a group of auditory neuropathy/dys-synchrony children who were long-term hearing aid users.

Results for 9 of the ten implanted children with auditory neuropathy/dys-synchrony were similar to those of the general population of paediatric implant recipients. (One child, who gained little perceptual benefit from his device, showed severely delayed spoken language development). Results for the group of aided auditory neuropathy/dys-synchrony subjects were comparable to those for their implanted counterparts suggesting that affected children should not automatically be considered cochlear implant candidates.

from the International Journal of Audiology

Do French–English Bilingual Children Gesture More Than Monolingual Children?

Abstract Previous studies have shown that bilingual adults use more gestures than English monolinguals. Because no study has compared the gestures of bilinguals and monolinguals in both languages, the high gesture rate could be due to transfer from a high gesture language or could result from the use of gesture to aid in linguistic access. In this study we tried to distinguish between those causes by comparing the gesture rate of 10 French–English bilingual preschoolers with both 10 French and 10 English monolinguals. All were between 4 and 6 years of age. The children were asked to watch a cartoon and tell the story back. The results showed the bilingual children gestured more than either group of monolinguals and at the same rate in both French and English. These results suggest that that the bilinguals were not gesturing because they were transferring the high gesture rate from one language to another. We argue that bilinguals might gesture more than monolinguals to help formulate their spoken message.

from the Journal of Psycholinguistic Research

An evaluation of frequency transposition for hearing-impaired school-age children

A key objective when fitting hearing aids to children is to maximize the audibility of high frequency speech cues which are critical in the understanding of spoken English. Recent advances in digital signal processing have enabled the development of hearing aids which offer linear frequency transposition as a new way of accessing these important speech sounds. This study examined the changes in aided performance observed in children with hearing impairment who trialed the alternative hearing aid technology. The rationale behind the research was to evaluate the benefits of this new technology, as applied in the commercially available Widex Inteo hearing aid, for a group of children with sloping high frequency sensori-neural hearing loss (n = 6). The participants were aged between 9 and 14 years and all attended mainstream schools. Outcome measures were comprised of tests of speech perception and speech production and questionnaires. Speech perception abilities were measured using CNC word and phoneme tests, with and without frequency transposition in both the audiovisual and visual alone modalities at 6 weekly intervals for 24 weeks. The Goldman-Fristoe 2 Test of Articulation was the chosen measure of speech production and was administered prior to the new aid fitting and after 24 weeks of frequency transposition use. Results showed statistically significant improvements for the group as a whole on both these measures. These results were supported by positive reporting on the questionnaires completed by key stakeholders. Copyright © 2009 John Wiley & Sons, Ltd.

from Deafness and Education International

Modelling the architecture of phonetic plans: Evidence from apraxia of speech

Abstract
In theories of spoken language production, the gestural code prescribing the movements of the speech organs is usually viewed as a linear string of holistic, encapsulated, hard-wired, phonetic plans, e.g., of the size of phonemes or syllables. Interactions between phonetic units on the surface of overt speech are commonly attributed to either the phonological encoding stage or the peripheral mechanisms of the speech apparatus. Apraxia of speech is a neurogenic disorder which is considered to interfere with the mechanisms of phonetic encoding. Analyses of apraxic speech errors have suggested that phonetic representations have a non-linear, hierarchically nested structure. This article presents a non-linear probabilistic model of the phonetic code, which embraces units from a sub-segmental level up to the level of metrical feet. The model is verified on the basis of accuracy data from a large sample of apraxic speakers.

from Language and Cognitive Processes

Competing conceptual representations trigger co-speech representational gestures

Abstract
Various interconnections between the processing of speech and gesture have been demonstrated in the literature. However, it is not well understood what cognitive factors influence intra-speaker variation of the gesture frequency. This study investigates the hypothesis that provides a unifying explanation for the previous findings on this issue; namely, competing representations in the conceptualisation process for speaking trigger representational gestures. Twenty adult participants described complex geometric figures. In the easy condition, the organisation of the lines necessary for conceptualisation were highlighted by making some lines darker, but in the hard condition, unnecessary competitor conceptualisations were highlighted. The descriptions in the two conditions were lexically comparable. However, the rate of representational gestures (but not that of beat gestures) was higher in the hard condition than in the easy condition. This finding is compatible with the idea that gestures may play a role in the conceptualisation process for speaking.

from Language and Cognitive Processes

Modelling the architecture of phonetic plans: Evidence from apraxia of speech

In theories of spoken language production, the gestural code prescribing the movements of the speech organs is usually viewed as a linear string of holistic, encapsulated, hard-wired, phonetic plans, e.g., of the size of phonemes or syllables. Interactions between phonetic units on the surface of overt speech are commonly attributed to either the phonological encoding stage or the peripheral mechanisms of the speech apparatus. Apraxia of speech is a neurogenic disorder which is considered to interfere with the mechanisms of phonetic encoding. Analyses of apraxic speech errors have suggested that phonetic representations have a non-linear, hierarchically nested structure. This article presents a non-linear probabilistic model of the phonetic code, which embraces units from a sub-segmental level up to the level of metrical feet. The model is verified on the basis of accuracy data from a large sample of apraxic speakers.

from Language and Cognitive Processes

Competing conceptual representations trigger co-speech representational gestures

Various interconnections between the processing of speech and gesture have been demonstrated in the literature. However, it is not well understood what cognitive factors influence intra-speaker variation of the gesture frequency. This study investigates the hypothesis that provides a unifying explanation for the previous findings on this issue; namely, competing representations in the conceptualisation process for speaking trigger representational gestures. Twenty adult participants described complex geometric figures. In the easy condition, the organisation of the lines necessary for conceptualisation were highlighted by making some lines darker, but in the hard condition, unnecessary competitor conceptualisations were highlighted. The descriptions in the two conditions were lexically comparable. However, the rate of representational gestures (but not that of beat gestures) was higher in the hard condition than in the easy condition. This finding is compatible with the idea that gestures may play a role in the conceptualisation process for speaking.

from Language and Cognitive Processes

Are the same phoneme and lexical layers used in speech production and comprehension? A case-series test of Foygel and Dell’s (2000) model of aphasic speech production

In this paper, we investigate the claim that although the same lexical units are involved in speech production and comprehension, there are separate input and output phoneme layers (Foygel and Dell, 2000). Data from a case series of aphasic patients are used to test this claim by examining the relationship between performance on a test of picture naming and performance on tests of phonological input. Estimates of each patient’s semantic-lexical and phonological impairments in speech production were derived from Foygel and Dell’s computational model of picture naming. It was found that the strength of the semantic-lexical impairments in speech production was significantly correlated with performance on auditory comprehension tests. This finding is consistent with the claim that the same lexical units are involved in speech comprehension and production. Conversely, the correlations between the strength of the phonological lesions in speech production and performance on tests of phonological input were non-significant, consistent with Foygel and Dell’s claim that there are distinct input and output phoneme layers.

from Cortex

A mediating role of the premotor cortex in phoneme segmentation

Abstract
Consistent with a functional role of the motor system in speech perception, disturbing the activity of the left ventral premotor cortex by means of repetitive transcranial magnetic stimulation (rTMS) has been shown to impair auditory identification of syllables that were masked with white noise. However, whether this region is crucial for speech perception under normal listening conditions remains debated. To directly test this hypothesis, we applied rTMS to the left ventral premotor cortex and participants performed auditory speech tasks involving the same set of syllables but differing in the use of phonemic segmentation processes. Compared to sham stimulation, rTMS applied over the ventral premotor cortex resulted in slower phoneme discrimination requiring phonemic segmentation. No effect was observed in phoneme identification and syllable discrimination tasks that could be performed without need for phonemic segmentation. The findings demonstrate a mediating role of the ventral premotor cortex in speech segmentation under normal listening conditions and are interpreted in relation to theories assuming a link between perception and action in the human speech processing system.

p>

from Brain and Language

Impact of newborn hearing screening

Conclusions:
The results of this extensive study of profoundly deaf children with CIs in Flanders indicate that a newborn hearing screening program results in earlier intervention in deaf children, which beneficially influences the auditory receptive skills and speech intelligibility. Laryngoscope, 2009

from Laryngoscope