Determining Appropriateness of a Mobile Technology Application to Treat Prosodic Deficits in ASD

Background: For those individuals with ASD who acquire spoken language, approximately 50–80% exhibit prosodic deficits including monotone or sing-song speech quality (Baltaxe & Simmons, 1985). Despite the limiting nature of these deficits, there are few available interventions (Diehl & Paul, 2009). The emergence of mobile technology provides speech-language pathologists (SLPs) with another medium to deliver intervention. While most SLPs use iPads for treatment (Fernandes, 2011), little is known regarding who is best suitable for technology-based interventions. 

Objectives: The purpose of this study is to 1) characterize the speech and language skills of those with ASD for whom SpeechPrompts, an application for iPad to treat prosodic disorders, might be suitable and 2) evaluate the application features that are most useful for clinicians. 

Methods: Students, 5-15 years of age (M = 8.80, SD = 3.16) with a diagnosis of ASD and prosody deficits, were enrolled (N = 11). Students were recruited from school-based speech-language pathologists (SLP; N = 5). Students’ prosody was assigned a global prosody rating pre-treatment. SpeechPrompts includes four main components with each targeting a different prosodic construct (Intensity, Stress, Rate, and Pitch). SLPs used the software for 16 weeks, with each of their recruited students, based on clinical judgment. A student engagement rating was assigned at the conclusion of each session.  

Results: Students received an average of 14 software sessions (SD = 9.79). Number of sessions was highly positively correlated with students’ social dysfunction as measured by the Social Responsiveness Scale-2 (Constantino, 2012) r = 0.69, p = 0.02. SLPs used the software with students whose verbal IQ standard scores ranged from 72-99 (M = 86.8; SD = 11.2). The software was used equally with students whose prosody was rated as monotone (n = 5; 45%) or sing-song (n = 6; 54%). Student engagement ratings did not differ between the monotone (M = 1.90, SD = 0.21) or sing-song (M = 1.97; SD = 0.40) groups. No significant correlation was found between prosody ratings and engagement (r = .12; p = 0.72). The majority of SLPs (80%) rated the main application components as useful to extremely useful with the exception of the Rate component.  The components that addressed pitch (36.7%) and stress (35.3%) were used most frequently while intensity (24.3%) and rate (3.7%) were used less frequently.

Conclusions: Results of this pilot study suggest that SLPs used the software primarily with high-functioning students of low average verbal IQ. The application appears to be appropriate for students who present with flat prosody as well as those who present with exaggerated speech prosody. Students with monotone prosody and students with sing-song prosody were highly engaged while using the software. SLPs rated the application from useful to extremely useful for the majority of features and used visualization of pitch curves and waveforms most frequently with their students. Although further efficacy testing is necessary, this preliminary study suggests that SpeechPrompts might be an engaging application for high-functioning students with ASD who present with either flat or exaggerated speech prosody.