Freespeech: Large-Scale Data from a New AAC Application Characterizes Usage for Young Children with ASD – One Size Does Not Fit All

Friday, May 15, 2015: 11:30 AM-1:30 PM
Imperial Ballroom (Grand America Hotel)
L. Boccanfuso1, J. C. Snider2, E. Schoen Simmons3, M. C. Lyons1, E. S. Kim2, C. A. Wall3, L. Whitaker4, M. Perlmutter2, K. Konwar5, R. Schrock6 and F. Shic7, (1)Yale University, New Haven, CT, (2)Child Study Center, Yale University School of Medicine, New Haven, CT, (3)Child Study Center, Yale University, New Haven, CT, (4)Yale University School of Medicine, New Haven, CT, (5)University of British Columbia, Vancouver, BC, Canada, (6)Keysight Technologies, Inc, Santa Rosa, CA, (7)Yale Child Study Center, Yale University School of Medicine, New Haven, CT
Background:  A recent search for Android and Apple autism applications (apps) yields approximately 1802 results and a search for alternative and augmentative communication (AAC) apps returns 553 hits.  The current proliferation of apps designed to help children with autism has largely stemmed from research and pragmatic observations that children with ASD tend to be motivated by computer apps.  However, very little research exists which methodically identifies the features of AAC apps that are most useful for this population and, as importantly, provides a definitive characterization of how very young children with ASD tend to use AACs most.

Objectives:  This evaluative study represents one of the largest usage-based statistical surveys of a computer app designed to augment and promote communication.  User and usage data collected over a period of approximately 2 years was analyzed to develop a characterization of usage patterns to deliver an evidence-based model of users and their activities.  We classify usage between users with ASD and users with other communication impairments (e.g., dysarthria, apraxia, etc.), analyze comparative data for usage of young children (0-5 years) with ASD and their same-aged counterparts with other communication impairments and provide an illustrative case study of young children with ASD at an individual level using massive data.

Methods:  We conducted a comprehensive analysis to characterize word selection frequency, breadth of vocabulary used, and frequency of functional and social language usage across a large set of FreeSpeech users.  Data from over 2.7 million events, representing more than 6,000 anonymous users was examined to determine general usage patterns for specific app features provided.  Further, member events recorded for over 230 individual users, who provided information pertaining to age, gender, medical diagnosis, speech and language disorder, were evaluated to investigate common usage traits among children with ASD and distinguishable linguistic preferences between young children with ASD and other age-matched children without ASD.

Results:  Highest frequency content was almost equally divided between functional and social words (“I want” and “hello”, respectively).  Although young children with ASD comprise a relatively small percentage of users who completed the survey (<10%) this age group represents approximately 23% of all events, averaging 3,134 events per user compared to 448 for other groups.  Among frequent users with ASD, significantly fewer unique words and a higher rate of repetition (27.95 to 7.27) was indicated compared to their age-matched non-ASD counterparts.  However, case studies of longer-term users (>6 months and >15,000 events) reveal that usage frequency increases and content is often supplemented with proper names, specialized activities and feelings after an introductory period.

Conclusions:  Outcomes from this study provide evidence that young children with ASD using AACs may have characteristically different usage needs compared to their age-matched counterparts with other communication disorders.  Significant changes in usage trends are also apparent in young users, providing evidence that AAC apps fulfill diverse roles.  This preliminary study motivates further investigation into how AAC apps are used by young children with ASD and offers valuable insight for designing improved apps which are most beneficial to this population.