Real-Time AI Feedback to Enhance Teacher–Child Interactions: A Randomized Trial in Early Childhood Classrooms

How to cite

Royer, Rebecca; Suskind, Dana; List, John; Pernaudet, Julie; Sailopal, Ajay. 2025. "Real-Time AI Feedback to Enhance Teacher–Child Interactions: A Randomized Trial in Early Childhood Classrooms." AEA RCT Registry. AEARCTR-0016498. https://doi.org/10.1257/rct.16498-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Rebecca Royer* , Dana Suskind , John List , Julie Pernaudet , Ajay Sailopal

* Primary investigator

Registry ID
AEARCTR-0016498
DOI
https://doi.org/10.1257/rct.16498-1.0
First registered
Last updated
Status
In development
Trial dates
2025-08-11 – 2026-07-17
Countries
Private
JEL codes
C93, D83, I21, J13
Registry record
http://www.socialscienceregistry.org/trials/16498

Abstract

Socioeconomic disparities in child skills emerge rapidly during the critical years of brain development (0-5 years). Yet, research in the first few years of life is limited due to a lack of data from children's natural learning environments. In this study, we leverage Luet, an AI-driven wearable device paired with an app, to measure language engagements in children's natural early learning environments. We design a randomized controlled trial (RCT) with a sample of 30 daycare classrooms in which infants and toddlers wear Luet every school day for nine months. We randomly assign half of classrooms to the treatment group, which consists of early childhood curriculum, coaching, and real-time feedback to teachers about each child's language engagements in class. The other half of classrooms are assigned to a control group. In this pre-analysis plan, we outline the study design, research questions, and methods for this project.

Intervention

In our study design, consented children in all classrooms will wear Luet throughout the nine-month period, and all teachers will have access to an app that provides real-time and historical data on each child's Luet usage. Each classroom will be assigned to either a physical development group or to a language development group. In the physical development group, teachers will receive information about physical development in early childhood. In the language development group, teachers will receive an intervention consisting of coaching, app feedback about language inputs, and an early childhood education curriculum.

Experimental design

Our primary research questions focus on how our intervention moves beliefs and behavior through the lens of the theory of change. To do so, we test for treatment effects first on teacher beliefs about child development, then on teacher behavior measured through teacher-child interactions, and finally on child outcomes. Secondary research questions include heterogeneity of treatment effects by classroom and child characteristics, changes to disparities in child development, changes in children's skills measured through other algorithms applied to the audio data, and spillovers to parental knowledge and investment.

Design details

Randomization method
We will use a rerandomization procedure (by a computer) combined with a staggered design.
Randomization unit
Randomization is at the daycare center. We anticipate each center will have 1-3 classrooms in the study.
Clusters
Estimated 16 centers and 30 classrooms.
Observations
Estimated 260 children in the study.
Arms
8 centers in treatment and 8 centers in control (15 classrooms in each group).
Primary outcomes
Our key outcomes consist of three categories: teacher beliefs (measured through SPEAK-CAT scores), teacher-child interactions (measured through conversational turn counts at the child level), and child outcomes (measured through the NIH Baby Toolbox and ROWPVT).
Secondary outcomes
Secondary outcomes for teacher beliefs include teacher beliefs about growth mindset, beliefs about AI, and teacher reports about their relationships with the children in class. Secondary outcomes for teacher-child interactions include adult word count measured at the child level. Secondary outcomes for child outcomes include MacArthur-Bates Communicative Development Inventories, ASQ:SE, and Woodcock Johnson-IV. Long term outcomes include K-12 academic and disciplinary records as well.

Cited by

No citation data available for this registration yet.

Keywords