Sentiments Toward AI Assistance

How to cite

Avoyan, Ala; Dedyukhin, Ivan. 2025. "Sentiments Toward AI Assistance." AEA RCT Registry. AEARCTR-0017375. https://doi.org/10.1257/rct.17375-1.0

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

Ala Avoyan* , Ivan Dedyukhin

* Primary investigator

Registry ID
AEARCTR-0017375
DOI
https://doi.org/10.1257/rct.17375-1.0
First registered
Last updated
Status
In development
Trial dates
2025-12-11 – 2026-05-10
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/17375

Abstract

This study investigates how people perceive the strategic decision-making abilities of others who use AI tools. We are particularly interested in whether people think AI tools can help less experienced individuals perform as well as experts, and how these beliefs vary based on the evaluator's own age, education, and familiarity with AI.

Intervention

Participants are randomly assigned to play a strategic game against one of four opponent types that vary in their characteristics and available resources. The opponent types differ along two dimensions that we communicate to participants before they make their decisions. Participants make choices in the game knowing their opponent's type but without direct interaction. The game structure incentivizes participants to consider their opponent's likely strategic sophistication when making their own decisions.

Experimental design

This is a between subject experiment where participants make a single strategic decision in a game. Participants are randomly assigned to face one of several opponent types and are informed about their opponent's characteristics before making their choice.

Design details

Randomization method
Computer based randomization using random number generation.
Randomization unit
Individual participant level.
Clusters
Not applicable
Observations
400
Arms
About 100 participants in each of the 4 treatments (this does not count lab subjects).
Primary outcomes
The choice in the game, response times, their cognitive uncertainty measures.
Secondary outcomes
Cognitive Ability, Familiarity with AI, AI use with the task, Risk and Uncertainty attitudes, Demographic characteristics.

Cited by

No citation data available for this registration yet.

Keywords