An Experimental Comparison of Cognitive Forcing Functions for Execution Plans in AI-Assisted Writing: Effects On Trust, Overreliance, and Perceived Critical Thinking

Ahana Ghosh, Advait Sarkar, Siân Lindley, Christian Poelitz · 2026

Read An Experimental Comparison of Cognitive Forcing Functions for Execution Plans in AI-Assisted Writing: Effects On Trust, Overreliance, and Perceived Critical Thinking by Ahana Ghosh, Advait Sarkar, Siân Lindley, Christian Poelitz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-18033.txt.

Introduction

Ahana Ghosh, Advait Sarkar, Siân Lindley, and Christian Poelitz test designs that make AI users think before they accept help. Their 2026 preprint studies cognitive forcing functions in AI-assisted writing. Generative tools speed up writing and coding, but they also invite overreliance and lazy thinking. Forcing functions are design moves that make a user engage with the task to proceed. The team runs an experiment on such moves built into execution plans.

They measure effects on trust, overreliance, and how useful people find the tool. The trade-offs are real: friction can guard thought but also annoy. The results guide designers who want AI help without dull users. Keeping human judgment awake is a core design problem of the AI era.

Catalog record

CREATORAhana Ghosh, Advait Sarkar, Siân Lindley, Christian Poelitz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE189 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-18033

Browse the full library · About the Tower of Babel