Copyright @ 2011, Peter, S.

Critical Review:
In children with phonological/articulation disorders, do non-speech oral motor exercises improve speech
production compared to direct speech therapy?

Sherry Peter
M.Cl.Sc (SLP) Candidate
University of Western Ontario: School of Communication Sciences and Disorders

This critical review examines the evidence evaluating the efficacy of non-speech oral motor exercises (NSOMEs) as
a treatment approach for children with phonological/articulation disorders. Research studies include one randomized
clinical trial design, one single group pre-test post-test design and one single subject design. Overall, the evidence
does not support the use of NSOMEs to treat children with phonological/articulation disorders. Future and clinical
recommendations are discussed.

Introduction
The use of NSOMEs is based on a number of
Children with speech sound disorders, such as assumptions that are not strongly supported in the
phonological/articulation disorders, reportedly literature. The first assumption is that the muscles
dominate the caseload receiving speech and language used for speech are used the same way for non-
services (Lass & Pannbacker, 2008). Speech- speech activities (Bowen, 2005). Conversely, there is
Language Pathologists (SLPs) often use a variety of research indicating that there are differences in the
phonetic or phonemic-based approaches to treat organization of the nervous system for non-speech
children with phonological/articulation disorders. and speech movements. As a result, non-speech
Another treatment approach that is commonly used movements may not convert into speech movements
by SLPs is the use of non-speech oral motor exercises (Bowen, 2005). The second assumption is that if non-
(NSOMEs) (Lof & Watson, 2008). Lof and Watson speech activities are taught with relevance to speech,
(2008) documented that approximately 85% of SLPs, the underlying movements can be integrated for
who are certified members of the American Speech speech. Forrest (2002) argued against this claim
and Hearing Association (ASHA), use NSOMEs as a stating that there is evidence suggesting that speech
primary treatment activity or in conjunction with should be taught as whole units rather than discrete
other speech sound treatment approaches. parts. Since NSOMEs are taught discretely, they are
not likely to integrate into speech behaviours. Lastly,
NSOMEs can be defined as a set of therapeutic it is assumed that NSOMEs strengthen the
activities that do not require speech production articulators needed for speech. However, according
(Ruscello, 2008). NSOMEs can be further to Forrest (2002), typically developing children use
categorized into active muscle exercises, passive approximately 10-20% of maximal lip force abilities
muscle exercises and sensory stimulation. Active and 11-15% of maximal jaw force abilities when
muscle exercises, such as stretching, are used to build producing speech. This suggests that minimal
muscle strength. Passive muscle exercises, such as strength is essentially needed for speech sound
passive range of motion, are used to maintain joint production (Forrest, 2002).
flexibility, modify tone and improve vascular
circulation. Additionally, sensory stimulation, such as In addition to a weak theoretical foundation, the
massage or vibration, is used to enhance muscle limited evidence base does little to resolve the
function (Ruscello, 2008). Therefore, the focus of dispute among researchers. To date, there have been
implementing NSOMEs in speech therapy is to few high quality studies published that have
improve the overall structure and function of the evaluated the effectiveness of NSOMEs to treat
speech mechanism (Ruscello, 2008). children with phonological/articulation disorders.
However, the existing findings indicate that the use
Currently, there is a lack of theoretical and empirical of NSOMEs in therapy is not empirically supported.
evidence supporting the efficacy of NSOMEs in SLPs are advised to be cognisant of the available
therapy. However, there are SLPs who continue to research findings when selecting treatment methods
use this approach despite limited research in the field. for their clients. The use of evidence-based practice
For this reason, a long-standing debate exists among may assist clinicians in providing effective services
researchers about whether it is appropriate to use a to clients.
treatment approach without sufficient and favourable
evidence available.
 Copyright @ 2011, Peter, S.

Objectives for six weeks, then twice a week for eight weeks for a
period of 14 weeks.
The primary objective of this review is to critically
evaluate the current literature examining the efficacy Two independent observers evaluated pre-test and
of NSOMEs as a treatment approach for children post-test scores on a test battery to determine if
with phonological/articulation disorders. A secondary therapy facilitated any gains in speech production.
objective is to provide future and clinical The test battery included a word repetition test to
recommendations. evaluate tongue tip placement in the production of
/n/, /d/, /l/, /n/, /s/, and /z/, a picture articulation test,
Method the Goldman-Fristoe Test of Articulation (GFTA)
(Goldman & Fristoe, 1969) and a tongue thrust
Search Strategy evaluation.
Databases searched were CINAHL and PubMed.
They were searched using the following terms: ((oral Nonparametric tests were used to interpret results.
motor exercises) AND (articulation disorders or The Sign Test was used to determine if
phonological disorders)), ((nonspeech oral motor improvements in speech production were due to
exercises) AND (articulation disorders or chance or to intervention. The performance of both
phonological disorders)). The search was limited to groups improved significantly with the exception of
articles in English. There was no limitation on the the scores on the tongue thrust evaluation for the
date of articles. group receiving articulation therapy only. The Fisher
Exact Test was used to determine if the differences
Selection Criteria between the two groups were significant. Results
Three articles were selected to illustrate the direct revealed that the differences between the groups were
implementation of NSOMEs to treat children with only significant for the tongue thrust evaluation (0.05
phonological/articulation disorders. level of confidence) and not for the word repetition
test, picture articulation test or the GFTA (Goldman
Data Collection & Fristoe, 1969). Inter-rater reliability was 97.4% for
The literature search yielded one randomized clinical presence or absence of tongue thrust, 87.8% for
trial design, one single group pre-test post-test design dentalization of /t/, /d/, /n/, and /l/, and 90.2% on
and one single subject design. correctness of /s/ and /z/ productions. The results of
the study indicated that both groups exhibited equal
Results improvements in speech production. Oral
myofunctional therapy did not improve speech
Randomized Clinical Trial Study production, but it was effective in improving tongue
Early research suggests that the use of NSOMEs does thrust behaviours.
not improve speech production compared to direct
speech therapy. Christensen and Hanson (1981) used Based on research design and methodology, level IIa
a between groups design to determine if oral evidence was obtained from the study. Participants
myofunctional therapy facilitated an improvement in were specified and a description of treatment
speech production in children with articulation procedures was provided. The authors used a
disorders and tongue thrust behaviours. Participants comparison group, appropriate blinding and valid
included six boys and four girls between the ages of statistical measures to evaluate the results of the
5;8 and 6;9. All participants were from white middle study. However, a small sample size was included
class families with parents who were willing to and the authors did not discuss selection bias or
participate directly in the study. Each participant had randomization procedures. Thus, the evidence for the
a severe anterior tongue thrust, a frontal lisp, normal use of NSOMEs as a treatment approach is equivocal.
hearing, normal developmental milestones, no
physiological or psychological issues and no Single Group Pre-test Post-test Study
involvement in previous speech therapy. Children More recent research complements earlier reports
were assigned to groups receiving either articulation suggesting that the use of NSOMEs in therapy does
therapy only or a combination of articulation therapy not improve speech production. Guisti Braislin and
and oral myofunctional therapy, which included the Cascella (2005) examined the efficacy of an oral
use of neuromuscular facilitation techniques. All motor approach, Easy Does it for Articulation: An
participants received a total of 22 individual half- Oral Motor Approach (Strode & Chamberlain, 1997),
hour sessions. They attended sessions once a week without direct articulation therapy for children with
mild articulation disorders. Participants included two
 Copyright @ 2011, Peter, S.

boys and two girls in grade one between the ages of an informal test of volitional movements (VOM), the
6;4 and 6;8. All participants were white and English GFTA-2 (Goldman & Fristoe, 2000), the Peabody
speaking. They all passed a hearing screening, had Picture Vocabulary Test-3 (PPVT-3) (Dunn & Dunn,
normal academic and cognitive abilities, received a 1997), the Clinical Evaluation of Language
diagnosis of a mild articulation disorder of unknown Fundamentals-Preschool (CELF-P) (Wiig, Secord &
origin, and had adequate oral and speech motor Semel, 1992) and a non-word repetition test. A
production as rated on the Kaufman Speech Praxis language sample and speech recordings were also
Test for Children (KSPT) (Kaufman, 1995) and the collected. Participants were included in the study if
Oral Speech Mechanism Screening Examination they scored below 85 on the GFTA-2 (Goldman &
(OSMSE) (St. Louis & Ruscello, 2000). Children Fristoe, 2000) and had speech difficulties with 3
were assigned to two groups of two. They received a unrelated speech sounds on a 200-word probe test.
total of 15 individual half-hour sessions. They All children received both types of training
attended sessions twice a week for a period of 7 alternatively on two unrelated target speech sounds.
weeks. One sound was treated using a traditional phonetic
approach and one sound was treated with NSOMEs.
The GFTA-2 (Goldman & Fristoe, 2000) was Another unrelated target sound was selected as a
administered at pre-test and post-test to examine if control. Each participant received both treatments
there were any improvements in speech production. during a 60-minute session twice a week. Each type
Descriptive measures were used to interpret the of treatment included imitation and spontaneous
results. The raw scores for the group at pre-test production phases. During sessions, participants
ranged from 4 to 13 errors (mean = 9, standard moved to spontaneous productions when they scored
deviation = 4.70) and the raw scores at post-test 15/20 for imitated productions across 3 consecutive
ranged from 2-10 errors (mean = 6.5, standard sets. The program was completed once participants
deviation = 3.41). Overall, participants made 2.5 correctly produced 20 out of 30 target sounds on a
fewer errors at post-test. Based on the standard sound-specific generalization probe. However, if
deviations, the number of errors did not significantly participants completed 20 sessions before achieving
change for participants. Inter-rater reliability ranged this score, the program was considered completed.
from 84% to 97%. The results of the study indicated
that all participants did not exhibit significant Statistical measures were used to interpret results.
improvements in speech production after receiving Using participants’ scores from sound-specific
oral motor therapy. generalization probes, the difference in percentage of
correct production of a target sound from pre-
Considering research design and methodology, level treatment to post-treatment was calculated. One of
III evidence was obtained from the study. the participants did not complete the post-treatment
Participants were specified and a description of sound probe due to a scheduling conflict.
treatment was provided. Outcomes were evaluated Consequently, all participants’ scores were based on
with appropriate blinding. However, the study had the second last sound probe rather than the post-
several weaknesses including small sample size, short treatment sound probe. The mean average change of
study duration and the lack of a control group. PT was a 30% increase from pre-treatment while the
Though the authors used valid descriptive measures, mean average change of NSOMEs was a 3%
they did not include a statistical analysis and thus the increase. A paired t-test was calculated and the
findings cannot be judged as statistically significant. results were statistically significant for PT. Eight of
The evidence from the study can be considered the nine participants showed more improvement
equivocal for the use of NSOMEs in therapy. producing the sound treated with PT compared to the
sound treated with NSOMEs.
Single Subject Study
The following study also provides evidence that the A qualitative evaluation of PT and NSOMEs was
use of NSOMEs is not effective for improving speech completed. The authors found that if NSOMEs
production. Forrest & Iuzzini (2008) examined if oral preceded PT in a treatment session, it did not result in
motor training improved speech development in improved speech production compared to if PT
children with phonological/articulatory disorders preceded the use of NSOMEs. They also found that
(PADs) compared to production training (PT). children with the lowest VOM scores did not show
Participants included nine children between the ages any significant gains when compared to children with
of 3;3 and 6;3. They all spoke English, had normal the highest VOM scores. Therefore, NSOMEs did not
oral structures and passed a pure tone hearing improve oral motor function. The overall findings
screening. In addition, participants were administered
 Copyright @ 2011, Peter, S.

suggested that oral motor training did not improve methodological weaknesses, such as study duration
speech development in children who have PADs. (7.5 treatment hours) and a small sample size.

Based on research design and methodology, level I A second limitation is that participants may have
evidence was provided. This is considered to be the been beyond the age in which their sound acquisition
highest quality of experimental evidence available. patterns could be influenced. According to Forrest
Participants were specified and a description of (2002), early sound correction is largely dependent
treatment procedures was included. The authors used on oral motor development and shaping rather than
valid statistical measures and they adequately later sound correction. Participants in the reviewed
controlled for order effects by randomizing the order studies ranged from ages 3;3 to 6;9, with the majority
of specific treatments received. However, the study of children being in the older end of the range
had a few weaknesses, which reduces the overall (Christensen & Hanson, 1981; Guisti Braislin &
strength of the evidence. The authors used a small Cascella, 2005; Forrest & Iuzzini, 2008). Therefore,
sample size and they did not provide any information the age of participants may have been another factor
about blinding assessors or reliability of outcome that affected the overall findings.
measures. Also, results may have been slightly
skewed due to a participant dropping out of the study A final concern is that some researchers do not
in the second last session. Though the authors provide detailed descriptions of the types of
provide recent and strong evidence against using NSOMEs used in therapy (Christensen & Hanson,
NSOMEs in therapy, the evidence can be considered 1981; Forrest & Iuzzini, 2008). It is unclear which
equivocal until additional well-designed studies are exercises were used in some studies (e.g., active
conducted to replicate similar findings. muscle exercises, passive muscle exercises and/or
sensory stimulation) and whether any of them were
Discussion observed to be more or less effective than others. If
this information was provided, it may help to support
The purpose of this review was to determine if the or refute assumptions that NSOMEs can convert into
use of NSOMEs is an effective and appropriate speech activities or can be integrated for speech
treatment approach for children with phonological/ (Bowen, 2005).
articulation disorders. Overall, the studies used valid
measures, but they had a number of methodological Conclusion
flaws, such as small sample size (Christensen &
Hanson, 1981; Guisti Braislin & Cascella, 2005; The best available research concludes that the use of
Forrest & Iuzzini, 2008), and insufficient information NSOMEs is not an effective treatment approach
about randomization procedures (Christensen & alone or in combination with direct speech therapy
Hanson, 1981), blinding procedures and reliability of for children with phonological/articulation disorders.
outcome measures (Forrest & Iuzzini, 2008). Overall, the existing evidence can be considered
Therefore, based on the reviewed studies, evidence is equivocal. Further research is needed to provide more
either weak or lacking in support of using NSOMEs compelling evidence for or against the use of
in therapy to improve speech sound production. NSOMEs in a clinical setting.

When reviewing the results of the studies, the reader Future Recommendations
should be mindful of certain limitations that may
have unintentionally affected the findings. One Based on the weaknesses of the studies discussed in
limitation involves the minimal progress made by this review, the following recommendations are made
participants. Christensen and Hanson (1981) noted to improve the strength of evidence in future studies:
that some participants in their study did not show any
progress compared to other participants regardless of 1. Well-designed and randomized studies
which type of treatment they received. This may be should be employed that offer stronger
related to the fact that some participants presented levels of evidence and larger sample sizes
with co-existing language problems, which possibly should be incorporated so findings can be
influenced the findings of the study. Additionally, generalized to encompass all children with
Guisti Braislin and Cascella (2005) reported that all phonological/articulation disorders.
participants did not make any progress during their 2. Individual characteristics, such as age and
study. Though the authors concluded that NSOMEs co-existing language problems, should be
are ineffective for improving speech production, a considered to determine if certain factors
lack of progress may have been associated with influence the outcome of studies.
 Copyright @ 2011, Peter, S.

3. A detailed description of treatment should motor treatments. Language, Speech and

be provided to determine if there are Hearing Services in Schools, 39, 408-421.
effective types of NSOMEs. Lof, G.L., & Watson, M.M. (2008). A nationwide
survey of nonspeech oral motor exercise
Clinical Implications use: Implications for evidence-based
practice. Language, Speech, and Hearing
SLPs should carefully evaluate the available research Services in Schools, 39, 392-407.
on NSOMEs and incorporate scientific evidence into Ruscello, D.M. (2008). Nonspeech oral motor
their daily practice. If clinicians choose to use treatment issues related to children with
NSOMEs in therapy, clients or parents of clients developmental speech sound disorders.
should be made aware that this approach is, at most, Language, Speech, and Hearing Services in
considered experimental. However, it is strongly Schools, 39, 380-391.
suggested that SLPs use evidence-based speech St. Louis, K.O., & Ruscello, D.M. (2000). Oral
sound treatments to ensure positive gains in their Speech Mechanism Screening Examination
clients’ communication abilities. (3rd ed.). Austin, TX: Pro-Ed.
Strode, R. & Chamberlain, C. (1997). Easy does it for
References articulation: An oral motor approach. East
Moline, IL: Linguisystems.
Bowen, C. (2005). What is the evidence for oral Wiig, E., Secord, W., & Semel, E. (1992). Clinical
motor therapy? Acquiring Knowledge in Evaluation of Language Fundamentals-
Speech, Language and Hearing, Speech Preschool. San Antonio, TX: The
Pathology Australia, 7, 144-147. Psychological Corporation.
Christensen, M., & Hanson, M. (1981). An
investigation of the efficacy of oral
myofunctional therapy as a precursor to
articulation therapy for pre-first grade
children. Journal of Speech and Hearing
Disorders, 46, 160-167.
Dunn, L.M., & Dunn, L.M. (1997). Peabody Picture
Vocabulary Test (3rd ed.). Circle Pines, MN:
American Guidance Service.
Forrest, K. (2002). Are oral-motor exercises useful in
the treatment of phonological/articulatory
disorders? Seminars in Speech and
Language, 23(1), 15-25.
Forrest, K., & Iuzzini, J. (2008). A comparison of
oral motor and production training for
children with speech sound disorders.
Seminars in Speech and Language, 29, 304-
311.
Goldman, R., & Fristoe, M. (1969). Goldman Fristoe
Test of Articulation. Circle Pines, MN:
American Guidance Service.
Goldman, R., & Fristoe, M. (2000). Goldman Fristoe
Test of Articulation (2nd ed.). Circle Pines,
MN: American Guidance Service.
Guisti Braislin, M.A., & Cascella, P.W. (2005). A
preliminary investigation of the efficacy of
oral motor exercises for children with mild
articulation disorders. International Journal
of Rehabilitation Research, 28, 263-266.
Kaufman, N.R. (1995). Kaufman Speech Praxis Test
for Children. East Detroit, MI: Wayne State.
Lass, N.J., & Pannbacker, M. (2008). The application
of evidence-based practice to nonspeech oral