Structuring Shared Reading Over Video Chat to Promote Joint Attention and Engagement

• In Event: Unifying Neurobiological, Computational, Environmental, and Clinical Studies to Advance Developmental Research: A Multimethodological Approach

Abstract

Children frequently share social experiences with family members, teachers, and friends via video chat. However, establishing shared attention without physical proximity can be challenging, and video chat includes additional complexities (McClure & Barr, 2016). It can be difficult for partners to visually signal that attention is shared, or to what it is directed, because gaze is often misaligned. That is, a person appears to look directly at their partner when they look at the camera, not when they are looking at the person’s image on screen. This may limit the ability for people to follow one another’s attention to things in each others’ environments. It may also be difficult to follow children’s attention if they refer to objects they do not realize are beyond the view of the webcam (McClure & Barr, 2016). These issues complicate the establishment of shared reference necessary for engaging in joint visual attention (JVA), which may have ramifications for children’s engagement and opportunities for learning.
Using a repeated-measures experimental design, we investigated 3- to 5-year-olds’ (n = 38) joint attention and engagement during three structured video chat shared reading sessions. that were designed to promote different amounts and types of joint visual attention (Table 1). They read: 1) a single physical book which the researcher held up to the camera, 2) two physical books (the child and researcher each had a copy), and 3) a screen-share digital book. We tested whether our manipulation resulted in quantifiable differences in joint attention and children’s engagement in the task.
Using bootstrapped repeated-measures analyses appropriate for non-normal data, we found that although child initiations of joint visual attention were rare, there were significant condition differences (p = .021), with the most initiations occurring in the two-book condition (Table 2). When only the researcher had the book, all instances were across screen (child referring to researcher’s environment). In the two-book condition, only 13.3% were across screen, and the remaining 86.7% were within screen (referring to something in the child’s environment). Of these, 31.1% were to something out of view of the camera.
Child engagement showed a similar pattern to JVA initiations, with the most engaged behaviors (e.g., smiling, vocalizing) occurring in the two-book condition, although the bootstrapped p value was not significant (p = .057). The proportion of the session in which children attended to the screen was lowest in the two-book condition (p = .003) but the proportion of time on task was high and did not differ across conditions (p = .611).
Our results indicated that the frequency and type of joint attention children initiated varied by condition. Only when children had their own copy of the book did they initiate joint attention in their partner’s environment and their own, although many within-screen initiations referred to things out of view of the camera. These ‘errors’ did not appear to substantially impact children’s engagement or time spent on task, suggesting that the two-book structure may be best for providing opportunities to learn about sharing attention both within and across screen.

Group Authors

Gabrielle Strouse, [email protected], University of South Dakota
Lauren Myers, [email protected], Lafayette College
Taylor Stone, [email protected], Boston College
Ryan Vinzant, [email protected], University of South Dakota
Emily Clark, [email protected], University of South Dakota
Jessica Lehmen, [email protected], University of South Dakota

Authors

• Gabrielle A. Strouse, University of South Dakota
  Presenting Author

• Lauren J. Myers, Lafayette College

• Taylor Stone, Boston College

• Ryan Vinzant, University of South Dakota

• Emily Clark, University of South Dakota

• Jessica Lehmen, University of South Dakota