Video self-modeling improves self-perception after speech restructuring

Background

Maintaining gains after intensive speech restructuring programs is a well-known challenge for people who stutter. Relapse is common, and many people find that confidence built during a program fades when they return to everyday speaking situations. Video self-modeling - watching edited footage of oneself speaking well - had shown promise in other areas but had not been rigorously tested for stuttering.

What the researchers did

Eighty-nine adults and adolescents were recruited across multiple Australian and New Zealand sites: the Australian Stuttering Research Center, La Trobe University, Macquarie University, Royal Prince Alfred Hospital, University of Newcastle, and the Stuttering Treatment and Research Trust Auckland. Eligible participants had to have at least 2% syllables stuttered at baseline, sufficient English, stuttering onset before age 12, no psychogenic onset, and no treatment in the previous 6 months.

All participants had recently completed the Camperdown program (a modified 5-day intensive speech restructuring variant). They were then randomly assigned to either a video self-modeling (VSM) group or a standard maintenance group. Those in the VSM group received DVD videos of approximately 5 minutes, filmed on Day 4 of the intensive week, featuring three contexts: a 90-second monologue, a 90-second telephone call, and a 90-second conversation - each edited to show only fluent speech. Participants watched their VSM DVD daily for one month (4 weeks). Clinicians, fellow participants, and data analysts were all blinded to allocation; the randomizing researcher distributed sealed-envelope DVDs only. Outcomes were assessed at one month and six months, including objective stuttering frequency, self-rated severity, satisfaction with speech, and quality of life.

What they found

On the primary objective measure - percent syllables stuttered - the two groups did not differ at either follow-up point. However, the video self-modeling group showed meaningful advantages in how they perceived their own speech. They rated their worst recent stuttering as roughly 10 percentage points lower than controls, and reported about 20 percentage points greater satisfaction with their everyday speech. Speech naturalness was unaffected, indicating the intervention did not produce an unnatural speaking pattern.

Why this matters

This study highlights an important distinction between objective speech measures and a person’s lived experience. While the video did not change how often stuttering occurred, it changed how people felt about their communication. For many people who stutter, self-perception and confidence are at least as important as measurable fluency. The finding also laid groundwork for later VR-based self-modeling research, where immersive environments could make the experience of watching oneself speak even more compelling.

Limitations

The study relied on a specific speech restructuring program, so results may not generalize to other approaches. The video clips were two-dimensional and watched on a standard screen, limiting the sense of immersion. Longer-term follow-up beyond six months was not conducted.