About the presenters: Brian D. Humphrey, M.A., CCC-SLP, is a program instructor and clinical supervisor in speech-language pathology at Nova Southeastern University, in Fort Lauderdale, Florida. He studied linguistics and psycholinguistics at the University of Rochester and communication disorders at the University of Minnesota. His current areas of interest are fluency disorders, multicultural issues, and clinical applications of technology.

John Van Borsel, Ph.D in neurolinguistics, is a Professor of Logopedics and neurolinguistics at the Ghent university and an associate editor of Journal of Fluency Disorders. His research interests include speech disorders in general with three lines of research: fluency disorders (developmental and acquired), speech (mainly articulation and phonological disorders) in genetic syndromes and the voice in transsexuals He has publications in: Int. J. of Language and Communication Disorders, J. of Fluency Disorders, J. of Communication Disorders, Folia Phoniatrica et Logopaedica, J. of Voice, Dementia and Geriatric Cognitive Disorders, American J. of Speech Language Pathology, Clinical Linguistics and Phonetics, Child Language Teaching and Therapy

You can post Questions/comments about the following paper to the authors before October 22, 2001.

Word-Final Dysfluencies: Ten Infrequently Asked Questions

by Brian Humphrey and John Van Borsel
from Florida, USA and Belgium

It has been commonly thought that stuttering and related dysfluencies do not occur at the ends of words. However, word-final dysfluencies have been documented in a variety of patients.


1. What is meant by word-final dysfluencies?

Word-final dysfluencies are interruptions in the flow of speech that affect the last part of a word, but do not affect the first sound or sounds of a word. This pattern contrasts with the patterns expected for developmental stuttering (i.e. stuttering that develops before puberty), when speech and language growth is most rapid. People with developmental stuttering are usually dysfluent at the beginning of words or syllables, or they may repeat a whole word.

2. Are there any reports or descriptions of word-final dysfluencies?

There are a number of reports or descriptions of word-final dysfluencies in the literature. Some of these studies deal with word-final dysfluencies in neurogenic stuttering (i.e. stuttering following damage to the nervous system), and some of these studies concern patients with dysfluencies of developmental origin.

Word-Final Dysfluencies in Neurogenic Stuttering

Canter (1971) identified final repetitions and prolongations as features that distinguish neurogenic stuttering from developmental stuttering. Several authors have agreed with Canter’s view (Market et al., 1990, Rosenfield et.al., 1991, Helm-Estabrooks, 1993). However, only a very small number of studies have documented final repetitions and prolongations in cases of neurogenic stuttering. The small number of studies suggests that word-final repetitions may be less common in this population than is sometimes thought. Table 1 summarizes these case reports.

Table 1.

Reported Cases of Word-Final Dysfluencies in Neurogenic Stuttering

Study and Subject Data

Word-Final Dysfluencies


Lebrun & Leleux (1985); Age 60

repetitions of a few final sounds

most repetitions were word-initial; right handed, right brain damage

Ardila & Lopez (1986); Age 50


repetitions in all positions within words; right brain damaged; right handed

Rosenfield et al. (1991); 8 adults

types not specified; heard in all patients

dysfluencies in all positions within words; acquired dysfluency

Bijleveld et al. (1994); Age 65

3 final repetitions

mostly word-initial repetitions, blocks, and prolongations; right handed

Note: None of these studies specified the number of times a sound or part-word was repeated.

Word-final Dysfluencies of Developmental Origin

Word-final dysfluencies, apparently of developmental origin, have been reported in a number of patients. The youngest was 16 months old, and the oldest were developmentally disabled adults. Nearly all of the word-final dysfluencies were reported to be repetitions. Except for two cases, word-final repetitions involved only the final sound of a word. Table 2 summarizes these case reports.

Table 2.

Reported Cases of Word-Final Dysfluencies of Developmental Origin

Study and Patient Data

Word-Final Dysfluencies


Rudmin (1984) Onset: 16 months

on occasion, 1-2 repetitions of /p, t, k/

more frequent when fatigued; gone by age 30 months; no other dysfluencies

Mowrer (1987) Onset: 34 months


usually 2-3 repetitions of /s, t/; occasionally, up to 7 repetitions

over 50% of dysfluencies were final consonants; accompanied by head movements; gradual disappearance; stuttering also present

Camarata (1989 Onset: 25 months

repetitions of /p, t, k/

no other dysfluencies; phonological process to contrast /p,t,k/ vs. /b,d,g/?

Lebrun & Van Borsel (1990); 8 year old boy

repetitions of plosives, fricatives, & /r/

mainly when reading; word-final dysfluencies on 8.25% of words; no neurological symptoms

Lebrun & Van Borsel (1990); Age 17

repetitions of plosives and fricatives

15.5% dysfluent; 19% of dysfluencies were word-final; diagnosed with Down’s syndrome

Stansfield (1995)

2 adults
1 adult
1 adult:


repetitions repetition, hesitation repetitions, prolongations, blocks

repetitions were mainly plosives; typical stuttering was also present in all 4 subjects; no more than 33% of dysfluencies were word-final; all subjects had significant learning difficulties and attended adult training centers

Van Borsel et al. (1996
9-year-old boy





1 repetition; usually final parts of words: a syllable, or a vowel and following consonants
"stap ap", "noemt oemt"
gehoord hoord"
diamanten ten"

final part-word repetitions predominated in monologue; other dysfluencies were present; no prolongations or blocks; no tension or struggle; speaker of Dutch; 2 small subcortical lesions were not clearly correlated with dysfluencies

Humphrey (1997)
Humphrey (1997, May)
12-year-old boy





1 repetition; final parts of words; usually a vowel and following consonants, or a vowel and the following syllable:
"night ight", "above ove",
"trolley olley",
"daylight aylight"

final part-word repetitions predominated in monologue; other dysfluencies present; no typical stuttering, no prolongations or blocks; no tension or struggle; speaker of English; diagnosed with attention deficit disorder

Note: Plosives are speech sounds that stop the flow of air, like p, b, t, d, k, g. Fricatives are speech sounds that have a rush of air, like f, v, s, z, sh, th. Prolongations are dysfluencies that stretch a speech sound. Blocks are complete interruptions of speech.

3. Are there different types of word-final dysfluencies?

Word-final repetitions, prolongations, and blocks have been reported, as listed in question number two. Word-final repetitions have been the most frequently reported and discussed.

Word-final repetitions are part-word repetitions that follow a complete production of a word. These word fragments include the ending of the word but not the beginning of the word. Some word-final repetitions may remind the listener of an echo. Two different types of word-final repetitions are apparent. Most of the studies discussed in question number two reported word-final repetition of final sounds or consonant clusters only. Usually these dysfluencies involved plosive sounds; less often, word-final repetitions of fricatives and /r/ were reported.

A second pattern can be seen in the cases reported by Van Borsel et al. (1996) and Humphrey (1997; 1997, May). These strikingly similar cases displayed single repetitions of larger units, i.e. a syllable ("gehoord hoord"), a vowel and its following consonants ("noemt oemt", "light ight", or a vowel and its following syllable ("trolley olley").

Word-final prolongations and blocks have been observed in two adults with learning problems (Stansfield, 1995). For one subject, they accounted for 24.2% of the total dysfluencies. For the other subject, they were less than 1% of the total dysfluencies. The second subject also had word-final repetitions.


4. Are word-final dysfluencies unusual?

Word-final dysfluencies are considered unusual, except perhaps in cases of palilalia (see question number six). In reports of developmental stuttering, neurogenic stuttering, and other atypical stuttering, word-final dysfluencies are seldom mentioned.

Word-final dysfluencies may be under-reported at present. In pediatric cases that report outcomes, word final dysfluencies have been described as transitory or easily treated (Camarata, 1989; Mowrer, 1987; Rudmin, 1984). Because there seems to be little association with tension or struggle, word-final dysfluencies may often go unnoticed or unreported. If other types of dysfluency are present and heard as disruptive to speech, word-final dysfluencies may escape notice. We may not be in the habit of listening for word-final dysfluencies, so we may tend not to hear them.

5. What accounts for word-final dysfluencies?

The answer is unclear. Some features of word-final dysfluencies differ from the features of developmental stuttering, and may be governed by different factors. Subjects usually have not evidenced tension or struggle. Secondary behaviors are mentioned only by Mowrer (1987); his subject showed head movements. Because word-final dysfluencies have occurred independently in some cases, and have co-occurred with typical stuttering in other cases, it is possible that they are separate but related phenomena. The dysfluencies of the neurogenic stuttering patients have been attributed to their brain damage. Given the small number of reports, it is remarkable that word-final dysfluencies are noted in both neurogenic and developmental cases.

Several explanations have been offered for word-final dysfluencies of developmental origin. Rudmin (1984) theorized that his daughter had a motor programming difficulty. Mowrer (1987) suggests two factors that may have affected his patient: the mother’s attempts to correct his pronunciation of final consonants may have created linguistic demands beyond his abilities. Additionally, some traumatic events may have triggered the onset of the dysfluencies.

Camarata (1989) viewed his patient’s final consonant repetitions as part of a phonological error process. This 25 month old boy devoiced /b/, /d/, and /g/ at the ends of words: he replaced /b/ with /p/, /d/ with /t/, and /g/ with /k/. Dysfluent repetitions of these errors never occurred. However, final repetitions occurred on words that correctly ended with /p/, /t/, and /k/. The final consonant repetitions were taken to be a strategy for marking a difference between the boy’s correct /p/, /t/, and /k/ endings and the /b/, /d/,and /g/ endings that he produced incorrectly.

The two patients of Lebrun and Van Borsel (1990) displayed dysfluencies consistent with developmental stuttering, along with word-final dysfluencies. For these patients, the word-final dysfluencies were considered to be a type of stuttering. Stansfield (1995) suggested that the four learning-challenged adults in her study may have developed word-final dysfluencies if the demands of two or more coexisting communication disorders exceeded the patients’ capacities to respond.

Although the boy studied by Van Borsel et al. (1996) had two small focal brain lesions, a link between his brain damage and his dysfluency was not clear; developmental origin was considered possible. Because he tended to repeat the final syllable of a word, or a vowel and the following consonants, the authors suggested that his word-final dysfluencies could be a form of palilalia (see question number six). Because palilalia is sometimes mentioned as a feature of Tourette’s syndrome and of Parkinsonian disorders, it is intriguing that several researchers have pointed out correspondences between stuttering and movement disorders (Abwender et al., 1998; Bagheri et al., 1996; Molt, 1999; Palumbo et al., 1997).

Humphrey (1997, May) suggested that the final part-word repetitions produced by his patient may have had phonological and grammatical influences. This boy’s word-final dysfluencies displayed a previously unreported characteristic. Frequently a part-word repetition, separated by a pause from its "parent word", crossed a phrase boundary and appeared to function as a starter for the next phrase or sentence -- for example: "... in the daylight. [pause] aylight But once at night, [pause] ight when we were up on a trolley,[pause] olley high above, [pause] ove the ground, ..."

6. Are word-final dysfluencies and palilalia similar phenomena?

Most of the word-final dysfluencies discussed here are clearly distinct from the dysfluencies of patients with palilalia. The term palilalia is derived from Greek, "pali", meaning "again", and "lali" meaning "speech". It is used to refer to a set of speech patterns heard in adult neurogenic disorders, especially Parkinsonian disorders.

In adults, palilalia is most often seen in patients with post encephalitic parkinsonism and with pseudobulbar palsy (Brain, 1961), but it has also been observed in Alzheimer’s disease and mult-infarct dementia (Helm, 1979). Palilalia in adult neurogenic disorders has usually been described as multiple repetitions of a word or phrase, often with increasing rate and decreasing clarity or volume, often at the ends of words or utterances. Boller et al. (1973) provide a typical definition. Benke and Butterworth (2001) have presented evidence that palilalia can take a second form that is marked by repetition of utterances at a constant rate, alternating with silent intervals.

To some extent, palilalia resembles word-final repetitions because the final part of an utterance is often, but not always, affected (Van Borsel, in press). However, the repetitions in palilalia usually involve longer fragments, i.e. words and even whole phrases; and the number of times a fragment is repeated is often considerable, compared to the word-final repetitions outlined above. In some cases of palilalia, a patient may repeat a fragment more than 50 times in succession (LaPointe & Horner, 1981; Lebrun et al., 1987).

Yet, the distinction between word-final repetitions and palilalia is not always very sharp, as the cases reported by Van Borsel et al. (1996) and Humphrey (1997) show. These two patients displayed patterns that appeared related to patterns seen in palilalic patients. For comparison, here are some examples of their repetitions, together with examples of palilalic speech:

Final Part-Word Repetitions:

"gehoord, hoord", "diamanten ten", "stap ap", "noemt oemt" (Van Borsel et. al, 1996)

"First, irst you have ... have to take, get the bulb. First you have to take the wires, ires like so ... [describing how to put a table lamp together] (Humphrey, 1997):


"..die beine in die luftluftluftluftluftluft..." [translation: legs in the air] (Benke and Butterworth, 2001)

"... tua i allaweil steckenbleiben ... steckenbleiben ... ja, steckenbleiben ... tua i steckenbleiben ... steckenbleiben ... allaweil steckenbleiben ..." [translation: I always get stuck] (Benke and Butterworth, 2001)

7. Are word-final dysfluencies a type of stuttering?

The World Health Organization (1977) defines stuttering as disorders in the rhythm of speech in which the individual knows precisely what he/she wishes to say but at the time is unable to say it because of an involuntary repetition, prolongation, or cessation of a sound. In this sense, word-final dysfluencies would not be stuttering. Word-final dysfluencies occur after the individual has said what he/she wishes to say. They do not fit the patterns usually found in developmental stuttering, but they appear to be related. In some of the case reports, there is sufficient evidence to suggest that some subjects demonstrated developmental stuttering.

8. Are people with word-final dysfluencies usually aware of them?

More information is needed about this question. Some subjects were reported to be unaware of their word-final dysfluencies. For others, there is no clear information. The boy studied by Mowrer (1987) appeared to be aware, and demonstrated some secondary behaviors.

9. Can and should word-final dysfluencies be treated?

Given the varieties of word-final repetition and our limited information, no general statement is possible, but some guiding questions may be useful. Helm-Estabrooks (1993) has suggested a similar decision process for cases of neurogenic stuttering.

Is there a significant disruption of communication? If word-final dysfluencies are readily noticed, or distracting to others, a decision to treat may be appropriate.

Is the dysfluent person motivated to change? If not, the probability of success may be low.

Is there evidence of an underlying neurological condition? Have symptoms been worsening? Have the dysfluencies appeared relatively suddenly in someone older than five or six years? If so, a referral to a neurologist may be in order, and the treatment decision may depend on the findings.

How does the problem of final-word dysfluency rank with medical, communication, learning, or personal problems that the dysfluent person may have? If word-final dysfluencies are a low priority, a decision not to treat may be appropriate.

If a decision to treat is reached, the following questions may help to frame a treatment strategy:

Have diagnostic procedures sufficiently documented the behavior and its components? Analysis of the structure of word-final dysfluencies, as well as the contexts and conditions in which they occur, may be useful in developing treatment strategies. For example, could an effective language based intervention be designed for someone whose word-final dysfluencies appear to be language based? Would someone who shows little awareness of word final dysfluencies show gains when awareness and self-monitoring is addressed in treatment?

Have treatment strategies been developed to address some features of the patient’s dysfluent speech? For example, if a rapid speech rate is observed, does reducing speech rate decrease word-final dysfluencies?

10. In future case studies, what issues could be addressed?

Additional case studies are needed to document and compare the final dysfluencies of neurogenic stutterers, developmental stutterers, patients with Tourette’s syndrome, and palilalic patients. Because have very little information about treatment of word-final dysfluencies, treatment studies and reports could be valuable.

To improve documentation and comparison, detailed communication sample analysis is needed. In some cases, linguistic analysis as well as fluency analysis may be needed to shed light on the patterns observed. Better documentation of subjects’ awareness of their own dysfluencies is needed.

To illustrate the value of in-depth comparison, Table 3 compares the dysfluencies produced by T. (Van Borsel et al., 1996), and B.B. (Humphrey, 1997; 1997, May). Rank orders of occurrence were derived from percentages of dysfluency for each dysfluency type. For both patients, word-final dysfluencies predominated in monologue.

Table 3.

Comparison of Dysfluency Rank Orders Across Clients and Conditions

Dysfluency Type:


Monologue 1




Monologue 2






Average Rank:









Monologue During Masking





Average Rank:

Final Part Word Repetitions








Word Repetitions








Broken Words
















Phrase Repetitions








Initial Part Word Repetitions


Not observed




Not observed


Revisions were not ranked.

Comparison of data across subjects offers a surprising finding. Overall, these two patients produced the same types of dysfluency in similar rank order. The results suggest strongly that these two patients should be grouped together. For such similar patients, a search for common underlying factors may be rewarding.


Abwender, D. A., Trinidad, K. S., Jones, K .R., Como, P .G., Hymes, E., & Kurlan, R. (1998). Features resembling Tourette syndrome in developmental stutterers. Brain and Language 62, 455- 464.

Ardila, A., & Lopez M. V. (1986). Severe stuttering associated with right hemisphere lesion. Brain and Language 27, 239- 246.

Bagheri, M. M., Kerbeshian, J., & Burd, L. (1999). Recognition and management of Tourette’s syndrome and tic disorders. American Family Physician, April 15. Retrieved August 22, 2001 from http://www.aafp.org/afp/990415ap/2263.html

Benke, T., & Butterworth, B. (2001). Palilalia and repetetive speech: Two case studies. Brain and Language 78, 62- 81.

Bijleveld, H., Lebrun, Y., & Van Dongen, H. (1994). A case of acquired stuttering. Folia Phoniatrica et Logopaedica 46, 250- 253.

Boller, F., Boller, M., Denes, G., Timberlake, W.H., Zieper, I., & Albert, M., 1973. Familial palilalia. Neurology 23, 1117- 1125.

Brain, W. R., 1961. Speech disorders: Aphasia, apraxia and agnosia. Washington, D.C.: Butterworths.

Camarata, S. M. (1989). Final consonant repetition: a linguistic perspective. Journal of Speech and Hearing Disorders 54, 159- 162.

Canter, G. J. (1971). Observations on neurogenic stuttering: A contribution to differential diagnosis. British Journal of Disorders of Communication 6, 139- 143.

Helm, N. (1979). Management of palilalia with a pacing board. Journal of Speech and Hearing Disorders 44, 350- 353.

Helm-Estabrooks, N. (1993). Chapter 10: Stuttering associated with acquired neurological disorders. In Curlee, R.F. (ed.), Stuttering and related disorders of fluency, 205- 219. New York: Thieme.

Humphrey, B. D. (1997). Unusual dysfluency: Repetitions in final position in an adolescent boy: A poster publication. Florida Journal of Communication Disorders 17, 41- 42.

Humphrey, B. D. (1997, May). Unusual dysfluency: Repetitions in final position in an adolescent boy. Poster session handout presented at the annual convention of the Florida Association of Speech-Language Pathologists and Audiologists, Orlando, FL.

LaPointe, L. L., & Horner, J. (1981). Palilalia: A descriptive study of pathological reiterative utterances. Journal of Speech and Hearing Disorders 46, 34-38.

Lebrun, Y., Leleux, C., & Retif, J. (1987). Neurogenic stuttering. Acta Neurochirurgica (Wien) 85, 103-109.

Lebrun, Y., & Leleux, C. (1985). Acquired stuttering following right-brain damage in dextrals. Journal of Fluency Disorders 10, 137- 141.

Lebrun, Y, & Van Borsel, J. (1990). Final sound repetitions. Journal of Fluency Disorders 15, 107- 113.

Molt, L. (1999). The basal ganglia’s possible role in stuttering: An examination of similarities between stuttering, Tourette syndrome, dystonia, and other neurological-based disorders of movement. Paper presented at the International Stuttering Awareness Day Online Conference, 1999. Retrieved August 27, 2001 from http://www.mnsu.edu/comdis/isad2/papers/molt2.html

Market, K. E., Montague, J. C., Buffalo, M. D., & Drummond, S.S. (1990). Acquired stuttering: Descriptive data and treatment outcome. Journal of Fluency Disorders 15, 21- 33.

Mowrer, D E . (1987). Repetition of final consonants in the speech of a young child. Journal of Speech and Hearing Disorders 52, 174- 178.

Palumbo, D., Maughan, A., & Kurlan R. (1997). Hypothesis III: Tourette syndrome is only one of several causes of a developmental basal ganglia syndrome. Archives of Neurology 54, 475- 483.

Rosenfield, D.B., Viswanath, N.S., Callis-Landrum, L., Didanato, R., & Nudelman, H.B. (1991). Patients with acquired dysfluencies: What they tell us about developmental stuttering. In Peters, H.F.M., Hulstijn, W. Starkweather, C.W. (eds.), Speech motor control and stuttering. Amsterdam: Elsevier.

Rudmin, F., (1984). Parent’s report of stress and articulation oscillation in a pre-schooler’s disfluencies. Journal of Fluency Disorders 9, 85- 87.

Stansfield, J., (1995). Word-final disfluencies in adults with learning difficulties. Journal of Fluency Disorders 20, 1- 10.

Van Borsel, J., Van Coster, R., & Van Lierde, K. (1996). Repetitions in final position in a nine-year-old boy with focal brain damage. Journal of Fluency Disorders 21, 137- 146.

Van Borsel, J., Schelpe, L., Santens, P., DeVos, N., & DeVos, C. (2001, in press). Linguistic features in palilalia: Two case studies. Clinical Linguistics and Phonetics 13.

World Health Organization (1977). Manual of the international statistical classification of diseases, injuries and causes of death (vol 1). Geneva: author.

You can post Questions/comments about the above paper to the authors before October 22, 2001.

September 14, 2001