Sentence Repetition as a Measure of Early Grammatical Development in Italian

Devescovi, A., & Caselli, M.C. (2007). Sentence repetition as a measure of early grammatical development in Italian. International Journal of Language & Communication Disorders, 42(2), 187–208. 

Early identification of language impairment is essential in order to provide timely intervention, however, few standardized measures exist for children under four years of age. Devescovi and Caselli suggest that the Sentence Repetition Test (SRT) could serve as a suitable assessment of language development for young children based on the premise that a child’s ability to repeat a sentence is representative of her spontaneous language abilities. Consensus has not been reached on whether the SRT is an accurate measure of language ability; some say it overestimates ability, while others argue it underestimates it. Most significantly though, SRT has been recognized as a tool for identifying language impairment (e.g., Conti-Ramsden et al., 2001).

In the first of two studies, Devescovi and Caselli examine test-retest reliability of SRT in 100 Italian-speaking children aged 2;0 to 4;0. The authors find that calculating the mean length of utterance in words is sensitive to linguistic development between the ages of 2;0 and 2;6, and that number of sentences repeatedly correctly better discriminates the older age bands. It is important to note that MLU is calculated in words in this study by counting not morphemes, but whole words, which is common in studies of the development of Italian. Perhaps calculation of MLU using morphemes would offer better discrimination beyond 2;6.

In the second study, Devescovi and Caselli examine the correlation between SRT performance, verbal memory span (VMS), and spontaneous speech (structured conversation and picture description), noting that verbal memory span has been shown to predict MLU in spontaneous speech (Blake et al., 1994). SRT responses and spontaneous utterances are analyzed for three factors: number of verbs, omitted articles, and MLU (in words). Analysis revealed no correlation between any of these measures and VMS after controlling for age. Article omission and number of verbs in SRT were both related to spontaneous speech, while correlation between MLU (in words) of SRT and spontaneous speech was not significant.

The sentences used in the SRT included simple sentence structure and only those morphemes expected to develop by 4;0. While it is important to include sentences that could be repeated by the youngest children in the age span, it is possible that limiting the complexity of the stimuli may have limited the performance of the older children, causing a ceiling effect. Despite limitations of the task, positive correlations shown in the findings suggest that some form of SRT may be beneficial in assessing language of young children.

Blogger: Laura Pauls

Variability and Detection of Invariant Structure

Variability and Detection of Invariant Structure. Gomez, R. L. (2002). Psychological Science, 13, 5.

Recent theories of language learning suggest that language learners are particularly adept at tracking the statistical regularities that exist within natural language. For example, learners are able to sensitive to “adjacent dependencies”, that is, the higher probabilities that occur between adjacent syllables within words than between words. However, a key aspect of language learning involves learning patterns between “non-adjacent dependencies”. Some examples of nonadjacent dependencies experienced in everyday language use are the dependencies between auxiliaries and inflectional morphemes (e.g.: is running), and dependencies involving agreement (e.g.: the kids in the park are playing). What is of interest to researchers is how humans learn to track these nonadjacent relationships.

In an experimental task involving learning nonadjacent dependencies, subjects listen to a string of syllables and must learn that a particular non-adjacent pattern – for example, that a first element always occurs with a third element in the string. A key hypothesis in this learning regards the middle element. If the middle element varies more, the pattern between the first and third element becomes more salient and the learner shows greats sensitivity to learning the nonadjacent dependency.

In this study, adults and infants were exposed to an artificial language containing nonadjacent dependencies between words (e.g.: aXd, bXe, cXf). The set size of X varied between conditions, drawing X from a set size of either 2, 6, 12, or 24 elements. Learning was greater for when the set size from which X was drawn was highly variable (set size = 24). These results demonstrate that learning is adaptable may be dynamically guided by the statistical structure in their linguistic environment. It should be noted that these results have not been replicated in all subsequent studies.

Statistical learning in children with language impairment has been investigated only recently (see blogpost from Evans et al., 2009). It may be that these children have more difficulty recognizing such patterns in language. Current interventions involving explicitly teaching linguistic rules may facilitate this learning.

Blogger: Nicolette Noonan

Relationships Among Linguistic Processing Speed, Phonological Working Memory, and Attention in Children Who Stutter

Anderson, J. D., & Wagovich, S. A. (2010). Relationships among linguistic processing speed, phonological working memory, and attention in children who stutter. Journal of Fluency Disorders, 35, 216-234. 

Studies of speech reaction time (SRT) are thought to examine the efficiency with which a person processes and responds to language-based stimuli. The few studies that have examined SRT in groups of children who stutter (CWS) and their typically developing peers have revealed mixed findings.

One finding is that longer SRT’s correspond with higher receptive vocabulary scores in children who do not stutter, but that for CWS, there appears to be no relationship between vocabulary and SRT. It is proposed that the relationship between vocabulary scores and SRT’s in typically fluent groups is mediated by lexical competition, such that individuals with larger vocabularies take longer to select a word and thus have slower reaction times than children with poorer vocabularies. Anderson and Wagovich hypothesized that factors related to lexical processing, such as phonological working memory, may correlate with SRT’s in children who stutter.  These researchers also investigated whether attentional processes may affect SRT’s in CWS.

In this study, groups of children who do or do not stutter completed nonword repetition, picture naming, and their parents completed a temperament questionnaire about them. These measures were used to evaluate phonological working memory, SRT, and attention, respectively. Results revealed that performance on tests of nonword repletion related to SRT in the children who stuttered but not the typically fluent group. Anderson and Wagovich concluded that for CWS, phonological working memory is associated with SRT.

Nonword repetition is typically considered to be a measure of phonological short-term memory because it involves immediate repetition only, rather than additionally requiring information processing as in working memory tasks. More broadly, nonword repetition taps several processes involved in phonological processing (discrimination, encoding, storage, output). Further work is needed to understand what underlying process may account for the associations between nonword repetition and SRT in the CWS in this study.

Blogger: Alexandra Smith

Grammatical Difficulties in Children with SLI: Is Learning Deficient?

Grammatical Difficulties in Children with SLI: Is Learning Deficient? Hsu, H. J., & Bishop, D. (2011). Hum Dev. 53(5): 264-277.

One view of language learning suggests that children (and humans generally) are sensitive to the patterns of sounds that they hear in language. This statistical learning is based on the frequencies or probabilities of sounds in the language. First, young children store heard sentences in an exemplar-by-exemplar fashion (Ellis, 2002), memorizing individual items. In time, children develop more  abstract (syntactic) patterns of different types, and these frames allow them to produce correct grammatical sentences (Ellis, 2002).

Hsu and Bishop discuss the viewpoint that the grammatical deficits in SLI are due to an impairment in the ability to extract these statistical patterns from the input. As a result, children with SLI may require more exposure to learn abstract syntactic patterns. Indeed, Bishop, Adams, and Rosen (2006) found that even after daily training with a particular construction type, children with SLI did not achieve fluent automatic comprehension. It may be that the poor statistical learning in SLI results from other deficits, however. Limitations in short-term memory (Coady & Evans, 2008) or speech perception (Joanisse et al., 2000) may negatively impact statistical learning abilities in SLI.

Children with SLI may have difficulty identifying patterns in the language they hear. Isolating particular patterns to increase the frequency of exposure and highlight the form are approaches that would be consistent with this viewpoint.

Blogger: Areej Balilah. 

Cognitive Profiling and Preliminary Subtyping in Chinese Developmental Dyslexia

Ho, C. S. H., Chan, D. W. O., Lee, S. H., Tsang, S. M., & Luan, V. H. (2004). Cognition, 91, 43-75.

Even though reading impairment is thought to be just as common among children in China as it is in other countries, most of the pioneering research on dyslexia has been conducted in English with the English alphabet. As a result, the leading theories of dyslexia account for reading impairment in languages like English, but it is unclear whether they apply to languages like Chinese with logographic written symbols. According to one theory called the phonological deficit hypothesis, children with reading impairment may have difficulty mapping (or matching) print and sound due to noisy representations of words in their brains (or underspecified phonological representations).

Ho and colleagues suggest that in Chinese, the nature of the deficits underlying dyslexia might be different. This is because of the nature of Chinese script, as parts of the written symbols in Chinese don’t directly map onto sounds in the word in the same way they do in English. The authors tested 147 Cantonese-speaking children with dyslexia in Hong Kong. Measures included the ability to rapidly name pictures (rapid naming), manipulate sounds in words (phonological processing), read symbols (orthographic processing), and other visual-perceptual skills. Ho and colleagues found that more children had problems with rapid naming (57% of children) and orthographic processing (42%) than with phonological processing (29%) and visual processing (27%). In other analyses involving step-wise regression, results suggested that rapid naming and orthographic processing are related to reading impairment in Chinese. Unlike findings for English speakers, phonological processing was not importantly related to dyslexia in Chinese. These results have implications for the types of intervention strategies that might be used to help children in China improve their reading.

A final note: it is worth mentioning that the phonological processing tasks used in this study were rather simple, and children performed very well on them (even those classified with reading impairment). These tasks asked children to indicate which of two syllables out of a group of three shared sounds. It might be that phonology would have been found to play a larger role in Chinese dyslexia if other tasks had been included. This remains an ongoing debate.

Guest blogger: Jeff Malins from the Language, Reading and Cognitive Neuroscience Lab. Jeff is a PhD student in Neuroscience, and has used measures such as eyetracking, ERPs, and fMRI to study spoken word recognition in adult native speakers of Mandarin Chinese. He is also currently collaborating with researchers in Beijing on a project looking at Mandarin spoken word recognition in children with reading impairments

Guiding Principles and Clinical Applications for Speech-Language Pathology Practice in Early Intervention

Paul, D., & Roth, F.P. (2011). Guiding Principles and Clinical Applications for Speech-Language Pathology Practice in Early Intervention. Language, Speech, and Hearing Services in Schools, 42, 320-330.

This clinical forum note describes four guiding principles in early intervention. Early intervention is defined as services provided to those from birth to 3 years who have, or are at risk for, communication, speech, language, hearing, feeding, swallowing, and/or emergent literacy problems. The services encompass screening/assessment, goal setting/intervention, consultation, service coordination, transition planning, and advocacy. SLPs play a central role in early intervention.

Principle 1 encompasses the idea that services are family centred, and culturally and linguistically responsive. Such services support family involvement in the child’s development, and positive interactions. Principle 2 highlights the need for services to be developmentally supportive and promote children’s participation in their natural environments. Principle 3 refers to services that are comprehensive, coordinated, and team based. Domains of development are interdependent during early childhood requiring comprehensive and coordinated services. Finally, Principle 4 says that services are based on the highest quality evidence available. This evidence may be external and drawn from published research findings or internal including policy, informed clinical opinion, and patient preferences.

There is much to agree with in these principles. Optimal implementation remains a challenge in many settings, but is a good target!

Incidental language learning: Listening (and learning) out of the corner of your ear

Saffran, J. R., Newport, E. L., Aslin, R. N., Tunick, R. A., & Barrueco, S. (1997). Psychological Science, 8, 101-105.

Language learning is incidental; that is, children learn language without being aware that they’re learning it. Humans are sensitive to the patterns they hear in language. Frequency patterns across a broad range of natural experiences are encoded incidentally, and support language learning. One way to study language learning is to teach someone an artificial language, a language they could never have learned before.

One important key to language learning is knowing when one word ends and another begins. There are lots of statistical cues available in the language that can help new language learners figure out where the words are. In particular, there is a relationship between sounds that tend to occur within or between words. For example, some sounds are more likely to occur within a word. If infants are sensitive to these patterns, they may be better at learning language.

In this study, participants listened to a stream of artificial language while focused on something else for 21 minutes. Even though the participants were not told to attend to the language, both adults and children performed above chance level at selecting items from the artificial language. After 42 minutes of exposure, performance was significantly improved and significantly greater than chance.

Despite mere minutes of listening, the complex stimuli, and importantly, the absence of instructions to attend to the artificial language, both adults and children demonstrated statistically-based learning of word boundaries. These findings suggest that incidental learning is a robust phenomenon that may play a role in natural language acquisition. The fact that children and adults did not differ in their learning of words in the artificial language gives support to the idea that early-acquired aspects of language are learned equally well by younger and older learners. Further research is needed to address questions related to the development of this ability.

Nicolette Noonan is an incoming Masters student in the Speech and Language Sciences field.

Children’s Development of Self-Regulation in Speech Production

MacDonald, E., Johnson, E., Forsythe, J., Plante, P. & Munhall, K. (2012). Children’s Development of Self-Regulation in Speech Production, Current Biology, 22, 113-117.

This study examined the use of auditory feedback across development. Auditory feedback is the mechanism by which individuals control the characteristics of their own voice, such as intensity, rate of speech, and the pronunciation of vowels and consonants. It is involved in error correction and is integral to the maintenance of coherent speech: when auditory feedback is absent or lost, changes in vocal characteristics occur.

The researchers compared compensation for formant shifted auditory feedback in 2 and 4 year olds, and adults Participants wore headphones and a microphone, and were seated in front of a computer. In order to get a robot to travel across a playground in a computer game, the participants said /bεd/ (bed). After 20 initial trials during which they listened to their own voice over headphones, the a property of the participant’s vowel known as the first formant (F1) was shifted when played back on their headphones such that they would hear themselves saying /bæd/ (bad).  

MacDonald et al. (2012) found that, on average, children older than 4 years compensate for formant shifted auditory feedback, in effect ‘trying to make the word sound like ‘bed’. Children younger than 4 years of age did not tend to compensate for the manipulation. The researchers also found that speech variability decreased with age, with the youngest children tending to have highly variable word productions. Adults did compensate for the manipulations.

Future research on use of auditory feedback in very young children and toddlers, however, should determine whether motivation plays a role in production at this stage of development.  In this study the robot in the computer program moved forward regardless of whether the child made use of auditory feedback and said the word correctly. Perhaps the toddlers would have compensated if it were necessary for moving the robot forward.

These findings may indicate that children have poor vowel representations at this stage of development. The auditory feedback correction mechanism does not appear to be used for vocal error correction in very young children. Perhaps this age group has not had time to develop sufficiently robust vowel representations due to the rapid growth of vocal structures at this stage. The researchers suggested that perhaps very young children are reliant on their communication partners such as parents, rather than their own internal correction mechanism.

Blogger: Michaela Holmes

Comparison of Conversational-Recasting and Imitative Procedures for Training Grammatical Structures in Children With Specific Language Impairment

Camarata, S.M., Nelson, K.E., & Camarata, M.N. (1994). Comparison of conversational-recasting and imitative procedures for training grammatical structures in children with specific language impairment. Journal of Speech and Hearing Research, 37(6), 1414-1423.

This study compared the effectiveness of two language intervention strategies used to increase correct production of grammatical morphemes and complex sentences in children identified as specifically language impaired. A group of 21 children between the ages of 4:0 and 6:10 who met the criteria for SLI received targeted intervention strategies aimed at either imitation or conversation recasting in 24 intervention sessions over the course of 12 weeks. Each treatment session was divided equally between imitative treatment and recasting treatment, which were each aimed at different intervention targets. The order of the two interventions were counterbalanced across sessions.

Results compared children’s elicited or spontaneous productions in treatment, and in response to clinician input. Use of the conversational recast method resulted in fewer clinician presentations to achieve a greater number of correct spontaneous utterances compared to the imitation model. Additionally, significantly fewer sessions were required for spontaneous production within the conversation treatment versus the imitation treatment. Conversely, the imitative treatment resulted in a greater number of elicited but not spontaneous productions. The authors proposed that the success of the conversation recast treatment model might be due to the easy generalizability of the embedded targets as both treatment and future use of targets take place within actual conversation.

It was recommended that future research on this topic should include increased isolation of the two treatment methods. The results of this study suggest that use conversation recasting as a method of treatment when targeting grammatical morphemes and sentence structure in children identified with specific language impairment.

Blogger: Rosine Salazer

Errorless Leaning of Face-Name Associations in Early Alzheimer's Disease

Clare, L., Wilson, B.A., Breen, K., & Hodges, J.R. (1999). Errorless learning of face-name associations in early Alzheimer’s disease. Neurocase: The Neutral Basis of Cognition, 5(1), 37-46.

Dementia refers to the loss of memory and other mental abilities that are severe enough to interfere with daily life. Many patients with dementia complain of major problems in face-name recognition. There is evidence that people with Alzheimer’s (a common type of dementia) can learn and retain new information by using strategies such as elaboration of material at encoding and expanding rehearsal. Another effective strategy that has been used with brain injured and learning disabled people is errorless learning. Errorless learning aims to prevent or significantly reduce errors made by the learner during the learning process.

This paper reports a multiple baseline intervention single case study of a 72 year old man (VJ) with Alzheimer’s disease and memory problems for the last three years aimed at improving face-naming. The first phase, initial baseline, involved showing VJ Polaroid pictures of 14 club members and testing which names he already knew. The second phase, the intervention, consisted of teaching the remaining face name associations over 21 sessions in his home. Phase three, generalization, took place at the social club and required VJ to identify individuals in person. In phase four, post intervention baseline, sessions were conducted in which VJ’s ability to name the photographs was tested. Finally, phase 5 included a follow-up at 1,3,6, and 9 months.

Errorless learning principles were applied throughout the phases by instructing VJ to provide a name only if he was sure. If he was unsure of a name, prompts were given. Each name-face association was trained in the same way during the intervention phase. VJ was shown a photograph and told the name. Next a mnemonic was chosen that related the name to a physical feature (example: Caroline with the curl on her forehead). Next, vanishing cues were used during the learning trials. Finally, an expanding rehearsal procedure took place.

The results of the study showed a significant increase in the number of faces VJ correctly named, rising from 22% at baseline to 98% following training with improvements maintained after 3, 6, and 9 months. The results provide some evidence that the errorless learning technique may be useful in learning face-name relationships in individuals with Alzheimer’s. However, mnemonic and vanishing cues may also have contributed to the improvement in VJ’s memory.

Blogger: Allison Partridge

The Cognitive and Academic Profiles of Reading and Mathematics Learning Disabilities. Journal of Learning Disabilities

Compton, D.L., Fuchs, L.S., Fuchs, D., Lambert, W., & Hamlett, C. (2012). The cognitive and academic profiles of reading and mathematics learning disabilities. Journal of Learning Disabilities, 45, 79-95.

This study examined the profiles of strengths and weaknesses across cognitive dimensions and across academic domains in children with learning disabilities. Over 600 children completed academic measures of reading comprehension, word reading, applied problems, and calculations, as well as cognitive measures of nonverbal problem solving, processing speed, concept formation, language, and working memory. Data were available at 3 time points for each child allowing modeling of the final learning disability status for each academic measure (at 5th grade) based on cognitive or academic performance.

Results revealed specific rather than general learning profiles in those with a learning disability, but a relatively flat profile for those without a learning disability. Children identified with a learning difficulty in each of the four academic measures tested tended to be low on that measure only (i.e., reading comprehension) and not the other academic tasks (i.e., word reading, applied problems, and calculations). As well, the cognitive profiles of these learning disability groups reflected specific impairments. A learning disability in reading comprehension was associated with low language relative to the other cognitive dimensions measured. Word reading disability was linked to relatively low working memory and language, and applied problems learning disability to relatively low concept formation. A learning disability in calculations was associated with a flat cognitive profile.

Overall results suggest that a specific deficit profile may characterize children with learning disabilities. Nevertheless, caution is warranted in interpreting these findings given the sample size within each profile group.

Blogger: Lisa Archibald

Accelerating Preschoolers’ Early Literacy Development Through Classroom-Based Teacher-Child Storybook Reading and Explicit Print Referencing

Justice, L.M., Kaderavek, J.N., Fan, X., Sofka, A., & Hunt, A. (2009). Accelerating Preschoolers’ Early Literacy Development Through Classroom-Based Teacher-Child Storybook Reading and Explicit Print Referencing. Language, Speech, & Hearing Services in Schools, 40, 67-85.

Like phonological awareness, print knowledge (awareness of words on the page, how to hold a book, direction to read) has been shown to positively influence children’s early literacy achievements; however print knowledge has been given less attention in research and practice. In this study, Justice et al. evaluated a 30 week, 120 book print referencing shared reading intervention implemented by teachers. Effects were measured in terms of changes in print awareness of preschool children. The teachers of 14 classrooms used the print referencing style of storybook reading while 9 comparison control classrooms used their regular style of reading. 106 preschool children were assessed on 3 measures of print knowledge: print concept knowledge, alphabet knowledge, and name writing. Results showed that preschoolers in the intervention group made significantly greater gains in print concepts and alphabet knowledge over control peers.

This article is comprehensive and includes examples of the stories used with their print referencing focus. It offers practical knowledge for any SLP or teacher.

Blogger: Laura Vanderlaan

The Relation Between Music and Phonological Processing in Normal-Reading Children and Children with Dyslexia

Forgeard, M., Schlaug, G., Norton, A., Rosam, C., & Iyengar, U. (2008). The relation between music and phonological processing in normal-reading children and children with dyslexia. Music Perception 25(4), 383-390.

Research has shown that a core deficit of dyslexia is phonological. Other research has found that musical ability is related to reading and phonological ability in both normal-reading children and those with dyslexia. If musical ability is supported by the same underlying process as phonological skill, then it is possible that musical intervention could play a role in remediating the phonological deficits present in children with dyslexia.

Forgeard et al. examined this relationship between music ability and reading ability through a series of four studies. In the first two studies, they compared musical skills to phonological and reading skills among normal readers with and without musical training. Results of the first study confirmed the relationship between phonological processing and pitch processing. Findings from the second study showed that reading ability was related to auditory music skills. They also found that children receiving musical training showed greater improvement in phonemic decoding than children receiving no musical training.

The third study tested the hypothesis that phonological and reading deficits in children with dyslexia should predict deficits in music processing. Results showed that musical discrimination predicted phonological discrimination, which in turn predicted reading ability; however, musical ability did not predict reading ability. The final study compared musical processing abilities of normal-reading children, both with and without musical training, to those of children with dyslexia. The researchers found that normal-readers with musical training performed better than normal readers without musical training, who performed significantly better than children with dyslexia.

This study offers some evidence of the relationship between musical processing and phonological processing. Forgeard et al. suggest that musical intervention may be a helpful addition in intervention programs targeting phonological processing skills. Results should be interpreted with caution due to small sample sizes.

Blogger: Laura Pauls is completing an independent study examining the agreement between standardized language tests and parent and teacher concerns about language development. She is finishing her final year in the Masters of Clinical Science program in Speech Pathology.