Novel Mobile Technologies for Family-implemented Early Language Interventions in Augmentative and Alternative Communication

Abstract

Background: The proliferation of mobile technology has made a strong impact on early intervention efforts based on augmentative and alternative communication (AAC). AAC specialists are increasingly being asked to serve inclining numbers of children with developmental delay or disability who are younger than age 3. Early intervention has the potential to reduce severity levels of language and intellectual disability and is often associated with improved long-term intervention outcomes. The session will show a comprehensive intervention approach from targeting joint attention and social sharing skills to enhancing functional communication and parent-training. Early communication and language intervention with young children can be optimized when parents learn to capitalize upon the child’s strengths through the implementation of mobile AAC technologies during daily family routines. To this extent, a suite of evidence-based tablet applications, SPEAKall!, SPEAKmore!, and SPEAKplay!, was developed with funding from the National Institutes of Health to help minimally-speaking learners with developmental disabilities engage in functional communication, and acquire emerging language skills. As part of an open science approach, the app suite is available to the international AAC community at no cost and can be downloaded in most countries’ mobile app stores. A series of clinical studies seeks to document the impact of these novel technologies within a family-oriented early language program to facilitate communication and language development.

Method: Community uptake of the app suite was evaluated through a data analytics tool known as Google Analytics for Mobile Apps. Google Analytics is a programmable service that takes data on the use of any feature within the app. This instrumentation allows collecting clinically relevant usage data such as, for example, time spent using the app during treatments sessions, most frequently chosen graphic symbols, additional steps completed for message construction, number of messages communicated per session, and message complexity. In addition, empirical evaluation involved a practice-based evidence (PBE) approach through participatory, single-subject research designs to compare the frequency of target responses from baseline to intervention. For the current effort, a series of pre-experimental A-B designs (aka simple time series designs) were implemented. In the A-B design, a research-practitioner engages in repeated observations of target behavior in baseline until data are stable, followed by intervention. During intervention, data is collected again repeatedly on the target behavior, using an identical measurement system. Observed changes in the target behavior may be assumed to be a function of the independent variable (i.e., only correlational conclusions are possible). This design, although characterized by low internal validity and diminished confidence in conclusions, has practical value when more complex designs are impractical.

Particularly, a series of video cases of 10-15 min were recorded to target requesting, speech acquisition, and social interaction. Inter-observer agreement (IOA) for dependent measures (i.e., frequency of target behavior) and treatment implementation was established by two clinicians independently scoring 40% of all sessions. Mean IOA was 99% for requesting and social interaction, and 92% for emerging speech. Treatment integrity ranged from 80-98% across parents. Social validity of the technology-based interventions was assessed through a modified treatment acceptability rating form, administered with N=11 parents and clinicians. The survey elicited information on the perception of the app design and intervention outcomes asking questions about interface appeal, motivating elements, potential side effects, time needed to program, perceived skill improvement, etc.

Findings: Google Analytics data revealed consistent usage and improvement patterns across a period of six months, tracking more than N=16K learners. Particularly noteworthy findings involved increasing complexity of utterances, user loyalty, weekly usage patterns, and average time engaged in app activities. A practitioner-friendly effect size metric (i.e., Non-overlap of All Pairs) was calculated for three cases to estimate treatment effect; NAP values larger than 92% were interpreted as strong effects. In all three parent-child dyads infusing mobile AAC technologies lead to positive outcomes: NAP scores were 100% for requesting, 92% for emerging speech, and 90% for social interactions. The treatment acceptability survey revealed user rating means ranging from 4.4-4.9 on a 5-point Likert-type scale indicating general satisfaction.

Conclusions: A comprehensive approach of embedding AAC technology into proper family-centered instruction is key to strengthening the family’s capacity for moving their child from prelinguistic stages to early language learning. A critical element is jointly determining how to begin infusing technology into daily routines and coaching caregivers accordingly. Participatory research elements as part of language sampling and data tracking techniques can be used to document these efforts while making caregivers full partners in all aspects of assessment, intervention, and data collection. Using this approach enables collection and documentation of PBE in natural settings.