1 of 1

Exploratory Results

Do Interactive Educational Media Benefit All Children Equally?

The Role of Attentional Control in Computational Thinking Outcomes

Cam-Ly Pham, Jasmine Tran, Mark Warschauer

Digital Learning Lab

University of California, Irvine

Results

Limitations

Introduction

This research is based upon work supported by the National Science Foundation under Grants No. 1906321, No. 2115382, and No. 2415882, and by the UCI CP-LEADS under Grant No. R305B210006. Production of the conversational episodes is supported by the Corporation for Public Broadcasting. 

Acknowledgements

Methodology

Conclusions

Research Question

Does attentional control moderate the effect of interactive educational media on young children’s computational thinking?

  • 159 children ages 4–8 were randomly assigned to an interactive or non-interactive condition.
  • Children watched three episodes of Lyla in the Loop, a PBS KIDS show designed to teach computational thinking, one episode per session.
  • Attentional control was measured using the parent-reported Strengths and Weaknesses of ADHD Symptoms and Normal Behavior (SWAN) scale, ranging from -3 to 3 (M = 0.65, SD = 0.92).
  • We conducted an analysis examining whether attentional control moderated the effect of condition on computational thinking outcomes.
  • Because outcomes did not depend on a child’s attentional profile, educational technology can be designed to be broadly effective without requiring extensive individual personalization based on attentional control.
  • Replication with a larger and more diverse sample would strengthen confidence in the generalizability of these findings.
  • Future research could specifically recruit children with a wider range of attentional abilities, including those with clinical ADHD diagnoses, to see if interactive educational media provides them with a unique advantage.
  • Unlike traditional passive media, AI-driven conversational agents in educational shows prompt children to respond verbally, enhancing engagement and learning outcomes (Xu et al., 2024).
  • Research suggests children with attention difficulties show more sustained focus in structured, engaging digital environments, indicating interactive media may be especially beneficial for children with diverse attentional profiles (Mohammadhasani et al., 2018).
  • It is important to evaluate whether educational interventions equitably support children with diverse learning profiles to ensure that digital tools serve all students effectively.
  • The SWAN scale relies on parent reports, and parents may be less likely to report lower attentional control behaviors, or these reports may not perfectly reflect a child's focus during media consumption.
  • The sample skewed toward above-average attentional control (M = 0.65), restricting the range of scores and potentially limiting the ability to detect moderation effects among children with lower attentional control.
  • With a sample of N = 159, the study may not have been large enough to detect very small differences in how attentional control influences learning.

We found that attentional control did not significantly moderate the effect of interactivity on computational thinking outcomes.

Measure

Low Attentional

(n = 26)

High Attentional

(n = 29)

t

p

Response rate

.77 (0.23)

.79 (0.25)

0.33

.740

Response length

3.33 (1.95)

3.85 (2.63)

0.84

.402

Reprompt rate

.36 (0.24)

.36 (0.26)

-0.02

.984

  • To further explore the role of attentional control, we examined verbal engagement among children in the interactive condition, including response rate, response length, and reprompt rate.
  • Children with lower and higher attentional control did not differ significantly on any measure of verbal engagement, suggesting that attentional control did not meaningfully influence how children participated during interactive episode watching.

Note: This exploratory analysis utilized a smaller subsample (N = 55) compared to the full study (N = 159) due to missing data that prevented a full calculation of engagement metrics for all participants.

RQ