2003 IFA Congress: Montreal, Canada

The Effects of Cognitive Behavioral Therapy with People Who Stutter

Evelyn R. Klein and J. Amster
La Salle University Philadelphia, PA


This study investigated the effects of cognitive behavioral therapy with People Who Stutter (PWS). Eight adults ranging in age from 27 to 56 years comprised the sample of PWS. Measures of perfectionism, dysfunctional thoughts, and attitudes as well as stuttering severity were taken at baseline, 3 weeks, and 6 weeks after treatment. Use of cognitive behavioral treatment and stuttering modification revealed improvement in PWS’ ability to cope, accept, and reduce perceived negative effects of stuttering in addition to stuttering severity.

  1. Introduction
The relationship of attitude change and long-term outcomes has long been of interest in the treatment of People Who Stutter (PWS) (Andrews & Cutler, 1974; Guitar & Bass, 1978). Research indicates that PWS often remain vulnerable to relapse (Neilson & Andrews, 1992) with fluency failures following a predictable course. Although stuttering therapy can be helpful, it is essential that PWS reconsider their beliefs about themselves, communication, and their functioning in the world. Cognitive behavioral therapy techniques deal with automatic thoughts, underlying assumptions, and cognitive distortions that can have negative effects on an individual’s self-esteem. Starkweather (1984) theorizes that stuttering has two major components, feelings and overt behavior, and both of these components must be treated. He posits that if feelings alone are treated the client may relapse because the original overt behaviors remain, although reduced in severity. If only overt stuttering behaviors are treated, the fear of being disfluent likely remains and old reactions of struggle and avoidance may lead to a breakdown in treatment effects. However, as noted by Ryan (2003), cognitive-affective-emotive procedures do not yet have published efficacy research and generally do not provide replicable treatment strategies. Therefore, the purpose of this study was to determine the effects of cognitive behavioral therapy (CBT) followed by stuttering modification techniques with PWS. Changes in underlying thoughts, perfectionist tendencies, social anxiety, communication attitudes, and stuttering frequency, duration, and physical concomitants were measured before, during, and after treatment.

  1. Method

The sample consisted of 5 male and 3 female adults ranging from 27 to 56 years of age. The mean age was 44 years (standard deviation = 9.9 years). The racial composition of the participants was seven Caucasian and one Asian. The educational backgrounds of these individuals consisted of two people with high school diplomas, three with bachelor’s degrees, and three with graduate degrees (two master’s level and one doctoral level). All participants were employed at the time of the study. Participants volunteered to be part of this study after the opportunity was presented to them at their local National Stuttering Association (NSA) meetings and on their local NSA website. Seven of the eight participants had a history of previous speech therapy. Two of the eight had received previous psychotherapy. None of the participants had received cognitive behavioral therapy prior to this study. Treatment was free of charge. At baseline, all eight participants reported continually trying to hide the fact that they stuttered.


The study took place over seven weeks from May-July, 2003. The first visit involved obtaining informed consent, providing an overview of the project, and obtaining a baseline evaluation. Therapeutic intervention consisted of six individual and six group sessions. The authors of this article, a licensed/certified speech-language pathologist as well as licensed psychologist and a licensed/certified speech-language pathologist holding specialty recognition in fluency disorders, provided all evaluation and treatment sessions. Treatment sessions were weekly with each individual session lasting one hour and group sessions lasting 1.5 hours.

During the first meeting, baseline measures were taken. These included: a comprehensive client interview with personal history, Stuttering Severity Instrument-3 (SSI-3) (Riley, 1994), Amster Adaptation of the Burns Perfectionism Scale (Amster, 1995), Liebowitz Social Anxiety Scale (Sajatovic & Ramirez, 2001), Locus of Control Behavior Scale (Manning, 2001; Rotter, 1966), and the Modified Erikson Scale of Communication Attitudes (Erikson, 1991). The first three weeks of individual and group therapy focused on cognitive behavioral therapy (CBT). Weeks four through six introduced stuttering modification with a review of CBT concepts. Assessment measures were administered after the third week of treatment (following CBT treatment) and at the end of week six (following stuttering modification treatment).

The following account provides an overview of individual and group treatment sessions.

Week One

Individual Session: The first task was for participants to comment on the evaluation session and to develop a list of their three primary goals. All eight participants listed goals related to modifying their speech and the way they felt about it. Both the males and females in the study wanted to be able to speak fluently but if that did not happen they stated that they did not want stuttering to limit their lives in the way it had previously. Statements such as, “I want to be able to speak without being worried that I made a fool of myself” and “I want to stop trying to hide the fact that I stutter” were common. These goals formed the basic concepts and self-explorations influencing individualized treatment. Participants also explored their feelings after they told someone new that they stuttered. During this session, four of the eight common cognitive distortions were introduced as errors in logic (Freeman et al, 1990). These included: (1) all or nothing thinking, (2) not seeing the whole picture, (3) focusing on the negative, and (4) mind reading. Participants were taught how to use the automatic thought record to write about a situation, accompanying automatic thoughts, associated emotions, distortions in thinking, and an alternative response for a possible change in thought.

Group Session: All participants met as a group for the first time. The video, “I think, they think...” (Rapee, 1998) about social anxiety was shown to the group. The two professional group leaders asked questions to encourage discussion on how the people in the video were similar to or different from people who stutter. Participants anonymously wrote their own personal governing rules or “˜core beliefs’ on a card. These became the topics for group discussion. Group members explored changes in core beliefs that challenge irrational thoughts. One of the primary beliefs that pervaded group participants from the time most were young children was the strong message they received to always do their very best and to be as perfect as possible.

Week Two

Individual Session: During this session, four additional cognitive distortions were introduced and explored. These included: fortune telling, catastrophizing, should’ve-c0uld’ve-would’ve, and taking it personally. The automatic thought records from the previous week were reviewed. This process was informative for participants as it was the first time for most to think about the way in which they thought, what words they said to themselves when situations occurred, and how these thoughts impacted their day. Remarkably, participants began to relate their personal difficulties to their cognitive distortions. At this time, participants also began systematic desensitization training using guided imagery for reducing stress in a variety of situations including speaking.

Group Session: During week two, the group explored a model of anxiety (Freeman, et al., 1990). They related actual situations to their own perceptions based on core beliefs. They learned how anticipated notions of danger, challenge, or safety relate to feelings of power or powerlessness and how these feelings may affect their speech and ability to cope. A primary focus was on unconditional self-acceptance.

Week Three

Individual Session: Each participant explored cognitive distortions related to specific situations using the automatic thought record from the previous week. Socratic questioning as a technique helped increase participants’ awareness of distorted thinking and its effects on speech. Participants also learned to re-rate their automatic thoughts and subsequent emotions. Once again systematic desensitization was used to help modify an anticipated scenario given new thoughts.

Group Session: Team-work involved reading various scenarios and writing negative, positive, and neutral responses to them. This eye-opening activity helped participants reflect differently on one situation. Group discussions helped increase self-awareness and identify thought patterns in need of change. Participants named core beliefs leading to negative thoughts. The concept of “acting-as-if”  was taught. Participants explored how they would act if they believed they were capable of speaking in different situations.

Week Four

Individual Session: This week began work on stuttering modification although CBT concepts were continued with analysis of the automatic thought record. Speech production was a new main focus. Individual stuttering was explored with a review of each person’s own sound production, breathing, and vocal and articulatory tensions. Participants also analyzed forward vs. backward moving speech using an object on wheels to parallel their own speech production patterns.

Group Session: Pairs were formed so that participants could work together during the week to discuss a predetermined topic using the phone. Speech modification work began with a writing activity to compare blocks of writing with blocks of speaking. All participants received instruction about speech production specifically related to articulatory contacts, oral postures, airflow, and voicing. Using fist postures, each person rated their own tension levels as well as vocal tension levels for others within the group.

Week Five

Individual Session: The session began with review of the automatic thought record and a brief discussion of the relaxation tape that was previously given. Participants also discussed their progress with systematic desensitization. Stuttering modification included heightening awareness, negative practice, and feedback of speech production. Cancellations were practiced to obtain a sense of control over the moment of stuttering. Pull-outs were practiced to alleviate blocks, and preparatory sets were used to feel oral postures and relax prior to a stuttering event. Pseudostuttering, prolongations, and the use of light contacts helped reinforce the concept of forward moving speech. Individual sessions also included audiotape recording and role-playing events.

Group Session: Participants had opportunities to practice stuttering modification techniques using personalized cards with targeted examples. Once again automatic thoughts were explored. A recurrent theme was the tendency toward perfectionism and the pursuit for excellence. Throughout the session, stuttering modification was integrated with cognitive behavioral techniques using role- playing and scripted plays.

Week Six

Individual Session: Participants reviewed personal accomplishments with stuttering modification and role-played feared situations. As they made a thorough review of their initially identified individual goals, it became evident how much progress they made. Each person identified personal next steps.

Group Session: Participants shared insights and realizations resulting from their involvement in the program. Participants also engaged in role playing acts from scripted dialogues and concentrated on stuttering modification. Each person identified the part of the program that was most beneficial to him/herself. Future program ideas included making the program slightly longer (9 weeks) and gaining more practice talking in groups. One of the most beneficial aspects was noted to be a new awareness of individuals’ automatic thoughts and how what one says to him/herself influences how they feel and act.

  1. Results
This case study design was analyzed using a repeated measures analysis of variance (general linear model) and post hoc paired comparisons with the Statistical Package for the Social Sciences (version 11). All testing was done by the two doctoral level ASHA certified and state licensed speech-language pathologists. Graduate students in speech-language-hearing science observed several sessions.

Figure 1 indicates a statistically significant decrease in stuttering severity on the SS1-3 from baseline to midpoint (week 3) to end (week”˜6). This was due to participants’ decrease in frequency, duration, and physical concomitants of stuttering. At baseline, the mean percent syllables stuttered was 7.32. At midpoint the average decreased to 4.2 percent syllables stuttered and at the end of the study the mean was 2.72 syllables stuttered.


Figure 1. Individual changes as recorded by the SSI-3

Participants also reported changes in attempts to hide their stuttering. Based on a rating of 0=none, 1= minimal, 2=often, and 3=continual, it was noted that at baseline, the average attempt to hide stuttering was 2.63 (sd=.74). After completing six weeks of the CBT for PWS Program, the mean decreased to 1.13 (sd=.35).

Dimensions of perfectionism and their impact on functioning have been researched (Frost, et a1., 1990) and were of major interest to this study. To capture the notion of perfectionism with people who stutter, scores from the Amster Adaptation of the Burns Perfectionism Scale were analyzed. The scale is rated from +20 (most perfectionistic) to - 20 (least perfectionistic). Each item received from +2 (very much agree) to - 2 (very much disagree) points. Rated items included statements such as “I shouldn’t have to repeat the same mistake many times.” and “If I don’t set the highest standards for myself, I am likely to end up a second rate person.” Both child recollection and current views of perfection were measured at baseline, midpoint, and again at the end of this study. Participants’ childhood recollections of perfectionism appeared to be a stable measure and did not change to any degree during the course of this study. However, when participants rated themselves currently, as 158 Theory, research and therapy in fluency disorders adults, there was a statistically significant difference from baseline to midpoint. Scores changed from +9.75 to -2.38. Although there was an additional decline in perfectionist tendencies from midpoint to end (-2.38 to - 5.5), it was not statistically significant. Overall, statistical significance reached p=.001 from the beginning to end of the study. Cognitive behavioral therapy reduced perfectionist tendencies early in the study. At the end of the study, a statistically significant correlation of .76(p=.03) between perfectionism and stuttering severity was found. A lower level of perfectionism was related to a lower stuttering severity score as measured by the SSI-3. Figure 2 provides an overview of all measures and how they changed at different points in the study.


Figure 2. Means ( standard deviations) for all measures from baseline to midpoint to end of study.

  1. Conclusion
Cognitive Behavioral Therapy (CBT) had a significant impact on participants’ levels of perfectionism with a decrease in perfectionist tendencies from aasejine to midpoint when CBT was introduced. Attitudes became more positive and speech fluency began to lose its power over participants’ feelings of worth. Levels of perfectionism continued to decline throughout the six weeks and by the end of the study a statistically significant correlation between stuttering severity rating and level or perfectionism was found.

As treatment became focused on stuttering modification (during weeks 4 through 6), the overall percent of syllables stuttered declined. Participants began to feel greater awareness about their speech behaviors and gained a sense of control over them. Several commented on their interest in learning how speech sounds are produced along with their ability to control oral and vocal tension. During this second phase, communication attitudes also started to solidify and participants’ developed a more positive attitude toward communication as measured by the Erikson Communication Attitudes Scale (Andrews & Cutler, 1974). Participants stated that they were less embarrassed about the way they talked and felt less nervous while talking. Anxiety levels, as measured by the Liebowitz Social Anxiety Scale (Sajatovic & Ramirez, 2001), also began to decrease after the CBT training. Participants began seeking new opportunities to speak. One participant spoke with colleagues at lunch after years of sitting in silence. Another person began providing small training sessions to co-workers. Another participant spoke up in a group setting for the first time. Attempting to hide stuttering by avoiding speaking was no longer the main goal.

Prior to the study, participants were not aware of their automatic thoughts and the effect they had on their interpretation of life’s events. Cognitive behavioral therapy provided a new way to analyze situations that impact daily feelings and satisfaction or upset with life’s events. Participants learned to develop alternate responses to their most prevalent cognitive distortions. These included less focusing on the negazive, a halt to mind-reading, and a greater ability to look at the whole picture in a situation. Communication attitudes improved with the addition of stuttering modification therapy. It is interesting to note that locus of control (Manning, 2001; Rotter, 1966) did not change throughout the study. This may be due to the fact that vocabulary related to an internal and external sense of control over one’s own life was not directly addressed in the study.

In general, CBT followed by stuttering modification produced effective therapeutic results. Reducing perfectionism and the need to hide stuttering had an impact on speech fluency. Six weeks of treatment with CB-T and stuttering modification made a dramatic difference in perfectionist tendencies, communication attitudes, social anxiety, and stuttering severity for all eight participants. Unsolicited comments from seven of the eight participants stated that this program was more effective than previous therapeutic interventions. Six of the eight participants commented that this program had life-changing effects. A follow-up study to determine long-term effects of this program is underway.

Amster, B. (1995). Perfectionism and stuttering. First World Congress on Fluency Disorders, 540-543.

Andrews, G., & Cutler, J. (1974). Stuttering therapy: The relation between changes in symptom level and attitudes. Journal of Speech Hearing Disorders, 39, 312-319.

Erikson, R.L. (1991). Assessing communication attitudes among stutterers. Journal of Speech and Hearing Research, 12, 711-724. .

Freeman, A., Pretzer, J ., Fleming, B., & Simon, K. (1990). Clinical application ofcognittve therapy. New York, NY: Plenum Press.

Frost, R.O., Marten, P., Lahart, C., & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitive Therapy and Research, 14, 449-468.

Guitar, B., & Bass, C. (1978). Stuttering therapy: The relation between attitude change and long term outcome. Journal of Speech Hearing Disorders, 43, 392-400.

Manning, W.H. (2001). Locus of Control of Behavior Scale. In Clinical decision making in fluency disorders, (2”" ed. ). San Diego, CA: Singular Publishing.

Neilson, M., & Andrews, G. (1992). Intensive fluency training of chronic stutterers, In R Curlee. (Ed.): Stuttering and related disorders of fluency, Current Therapy of Communication Disorders Series, New York, Thieme. '

Rapee, R. (1998) I think, they think... Overcoming social anxiety. (Video). New York, NY: Guilford Publications.

Riley, G.D. (1994). Stuttering Severity Instrument for Children and Adults, 3"‘ Edition. Circle Pines, MN: AGS Publishing.

Rotter, J .B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80, 1-28.

Ryan, B.P. (2003). Treatment efficacy research and clinical treatment. Perspectives on Fluency and Fluency Disorders, 13, 31-33.

Sajatovic, M., & Ramirez, L.F. (2001). The Liebowitz social anxiety scale. In Rating Scales in Mental Health. Hudson, OH: Lexi-Comp, Inc. '

Starkweather, C.W. (1984). A multiprocess behavioral approach to stuttering therapy, in W.H.

Perkins (Ed): Current therapy of communication disorders: Stuttering disorders. San Diego, CA: Co1lege-Hill Press.


In preparation for the 2018 World Congress the IFA is implementing Japanese translations of some pages on the site. Choosing Japanese below to see these translations.

Not all pages are translated, but you can use Google translate to see a machine translation using the switch below

Google Translate

Follow the Joint World Congress