Encouraging Toddlers with ASD to Request: An Exploration of Expectant Pausing and Engagement Strategies

Background: Deficits in requesting are an early social communication risk factor for ASD in toddlers (Barbaro & Dissanayake, 2009) and are successfully targeted through intervention (Leung, 1994). Evidence-based interventions address a number of social communication deficits. However, little is known about how specific interactional strategies (i.e., ways interventionists interact with children to elicit or respond to behavior within an intervention session) function individually, and in combination, on requesting.

Objectives: The current study evaluated examiner use of two levels of two interactional strategies (i.e., positive affect, expectant pausing) on child requesting. Positive affect was defined as examiner use of animated facial expressions and expressive vocal intonation (levels: consistently positive affect, low affect with contingent positive affect). Expectant pausing was defined as the amount of time between an examiner’s prompt before s/he delivered the next prompt on a hierarchy of increasing support (levels: 2 seconds, 7 seconds). Child interest (none, moderate, high) was examined as a potential moderator of child requesting. Child expressive language and ASD symptom severity were considered as covariates of child requesting.

Methods: As part of a larger screening study, toddlers were recruited and randomized to one of four conditions (positive affect by expectant pausing) following confirmation of an ASD diagnosis made with gold-standard assessment tools and clinical judgment (n=58, M=28.24 months [4.61]). Within the assigned condition, unfamiliar examiners led children through a series of 5 hierarchical prompts in an attempt to elicit requesting.

Results: Generalized estimating equation (GEE) analyses revealed a significant three-way interaction between positive affect, child interest in materials, and trial number when controlling for expressive language (See Figure 1). In high child interest trials in which children were randomized to the consistently high engagement versus the low with contingent high engagement condition, children were more likely to request in earlier trials than in later trials. In contrast, in medium child interest trials, in which children were randomized to the low with contingent high engagement versus the consistently high engagement condition, children made more requests independent of trial number. Low child interest trials rarely resulted in a request, irrespective of engagement condition or trial number. Pause length was not associated with requesting. Children with higher levels of expressive language were more likely to request.

Conclusions: A major finding to emerge from this study is that a child’s level of interest in materials is an important consideration when attempting to elicit a request. This study provides initial evidence that children with varying levels of interest in materials may respond better to different levels of clinician use of positive affect (i.e., consistently high, versus low with contingent high). Further, beyond child interest in materials and clinician engagement condition, child developmental factors matter. Children with higher levels of expressive language were more likely to request. Additionally, the present study demonstrates a novel method for evaluating potential active ingredients of intervention in an experimental manner. Further exploration of individual interactional strategies may lead to both more personalized interventions for toddlers with ASD and more directive guidelines for clinicians.