Selective mutism: self-modelling, graded exposure and what the trials show

Background: what selective mutism is

Selective mutism describes a child who speaks normally in some places, usually at home with close family, but consistently cannot speak in others where talking is expected, most often at nursery or school. The US diagnostic manual, DSM-5, classes it as an anxiety disorder. To meet the criteria, the pattern must last at least a month, not counting only the first month at a new school; it must get in the way of learning or social communication; and it must not be explained by the child not yet knowing the language, or by a communication disorder such as stuttering (American Psychiatric Association, 2013); (Muris & Ollendick, 2021). The NHS stresses the point families most need to hear: the child is unable to speak in those moments, not refusing to (NHS, 2023).

How common is it? In a large US school district, teachers of children in the first three school years identified 16 of 2,256 pupils (0.71%) who met criteria (Bergman et al., 2002); a survey of urban primary schools in Ireland found 20 in a catchment of 10,927 (0.18%) (Sharkey & McNicholas, 2012). Published estimates range from 0.03% to 1.9% depending on the setting (Muris & Ollendick, 2021). The NHS gives about 1 in 140 young children, more often girls (NHS, 2023).

When does it start? Onset is usually before age 5, and it often comes to attention when a child starts school (Muris & Ollendick, 2021). In a mainly US survey of 230 families, diagnosis came on average two years after onset (Boneff-Peng et al., 2025).

Why it is classed as anxiety. A meta-analysis of 22 studies (837 children) found that 80% of children with selective mutism had another anxiety disorder, most often social anxiety (69%), with wide variation between studies (Driessen et al., 2020). Reviewers support the anxiety classification while noting that some children also have speech and language difficulties, developmental delays or autistic traits (Muris & Ollendick, 2015); (Muris & Ollendick, 2021).

Children learning a new language. Many children have a normal silent period when learning a second language. Selective mutism appears at least three times more common in immigrant, language-minority children, but telling the two apart needs careful assessment (Toppelberg et al., 2005).

Why it matters. Children in the US school study had improved six months later but were still impaired, so the authors argued for intervention rather than waiting (Bergman et al., 2002). Most young people recover by adolescence, but later anxiety disorders are common (Koskela et al., 2023). The hard moments are everyday ones: the register, the toilet, the doctor.

How the research has been done

Selective mutism is uncommon, so most studies are small. Most of the literature consists of case reports and single-case designs, in which one or a few children are measured repeatedly before and during treatment (Cohan et al., 2006); (Zakszeski & DuPaul, 2017). Since 2013 a few randomised controlled trials (RCTs) have appeared, the behavioural ones with roughly 20 to 30 children each, usually against a waiting list, which shows whether treatment beats the passage of time but not whether it beats another credible treatment.

Typical outcomes are parent ratings (Selective Mutism Questionnaire), teacher ratings (School Speech Questionnaire), words spoken at school, and remission judged by an assessor blind to group. Reviews report effect sizes such as Hedges' g, a standardised group difference where 0.8 or more is conventionally large; our methods glossary explains these terms. Measures vary so much that comparing studies is hard (Hipolito et al., 2023).

What the reviews and trials found

Narrative reviews. A review of 23 studies from 1990 to 2005 found most support for behavioural and cognitive-behavioural approaches, despite widespread methodological weaknesses (Cohan et al., 2006). A review of 21 studies from 2005 to 2015 named four recurring ingredients: reinforce, shape, expose and fade. Reinforcement rewards any step towards speaking; shaping builds from gestures to sounds to words; exposure climbs a ladder of feared speaking situations; stimulus fading starts with a person the child talks to and gradually brings in someone new. Treatments were generally reported as effective, but effect sizes were rarely given, and school-based programmes were usually run by researchers or clinicians rather than school staff (Zakszeski & DuPaul, 2017).

Meta-analyses. Pooling five RCTs (233 children), psychological treatments outperformed no treatment with a large effect (g = 0.87) (Steains et al., 2021). A broader review found 25 studies of non-drug treatments. Three trials of a combined behavioural and home–school approach against a waiting list gave a large effect on speaking (g = 1.06, 95% CI 0.57 to 1.56), but the remission estimate was imprecise and not significant, and two trials against active treatments showed no significant overall difference (Hipolito et al., 2023). A 2025 meta-analysis of 12 mostly uncontrolled studies (472 participants) found large improvements (g = 1.00), smaller for web-based than face-to-face treatment (Iimura et al., 2025).

Integrated Behavior Therapy for Selective Mutism. In this pilot RCT, blind assessors rated 75% of treated children as responders, while waitlisted children did not improve significantly. Words spoken at school rose from baseline but did not differ significantly between groups, and social anxiety fell on parent but not teacher ratings. Gains held at three months (Bergman et al., 2013). The trial registration (ClinicalTrials.gov NCT00458198) names graduated exposure as the main ingredient, alongside desensitisation, shaping and self-modelling.

The Norwegian home-and-school programme. Treatment began at home, where the child felt safest, then moved to kindergarten or school, using stimulus fading and rewards in short, frequent sessions. Adults practised defocused communication: sitting beside the child, avoiding direct questions and talking about a shared activity rather than about the child (Oerbeck et al., 2014). One year after treatment, gains had held; 78% of children aged 3 to 5 no longer had selective mutism, compared with 33% of children aged 6 to 9 (Oerbeck et al., 2015). Five years on, 21 of 30 children were in full remission and 4 still met criteria; 23% met criteria for social anxiety, and half said talking outside home was still somewhat challenging (Oerbeck et al., 2018). Defocused communication has not been tested on its own (Muris & Ollendick, 2021).

Intensive group treatment. In the trial of a five-day group programme running for most of each day, social anxiety and speaking in social situations improved by week 4, but selective mutism severity and speaking at home did not yet differ significantly from the waiting list. Teachers in the following school year rated more classroom speaking than the previous year's teachers (Cornacchio et al., 2019).

Other designs. In a study of nine young people comparing two components in alternation, exposure-based practice with the child was more effective than parent-focused contingency management (reward systems) (Vecchio & Kearney, 2009). An online parent-training pilot found improvement in selective mutism at school, but not in total scores or at home, among 13 completing families (Yamanaka et al., 2025). A Dutch RCT of a school-based behavioural protocol has been designed (Rodrigues Pereira et al., 2020); we did not find published results.

Medication. Medication is outside our scope. A systematic review found improvement in 66 of 79 children given SSRI antidepressants, but only 3 of 10 studies had an unmedicated comparison group (Manassis et al., 2016). The NHS describes it as generally for older children and never a replacement for behavioural approaches (NHS, 2023): a decision for a specialist with the family.

Video self-modelling, augmented self-modelling and audio feedforward

Self-modelling means repeatedly watching an edited recording that shows only yourself doing the desired behaviour. For selective mutism, a recording is typically made of the child answering questions where they do speak, for example with a parent, then edited so the child appears to answer the teacher in class. In an early single-case report, Kehle and colleagues described a child whose mutism resolved within five 5-minute viewing sessions (Kehle et al., 1990). They later used the term augmented self-modelling for the video combined with other behavioural techniques (Kehle et al., 1998), and later described it as relatively successful and among the least intrusive options (Kehle et al., 2012).

Audio feedforward is the sound-only version: children listen to tapes edited to depict them speaking in situations where they do not yet speak. A case report described success with three children at school and in the community, but also noted that the approach did not always work, partly because some children refused to make the recordings (Blum et al., 1998). Later reports combined video self-modelling with other techniques: role play plus video self-modelling increased a 9-year-old girl's spoken responses and initiations when ordering in a restaurant, meeting new adults and playing with new children (Lang et al., 2011), and video self-modelling plus stimulus fading and reinforcement increased speech in three 8-year-olds, with gains maintained at one month (Bork & Bennett, 2020).

The honest summary: these are encouraging single cases and series of three, almost always bundled with fading and rewards, so the effect of watching or hearing oneself cannot be separated. No controlled trial has tested it alone; its supported role is as one ingredient in a wider behavioural programme.

For whom and under what conditions

• Age. Younger children improved more in the Norwegian trial and its follow-ups; older age, greater severity and a family history of selective mutism predicted poorer long-term outcomes (Oerbeck et al., 2014); (Oerbeck et al., 2018). The long-term review points the same way (Koskela et al., 2023).
• Where it happens. Effective programmes work where the child cannot speak, especially school, not only in a clinic (Oerbeck et al., 2014); (Zakszeski & DuPaul, 2017).
• Exposure matters. In one small study, graded speaking practice was more effective than parent-run reward systems (Vecchio & Kearney, 2009).
• Parents. Parents understandably adapt to a child's anxiety, for example by answering for them. In an open trial, parental accommodation fell during treatment alongside symptoms (Shorer et al., 2023).
• Format. Weekly and intensive formats both show promise; web-based treatment showed smaller pooled gains than face-to-face (Iimura et al., 2025).
• Co-occurring needs. Speech and language problems, developmental delay or autism call for adapted support, including a speech and language therapist where needed (Muris & Ollendick, 2021). Bilingual children need assessment that separates a normal silent period from selective mutism (Toppelberg et al., 2005).

What this means in practice

• Take the pressure off first. Do not pressure or bribe a child to speak. Accept nods, pointing and smiles, respond warmly and without fuss if they do speak, and praise privately rather than in front of others (NHS, 2023).
• Work in small steps. Build a ladder from what the child can do now, for example from gestures, to sounds, to single words with a parent nearby, to a word with a new adult. This is the shaping, fading and graded exposure common to the trials (Zakszeski & DuPaul, 2017).
• Start where the child talks. The Norwegian programme began at home and then moved to school, bringing in a new person gradually while a trusted adult was present (Oerbeck et al., 2014).
• Change how adults talk. Sit alongside, comment on a shared activity and offer choices that can be answered with a nod, rather than direct questions.
• Make school a partner. Short, frequent sessions in school and a shared home–school plan run through the programmes reviewed here; a widely used practitioner manual sets out the steps (Johnson & Wintgens, 2016).
• Do not wait it out. The NHS advises families not to accept that a child will simply grow out of it (NHS, 2023). If a child has not spoken at school after the first month or so, or not speaking is affecting learning, friendships or wellbeing, talk to the GP, the school's special educational needs coordinator or a speech and language therapy service. Younger children have responded better to treatment (Oerbeck et al., 2015).

What this means for short personalised cartoons

Personalised video has an obvious appeal here, because self-modelling and feedforward are about children seeing or hearing themselves succeed (Kehle et al., 1990); (Blum et al., 1998). Several points need care.

• What plausibly transfers. Preparation for a specific moment, a picture of a small achievable step, and a guide that helps adults use shaping and fading in real settings. These fit the behavioural approach above.
• What does not transfer automatically. Self-modelling studies used the child's own real face and voice. A cartoon character based on a child is a different stimulus, and whether it works the same way is untested. See our reviews of personalisation and cartoons versus real video.
• The risk of pressure. A video in which the child's character chats freely could feel like an expectation. A safer story shows small, low-pressure steps, such as waving, pointing or whispering to one trusted person, at the child's own pace.
• Consent and comfort. Some children in the feedforward work refused to make recordings (Blum et al., 1998). Any personalised material should be made with the child's agreement and abandoned if it causes distress.

Cala Luna's personalised cartoons have not been tested in any trial, including with children who have selective mutism. At most, a cartoon or social story could be one preparation tool inside a plan agreed with school and, where needed, a professional. It is not a treatment for selective mutism.

Limits of the evidence

• Few trials: around five RCTs, each with roughly 20 to 30 children for the behavioural ones, mostly compared with waiting lists rather than other treatments (Steains et al., 2021); (Hipolito et al., 2023).
• Parents and teachers who rate outcomes usually know whether the child was treated.
• Outcome measures differ widely, and few studies report remission, wellbeing, school progress or quality of life (Hipolito et al., 2023).
• Self-modelling and feedforward rest on case reports and small case series, with components bundled together and a likely bias towards publishing successes.
• Much of the trial evidence comes from a few teams in the USA and Norway; long-term follow-up exists mainly for the Norwegian sample.

Open questions for research

• Does video self-modelling add anything to stimulus fading and reinforcement? A dismantling design, with and without the video component, could answer this even in small samples.
• Does an animated, personalised avatar work like real self-video, or does the child need to see and hear their own face and voice?
• Can class teachers deliver graded exposure with brief training and supervision, with outcomes comparable to researcher-delivered programmes?
• What is the minimum effective dose? We found no direct comparison of intensive one-week groups with six-month weekly programmes.
• How should assessment and treatment be adapted for bilingual and autistic children?
• Could the field agree a core outcome set, including remission, school functioning and the child's own view, so that future trials can be pooled?

Frequently asked questions

References

1. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association. doi.org/10.1176/appi.books.9780890425596
2. NHS. (2023). Selective mutism. NHS website (page last reviewed 17 February 2023). www.nhs.uk/mental-health/conditions/selective-mutism/
3. Bergman, R. L., Piacentini, J., & McCracken, J. T. (2002). Prevalence and description of selective mutism in a school-based sample. Journal of the American Academy of Child & Adolescent Psychiatry, 41(8), 938–946. doi.org/10.1097/00004583-200208000-00012 · PubMed
4. Sharkey, L., & McNicholas, F. (2012). Selective mutism: A prevalence study of primary school children in the Republic of Ireland. Irish Journal of Psychological Medicine, 29(1), 36–40. doi.org/10.1017/S0790966700017596 · PubMed
5. Muris, P., & Ollendick, T. H. (2015). Children who are anxious in silence: A review on selective mutism, the new anxiety disorder in DSM-5. Clinical Child and Family Psychology Review, 18(2), 151–169. doi.org/10.1007/s10567-015-0181-y · PubMed
6. Muris, P., & Ollendick, T. H. (2021). Current challenges in the diagnosis and management of selective mutism in children. Psychology Research and Behavior Management, 14, 159–167. doi.org/10.2147/PRBM.S274538 · PubMed
7. Boneff-Peng, K., Lasutschinkow, P. C., Colton, Z. A., & Freedman-Doan, C. R. (2025). An updated characterization of childhood selective mutism: Exploring clinical features, treatment utilization, and school services. Child Psychiatry & Human Development, 56(3), 683–695. doi.org/10.1007/s10578-023-01589-8 · PubMed
8. Driessen, J., Blom, J. D., Muris, P., Blashfield, R. K., & Molendijk, M. L. (2020). Anxiety in children with selective mutism: A meta-analysis. Child Psychiatry & Human Development, 51(2), 330–341. doi.org/10.1007/s10578-019-00933-1 · PubMed
9. Toppelberg, C. O., Tabors, P., Coggins, A., Lum, K., & Burger, C. (2005). Differential diagnosis of selective mutism in bilingual children. Journal of the American Academy of Child & Adolescent Psychiatry, 44(6), 592–595. doi.org/10.1097/01.chi.0000157549.87078.f8 · PubMed
10. Koskela, M., Ståhlberg, T., Yunus, W. M. A. W. M., & Sourander, A. (2023). Long-term outcomes of selective mutism: A systematic literature review. BMC Psychiatry, 23, 779. doi.org/10.1186/s12888-023-05279-6 · PubMed
11. Cohan, S. L., Chavira, D. A., & Stein, M. B. (2006). Practitioner review: Psychosocial interventions for children with selective mutism: A critical evaluation of the literature from 1990–2005. Journal of Child Psychology and Psychiatry, 47(11), 1085–1097. doi.org/10.1111/j.1469-7610.2006.01662.x · PubMed
12. Zakszeski, B. N., & DuPaul, G. J. (2017). Reinforce, shape, expose, and fade: A review of treatments for selective mutism (2005–2015). School Mental Health, 9(1), 1–15. doi.org/10.1007/s12310-016-9198-8
13. Steains, S. Y., Malouff, J. M., & Schutte, N. S. (2021). Efficacy of psychological interventions for selective mutism in children: A meta-analysis of randomized controlled trials. Child: Care, Health and Development, 47(6), 771–781. doi.org/10.1111/cch.12895 · PubMed
14. Hipolito, G., Pagnamenta, E., Stacey, H., Wright, E., Joffe, V., Murayama, K., & Creswell, C. (2023). A systematic review and meta-analysis of nonpharmacological interventions for children and adolescents with selective mutism. JCPP Advances, 3(3), e12166. doi.org/10.1002/jcv2.12166 · PubMed
15. Iimura, D., Tsujita, N., Aoki, M., & Hagihara, H. (2025). Meta-analysis of behavioral treatments for selective mutism: Findings from selective mutism questionnaire (SMQ) and school speech questionnaire (SSQ). Child and Adolescent Psychiatry and Mental Health, 19, 40. doi.org/10.1186/s13034-025-00891-8 · PubMed
16. Bergman, R. L., Gonzalez, A., Piacentini, J., & Keller, M. L. (2013). Integrated Behavior Therapy for Selective Mutism: A randomized controlled pilot study. Behaviour Research and Therapy, 51(10), 680–689. doi.org/10.1016/j.brat.2013.07.003 · PubMed
17. Oerbeck, B., Stein, M. B., Wentzel-Larsen, T., Langsrud, Ø., & Kristensen, H. (2014). A randomized controlled trial of a home and school-based intervention for selective mutism – defocused communication and behavioural techniques. Child and Adolescent Mental Health, 19(3), 192–198. doi.org/10.1111/camh.12045 · PubMed
18. Oerbeck, B., Stein, M. B., Pripp, A. H., & Kristensen, H. (2015). Selective mutism: Follow-up study 1 year after end of treatment. European Child & Adolescent Psychiatry, 24(7), 757–766. doi.org/10.1007/s00787-014-0620-1 · PubMed
19. Oerbeck, B., Overgaard, K. R., Stein, M. B., Pripp, A. H., & Kristensen, H. (2018). Treatment of selective mutism: A 5-year follow-up study. European Child & Adolescent Psychiatry, 27(8), 997–1009. doi.org/10.1007/s00787-018-1110-7 · PubMed
20. Cornacchio, D., Furr, J. M., Sanchez, A. L., Hong, N., Feinberg, L. K., Tenenbaum, R., Del Busto, C., Bry, L. J., Poznanski, B., Miguel, E., Ollendick, T. H., Kurtz, S. M. S., & Comer, J. S. (2019). Intensive group behavioral treatment (IGBT) for children with selective mutism: A preliminary randomized clinical trial. Journal of Consulting and Clinical Psychology, 87(8), 720–733. doi.org/10.1037/ccp0000422 · PubMed
21. Catchpole, R., Young, A., Baer, S., & Salih, T. (2019). Examining a novel, parent child interaction therapy-informed, behavioral treatment of selective mutism. Journal of Anxiety Disorders, 66, 102112. doi.org/10.1016/j.janxdis.2019.102112 · PubMed
22. Vecchio, J., & Kearney, C. A. (2009). Treating youths with selective mutism with an alternating design of exposure-based practice and contingency management. Behavior Therapy, 40(4), 380–392. doi.org/10.1016/j.beth.2008.10.005 · PubMed
23. Yamanaka, T., Yuruki, K., Koyama, Y., Koyama, H., & Inoue, M. (2025). A pilot study of an online behavioral parent training program for children with selective mutism: Feasibility and preliminary effectiveness. Child and Adolescent Psychiatry and Mental Health, 19, 144. doi.org/10.1186/s13034-025-00976-4 · PubMed
24. Rodrigues Pereira, C., Ensink, J. B. M., Güldner, M. G., Kan, K. J., de Jonge, M. V., Lindauer, R. J. L., & Utens, E. M. W. J. (2020). Effectiveness of a behavioral treatment protocol for selective mutism in children: Design of a randomized controlled trial. Contemporary Clinical Trials Communications, 19, 100644. doi.org/10.1016/j.conctc.2020.100644 · PubMed
25. Shorer, M., Ben-Haim, Z., Klauzner, N., Ben-Ami, N., & Fennig, S. (2023). The Integrated Behavior Therapy for Children with Selective Mutism: Findings from an open pilot study in a naturalistic setting. Clinical Child Psychology and Psychiatry, 28(2), 465–482. doi.org/10.1177/13591045221075526 · PubMed
26. Manassis, K., Oerbeck, B., & Overgaard, K. R. (2016). The use of medication in selective mutism: A systematic review. European Child & Adolescent Psychiatry, 25(6), 571–578. doi.org/10.1007/s00787-015-0794-1 · PubMed
27. Kehle, T. J., Owen, S. V., & Cressy, E. T. (1990). The use of self-modeling as an intervention in school psychology: A case study of an elective mute. School Psychology Review, 19(1), 115–121. doi.org/10.1080/02796015.1990.12087342
28. Kehle, T. J., Madaus, M. R., Baratta, V. S., & Bray, M. A. (1998). Augmented self-modeling as a treatment for children with selective mutism. Journal of School Psychology, 36(3), 247–260. doi.org/10.1016/S0022-4405(98)00013-2
29. Kehle, T. J., Bray, M. A., Byer-Alcorace, G. F., Theodore, L. A., & Kovac, L. M. (2012). Augmented self-modeling as an intervention for selective mutism. Psychology in the Schools, 49(1), 93–103. doi.org/10.1002/pits.21589
30. Blum, N. J., Kell, R. S., Starr, H. L., Lender, W. L., Bradley-Klug, K. L., Osborne, M. L., & Dowrick, P. W. (1998). Case study: Audio feedforward treatment of selective mutism. Journal of the American Academy of Child & Adolescent Psychiatry, 37(1), 40–43. doi.org/10.1097/00004583-199801000-00015 · PubMed
31. Lang, R., Regester, A., Mulloy, A., Rispoli, M., & Botout, A. (2011). Behavioral intervention to treat selective mutism across multiple social situations and community settings. Journal of Applied Behavior Analysis, 44(3), 623–628. doi.org/10.1901/jaba.2011.44-623 · PubMed
32. Bork, P., & Bennett, S. (2020). Video self-modeling, fading, and reinforcement: An effective treatment package for children with selective mutism. Clinical Child Psychology and Psychiatry, 25(2), 446–455. doi.org/10.1177/1359104519855110 · PubMed
33. Johnson, M., & Wintgens, A. (2016). The selective mutism resource manual (2nd ed.). Speechmark. www.routledge.com/The-Selective-Mutism-Resource-Manual-2nd-Edition-2nd-Edition/Johnson-Wintgens/p/book/9781909301337

Spotted an error, or know a study we should add? Email hello@calalunakids.com. Every study we draw on is in the research library; research terms are explained in our methods glossary.