Enhancing generalisation in biofeedback intervention using the challenge point framework: a case study

Abstract

Biofeedback intervention can help children achieve correct production of a treatment-resistant error sound, but generalisation is often limited. This case study suggests that generalisation can be enhanced when biofeedback intervention is structured in accordance with a "challenge point" framework for speech-motor learning. The participant was an 11-year-old with residual /r/ misarticulation who had previously attained correct /r/ production through a structured course of ultrasound biofeedback treatment but did not generalise these gains beyond the word level. Treatment difficulty was adjusted in an adaptive manner following predetermined criteria for advancing, maintaining, or moving back a level in a multidimensional hierarchy of functional task complexity. The participant achieved and maintained virtually 100% accuracy in producing /r/ at both word and sentence levels. These preliminary results support the efficacy of a semi-structured implementation of the challenge point framework as a means of achieving generalisation and maintenance of treatment gains.

Keywords: Biofeedback; challenge-point framework; generalisation; principles of motor learning; speech sound disorder.

Conflict of interest statement

Declaration of interest

This project was supported by NIH R03DC012883 to Tara McAllister Byun. Aspects of this research were presented at the annual convention of the American Speech-Language Hearing Association in Chicago (2013).

Figures

Figure 1

Lilianne’s accuracy in producing untreated /r/ words elicited without feedback in a previous biofeedback treatment study (McAllister Byun, Hitchcock, & Swartz, 2014). X-axis represents probe elicitation session (BL = baseline, TX = within-treatment, MN = maintenance). Y-axis represents per cent of tokens rated perceptually correct based on mode across three blinded clinician listeners.

Figure 2

Lilianne’s accuracy in producing untreated /r/ words elicited without biofeedback in the current study. X-axis represents probe elicitation session (BL = baseline, TX = within-treatment, MN = maintenance). Y-axis represents per cent of tokens rated perceptually correct based on mode across three blinded clinician listeners.

Figure 3

Lilianne’s accuracy in producing /r/ in untreated sentences elicited without biofeedback. X-axis represents probe elicitation session (BL = baseline, TX = within-treatment, MN = maintenance). Y-axis represents per cent of /r/ words in sentence-level utterances rated perceptually correct based on mode across three blinded clinician listeners.