Chronic Stress and Information Acquisition in Low-Income Individuals: A Randomized Field Experiment in Mumbai

How to cite

Deshpande, Aishwarya. 2026. "Chronic Stress and Information Acquisition in Low-Income Individuals: A Randomized Field Experiment in Mumbai." AEA RCT Registry. AEARCTR-0019100. https://doi.org/10.1257/rct.19100-1.0

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

Aishwarya Deshpande*

* Primary investigator

Registry ID
AEARCTR-0019100
DOI
https://doi.org/10.1257/rct.19100-1.0
First registered
Last updated
Status
On going
Trial dates
2026-07-03 – 2027-03-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/19100

Abstract

This pre-registration outlines a randomized controlled trial testing whether stress constrains information acquisition among 1,200 low-income parents of kindergarteners in Mumbai. The experiment implements a 2x2 factorial design: parents receive either (i) a stress-relief breathing routine (with a tactile bracelet cue) or a bracelet-only control condition, and (ii) daily curiosity prompts that vary in signal precision---high-precision prompts elicit bounded, verifiable responses (e.g., counting, binary choices), while low-precision prompts require open-ended interpretation (e.g., ``why'' or ``how'' questions). This design isolates whether stress disproportionately impairs the processing of ambiguous signals. I pre-register four confirmatory hypotheses on emotional regulation, belief updating accuracy, engagement and habit formation, and a focal Stress x Precision interaction testing whether stress reduction helps parents extract value from harder-to-interpret information. The plan specifies sampling (N=1,200 across ~20 schools), interventions (delivered via tactile bracelet cues and physical card decks), timeline (24 weeks with midline at Week 12), power calculations (accounting for 20-25% attrition), and intent-to-treat estimation using ANCOVA with pre-specified indices, mechanisms, heterogeneity tests, and treatment-on-the-treated estimates. Multiple-testing corrections, robustness checks, and attrition-bounding procedures are detailed. All curiosity activity prompts and their information-cost coding rules (Dimensions A--E) are included as appendices.

Experimental design

The experiment is a 2x2 design which cross randomizes a stress relief intervention with a signal precision intervention in a sample of approx 1200 low-income individuals in an urban Indian sample.

Design details

Randomization method
Stratified randomization done in an office by a computer via python code.
Randomization unit
The unit of randomization is the individual (the parent).
Clusters
Approximately 20 schools.
Observations
1200 parents of kindergartner children.
Arms
300 parents in the stress relief + high precision information arm, 300 parents in the stress relief + low precision information arm, 300 parents in the placebo + high precision information arm and 300 parents in the placebo + low precision information arm.
Primary outcomes
Measurement done at week 24 is the primary endpoint.

Cited by

No citation data available for this registration yet.

Keywords