Computer-based receptive language training for young children with Autism
Laura Schreibman, Ph.D., University of California, San Diego
Developing modes of communication and speech in young children with autism is a primary focus of many interventions, but the development of receptive language abilities has generally received far less targeted attention. To address this area of need, Laura Schreibman, Ph.D. and her colleagues, Aubyn Stahmer, Ph.D., and Joseph McCleery, M.A., at the University of California, San Diego Autism Research Laboratory have received a grant from Cure Autism Now to investigate the clinical efficacy of a computer-based training program for teaching word comprehension skills to young, non-verbal and minimally verbal children with autism.
In typical infant development, the advent of joint attention (the ability to coordinate attention between an object and a person in a social context) precedes the acquisition of receptive language (learning the names of objects). It is theorized that the noted difficulties with joint attending in autism serve to limit vocabulary development. Learning language in a natural environment requires that a child must be attending to an object at the same time as he hears its label - i.e., see mom point to the ball, hear "ball" and understand that the round thing rolling across the floor is a "ball." If a child has difficulty with joint attention and does not look where mom is pointing, he or she may never learn to associate the object with the word. Although many interventions are being developed to teach joint attention skills to children with autism, Dr. Schreibman's laboratory is aiming to teach receptive language by designing a method that bypasses the need for joint attention.
Dr. Schreibman's team plans to teach receptive labels using a computer training approach that has previously been shown to effectively teach typically-developing infants. The computer program presents repeated pictures paired with their auditory labels. It has recently been demonstrated that this simple technique can teach typically-developing infants of 18 months new labels after just 12 trials of picture/label pairings. This approach offers the benefit of rapid, multiple trials and repetition (as in discrete trial training) but use of the computer limits the need for joint attention with another person. The grant from Cure Autism Now will allow Dr. Schreibman's laboratory to examine whether this is an effective method to teach infants with autism as well.
It is expected that removing the social context of the therapy setting may improve learning and increase attention to the task. Since word comprehension skills have been shown to be necessary for the development of later vocabulary, grammar and cognition in typically-developing children, the team also predicts that gains in receptive vocabulary will result in pivotal gains in these later-developing, higher-level skills. If the early trials are successful, the team plans to extend the research and hopes that this program might eventually offer an effective training opportunity for use in home and community settings. As Dr. Schreibman explains, "Our ultimate goal with this research is to understand more about how and what these children learn. We do not want to replace treatments based on social interaction but we expect that knowledge gathered from this research will allow us to develop more effective treatments for these children."
In addition to developing a possible treatment program, this project, under the supervision of Rita Ceponiene, M.D., will investigate the use of event-related EEG potentials (ERPs) as a possible measure of assessing learning of receptive language. ERPs are a non-invasive technique used in the study of human brain function-electrodes placed on the scalp monitor the electrical activity generated by neuronal responses to a stimulus. The brain waves produced by these responses can be compared to show evidence of differences in the timing, sequence, amount and general location of brain activity. Preliminary data show patterns that indicate that object recognition, word recognition and the integration of the two (object label) can be exhibited through ERP testing. If validated as a measure of assessing receptive language, most importantly this ERP technique would allow researchers a way of measuring treatment success in children who, because of motor or social challenges, simply may not be able to indicate what they have learned through traditional behavioral measures usually employed (point to..., touch the...) In doing so, the grant reviewers noted that this project will add valuable information to the educational field by providing insight into the developmental course of word processing in both typically-developing and autistic children.