Promoting Joint Attention for Toddlers with Early Indicators of Autism: a Parent-Mediated Approach

Background: Joint attention has been demonstrated to be a foundational competency that supports later development of higher level social-communication skills; however, its development is disordered in autism. Intervention models that promote joint attention are needed for implementation during the toddler period when intervention can be most effective. Also needed are early intervention models that follow developmentally appropriate family-centered approaches for toddlers consistent with recommended early intervention practice.  For toddlers with autism who have difficulty with interaction, primary caregivers may have the best chance of eliciting early nonverbal communication successfully. A three-site research project has implemented a manualized intervention, the Joint Attention Mediated Learning (JAML) model that responds to these needs.

Objectives: When performance during and following the intervention period is compared to baseline performance, participants will improve in (a) focusing on faces, (b) turn-taking, (c) responding to joint attention, and (d) initiating joint attention.

Methods: In the first two years of a three-year project, 22 participants participated in the parent-mediated four-phase JAML intervention in a single case experimental research design using a multiple baseline probe across targeted outcomes. Participants entered new phases of intervention as they demonstrated proficiency in preceding phases, showing stable performance above a baseline trend line that was used to predict future performance.  Researcher-constructed measures for each targeted outcome were assessed and independently coded from 10-minute video-recordings taken at weekly intervention sessions. The four phases of intervention corresponded to the four targeted outcomes: focusing on faces, turn-taking, responding to joint attention, and initiating joint attention. In a series of systematic paired replications, the study explored variations in frequency of intervention sessions, parents’ use of audio-recordings to report child progress, and the use of video modeling and reflection with parents. In addition, the intervention was implemented with one Spanish-speaking and additional bilingual families. Intervention was provided in weekly sessions in which trained interventionists provided individual guidance to parents in their homes. Parents carried out planned and naturally occurring interaction-based activities daily and reported each week on child participation. An intervention manual guided intervention activities and a parent manual presented mediated learning principles and descriptions and examples of activities to promote each of the four targeted outcomes.

Results: To date, across the three sites two participants have progressed through the four phases of intervention (i.e., achieved the four targeted outcomes), seven completed the first two phases, and six completed the first phase. Remaining participants are in the baseline condition. We expect the majority of participants to have ended the intervention before the IMFAR meeting in May, when updated data will be presented.  Of those participants who will have completed the intervention we expect that the majority will have achieved all four targeted outcomes.

Conclusions: Experimental results show that joint attention can be achieved by toddlers with early signs of autism using a naturalistic parent-mediated intervention model in which primary caregivers build on their intimate knowledge of their children’s preferences for activities and materials to interact with their toddlers in developmentally sequenced stages designed to build toward joint attention.