About the presenter: Brian Humphrey, a Board-Recognized Fluency Specialist, is an instructor and clinical supervisor in Communication Sciences and Disorders at Nova Southeastern University in Ft. Lauderdale, Florida. He teaches a graduate course in fluency disorders, and he has published and presented research about bilingual stuttering and atypical fluency disorders. In his "other life", he plays several musical instruments in a band, and he composes tunes for traditional dancing. For several months in 2005-2006, he experienced neurogenic stuttering.
About the presenter: Rosalee has treated people who stutter for over 30 years. Founder and Executive Director of the Montreal Fluency Centre, she is also Adjunct Professor at McGill University, School of Communication Sciences and Disorder. As a charter member of the Lidcombe Program Training Consortium she coordinates clinical training for this program in North America and has provided presentations and workshops to more than 3000 clinicians. Rosalee has published in peer reviewed journals and contributed chapters on Stuttering to various textbooks. She is interested in evaluation of clinical outcomes, bilingualism and stuttering and long distance biking.

You can post Questions/comments about the following paper to the author before October 22, 2008.

Stuttering and Bilingualism in Children and Adults

by Brian Humphrey (Florida, USA) and Rosalee Shenker (Montreal, Canada

Many speech-language clinicians feel challenged when called upon to evaluate or treat a person who stutters. Bilingual people may be affected by stuttering in one or more of their languages, and stuttering in a bilingual person may present additional challenges. We have independently studied bilingual children and bilingual adults who stutter. When we first learned of each other's work, we were pleased to note the complementary nature of our projects. Here is a summary of what we each have learned.

Stuttering and Bilingualism in Children: Rosalee C. Shenker

I first became interested in issues related to stuttering in bilinguals when presented with the standard question that parents ask: "if my bilingual child is stuttering should I eliminate a second language?" Providing an answer was not so straightforward. The question led me to the literature, so that I could give the parent an informed response. At the time, there were virtually no clinical data to suggest that the answer was a simple 'yes' or 'no'. Bilingualism is a continuum. Competence in each language varies across many modalities, including auditory and written comprehension, and verbal and written expression. Level of proficiency is defined in relation to each modality.

Since I work in a city where most children speak at least two languages, with a large multicultural population that speaks a home language that is often not one of the two languages that I speak, the answer became even more complex. An important part of the dilemma for me was: if we were to eliminate one language, which language should be eliminated - the mother's or the father's? Additionally, would this be possible to do? These questions led me to my current line of clinical exploration. To answer a question you often have to ask many. The questions that I am asking currently focus on two major themes:

A. Can stuttering be treated simultaneously in both/all languages spoken?

B. Will bilingualism affect treatment outcomes or treatment durations in a pediatric population?

Although definitions of bilingualism have varied, two types of early childhood bilingualism have been identified: simultaneous bilingualism and second-language learning. Individuals introduced to both languages from birth have been described as "simultaneous bilinguals", and those introduced to a second language at the age of four or later have been described as second-language learners.

I work with a pediatric population. My experience with treating bilingual children is best described by presenting a case study: the child's stuttering was treated using the Lidcombe Program in both languages that the child spoke. The Lidcombe Program makes use of verbal feedback to the child provided by the parent both in and out of the clinic.

This child was five years old at treatment onset. He was considered to be a "simultaneous bilingual", with both English and French spoken at home since birth; e.g. his mother always speaks in French and his father always speaks in English. Upon assessment, no other speech and language issues were of concern, but there was a family history of recovery from stuttering (mother and older sister). Stuttering was first noted two years prior to the assessment, and no previous treatment was provided.

This child appeared to be aware, but unconcerned with his stuttering, although frustration with abandoned utterances had developed: after starting a sentence, he sometimes ended with "I can't say it". Prior to treatment, his parents had been telling him to "slow down", and to "take a deep breath" when stuttering was noted in spontaneous conversation.

Stuttering was characterized by the referring Speech Language Pathologist as severe, with a high frequency of prolongations, blocks and initial repetitions of sounds and syllables. Some secondary behaviors were increased volume and loss of eye contact, turning of his head and neck to release a word, and rapid and audible breathing.

Stuttering was noted in both English and French. Pre-treatment, in a spontaneous conversational sample in English, 4.8% of the syllables were stuttered; and in French, 3.2% of the syllables were stuttered. In beyond-clinic samples, 12% of syllables were stuttered in English; and in French, 8.9% of the syllables were stuttered.

Severity ratings were assigned for the samples given by the parents. On the rating scale, 1 indicated no stuttering, 2 indicated mild stuttering, and 10 indicated extremely severe stuttering. The severity rating was 3 for the in-clinic samples and 7 for the beyond clinic samples. English was noted to be the more proficient language. Both English and French were heard at home and at school, but French predominated at school.

Treatment was initiated with the Lidcombe Program for Early Stuttering, a behavioral program in which parents are taught to provide verbal feedback to children, with an emphasis on praise for stutter-free speech. Feedback is provided initially in structured conversations and, as fluency increases, in unstructured conversations through out the day. Since both parents wished to participate, the treatment was provided in both English and French.

Initially, verbal contingencies for stutter-free speech were provided in structured speaking conversations, with parents alternating treatment from day to day. After three sessions, stuttering had reduced by more than 50% in both languages. The parents reported some spontaneous self-correction, less effort and more whole-word repetitions and phrase repetitions.

This child met the criteria for Stage Two of the Lidcombe Program in eight clinic visits over 12 weeks. He had achieved a criterion of less than one percent stuttered syllables, and daily severity ratings were mostly '1' , with an occasional '2'.

Stage Two had to be adjusted, since this child failed to meet the criteria for Stage Two at the first visit. The parents resumed weekly clinic visits, daily structured conversations and a vigilant home program and the Stage Two criteria were met four weeks later, after a total of 13 clinic visits over 17 weeks. From this point the criteria for Stage Two was sustained at less than 1% syllables stuttered; and severity ratings were mostly '1' with an occasional '2'. The child was discharged from Stage Two after 12 months with 0% stuttered syllables and a severity rating of 1. At this writing, he has been followed for four years, and he has remained stutter-free and fully bilingual, with a third language added in school. His parents enjoyed being part of the solution for their son's stuttering, and they welcomed the opportunities to actively participate in the treatment. Both parents commented on how relieved they were to be able to continue speaking their native languages.

Is this case representative of the clinical outcomes of early treatment of bilingual children? Our clinic recently performed a chart audit to evaluate treatment time using the Lidcombe Program for a group of 56 children. One of the questions asked was: "would bilingualism affect treatment times or treatment outcomes?" For this project, simultaneously bilingual children were considered to be bilingual, and children who spoke two or more languages were considered to be multilingual. Both were placed in one category for the purpose of this audit.

Of the 56 children whose files were audited, 26 were preschool age, and 30 were school age. Ages ranged from 2.9 to 11.9 years. There were 48 males and 8 females: 23 were bilingual or multilingual, and 33 were uni-lingual.

The uni-lingual group had a shorter median interval between onset of stuttering and treatment; on average they had been stuttering for 16 months prior to treatment. In comparison, the children in the multi-lingual group had been stuttering longer, for an average of (27 months). Little difference was noted in the median percentage of stuttered syllables at last treatment. For the uni-lingual children, the median was 0.9%, and for the multilingual children, it was 0.5%. The median number of weeks required to complete Phase 1 of the Lidcombe Program was essentially the same for the uni-lingual group (14.29) and the multilingual group (15.0). These findings show that language proficiency (bilingualism or multilingualism) did not have a significant effect on treatment outcomes for preschool children and school-age children who stutter.

Conclusions: Our clinical experience to date suggests that bilingualism does not affect clinical outcome for treatment of stuttering in young children. Nevertheless there is still too little research to answer 'yes' or 'no'. The research to date on bilingual stuttering does not provide compelling evidence to justify asking a child to 'become monolingual', especially when those around him speak or are spoken to in more than one language, as part of the daily routine.

Stuttering and Bilingualism in Adults: Brian D. Humphrey

My interest in bilingual stuttering came about because of an interesting clinical case. Although the subjects of my study were three adults and a teen, my first bilingual client who stuttered was a five-year-old boy. Both of his parents were bilingual in English and Spanish, and they asked for their son's treatment to be provided in English. As the boy's fluency in English improved, the parents reported improvement in Spanish as well. Very soon, his improved fluency in both languages led to dismissal from treatment.

I began to wonder how often stuttering treatment in one language would be likely to transfer to another language. A search of the literature turned up little information, so I set out to measure the effects of monolingual fluency treatment on two languages spoken by a person who stutters.

The study was designed focus on the percentages of reduction in dysfluency achieved by the subjects. It was expected that baseline levels of dysfluency and final levels of dysfluency would vary among the participants.

Over time, I obtained data for four adult and teenage subjects. As each subject was seen for treatment, fluency gains were monitored in a consistent manner for all subjects, and treatment approaches could be tailored to each subject's needs. Bilingual graduate students in speech-language pathology analyzed the conversational samples in English - the language of treatment, and in Spanish or French - the languages that were monitored.

For each subject, improvement of fluency in English was accompanied by clear improvement of fluency in the language that was monitored.

Across the four subjects, the percentage of syllables with stuttering dysfluencies in English was reduced by an average of 67.5%. In Spanish or French, the untreated languages, the percentage of syllables with stuttering dysfluencies was reduced by an average of 59.8%

These data indicate that fluency gains from treatment in one language can transfer to an untreated language, and that monolingual clinicians who treat bilingual adults who stutter may be encouraged to look for a decrease of stuttering in the untreated language.

A second study was prompted by a review article, written by Dr. Patrick Finn and Dr. Ann Cordes. They asserted that there had been no information about how well speech-language pathologists may be able to judge disfluencies in a language that they do not speak. The question intrigued me, so I designed a second study to find out whether, when listening to a communication sample in Spanish, bilingual speakers of English and Spanish may be better at making judgments about disfluencies than monolingual English-speaking judges.

The subjects for my second study were monolingual and bilingual graduate students in speech-language pathology who had taken a graduate-level course in fluency disorders.

First, they were familiarized with the criteria they were to use for judging disfluencies. Then, they were asked to watch and listen to two video recordings of the same person who stutters one recording was in English, and one was in Spanish. Previously, I had determined that the speaker in the video clips stuttered more in Spanish than in English. The graduate students serving as judges were asked to press a button whenever they thought that the speaker in the videos was disfluent. For the purposes of this study, stuttering dysfluencies and non-stuttering disfluencies were not scored separately.

When I analyzed the data, the bilingual judges and the monolingual judges both considered the video clip in Spanish to contain a greater percentage of disfluencies than the English-language video recording. The differences in performance between the two groups of judges were not statistically significant.

The results of the second study suggest that monolingual English-speaking clinicians who are familiar with identifying stuttering may successfully do so in an unfamiliar language, if the unfamiliar language is at least moderately related to English.

We may think about the issue of stuttering and bilingualism as a puzzle whose construction is in progress. Completion of this puzzle is a group effort, with participants from around the globe.

You can post Questions/comments about the above paper to the authors before October 22, 2008.

submitted DATE
Return to the opening page of the conference