Understanding the Division of Cognitive Effort in Human-AI Interactions

Read full text Summarise with VIRAT
Authors: Sunny Yu, Ilia Sucholutsky, Katherine M. Collins, Anonymous , Dan Jurafsky, Robert Hawkins, Ahmad Jabbar
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

As LLMs are increasingly treated as thought partners that collaborate with users on different cognitive tasks, it is important to understand the division of labor in this interaction: how much cognitive effort is a user offloading to such an AI-based thought partner in each interaction? What resource – time or effort – is conserved when people use AI to complete a task? We conduct a large-scale study to examine the systematic difference between completing a task independently and completing it with AI assistance across different cognitive tasks. We compare the time and effort participants predict a task would take to the actual time effort and hypothesize that a calibration error persists in how much offloading AI use affords: users systematically underestimate the time it takes to complete a task with AI assistance but are well calibrated about effort; when making predictions about completing a task independently, people are well-calibrated about time but overestimate effort.

Fields of Study & Subjects

Human-Automation Interaction and Safety Usability and User Interface Design Personal Information Management and User Behavior

Publication details

Publication date
1 January 2026
Language
EN
Document type
Preprint