Results of a Group Comparison Study of the Joint Attention Mediated Learning Early Intervention for Toddlers with Autism Spectrum Disorders

Background:

Findings from a small randomized group study of the Joint Attention Mediated Learning (JAML) intervention for toddlers with high risk for autism spectrum disorders (ASD) are presented.  JAML is a developmentally ordered parent-mediated intervention to promote preverbal social communication for toddlers with ASD.  This research was funded by Autism Speaks.

Objectives:

To conduct a preliminary assessment of JAML’s effects on toddlers’ demonstration of joint attention in parent-child interaction and on receptive and expressive language measures.

Methods:  

Participants: 24 parents and their toddlers with high risk for ASD were recruited with eligibility determined by failure on the M-CHAT questionnaire and follow-up interview.

Research procedures: Participants were evenly divided between randomly assigned intervention and control conditions.  Eleven intervention and twelve control participants completed the study.  Assessment data and video of parent-child interaction were gathered in three pre-intervention and three post-intervention sessions and additional video data were collected in 4- and 8-week follow-up sessions.    

Intervention procedures: Intervention Coordinators met weekly with parents to provide guidance on promoting the current targeted outcome focusing on faces (FF), turn-taking (TT), responding to joint attention (RJA), or initiating joint attention IJA), in order) through five mediated learning principles: focusing, organizing and planning, encouraging, giving meaning, and expanding.  Parents were asked to facilitate daily planned and incidental JAML-focused interaction with their toddlers.  Videos of parent-child interaction, recorded weekly, were used for parent reflection at weekly sessions.  New targeted outcomes were illustrated with video examples and parents reviewed print materials and video examples of other parents promoting the targeted outcome using one of five mediated learning principles.

Results:  

A hierarchical linear model (HLM) was used to estimate the effects of change over time and intervention between the randomly assigned intervention and control groups.  Measures were the number of observations for each of the four targeted outcomes: FF, TT, RJA, and IJA from 10-minute videos of parent-child interaction taken from the pre-intervention, post-intervention, and follow-up sessions.  A strong response to the intervention was shown for FF and RJA.  Group differences for TT and IJA in this small sample were not statistically significant.    

An HLM analysis of the Mullen and Vineland assessments showed interaction effects for the Vineland Expressive V-Score by time and treatment and the Mullen Receptive Language by treatment and time, (all F > 4.6, all p<.05).  In all cases the change in the treatment group was significant, but the change in the control group was not.

Conclusions:  

Findings support the viability of the JAML intervention for toddlers with early signs of ASD and their families.  The intervention builds on theoretical precursors (FF and TT) to promote engagement in joint attention, a known precursor of verbal forms of social communication.   JAML promotes preverbal social communication learning from within the parent-child relationship and is integrated within planned and routine activities, targeting approximately 30 minutes of focused interaction daily.  It requires professional and parent conceptual clarity on JAML’s content (current phase of intervention) and process (mediated learning) components, which are defined and illustrated in JAML materials.