Paper Number

ICIS2026-3008

Paper Type

Completed

Abstract

Generative AI is becoming integrated into human cognition, yet the mechanisms shaping this cognitive coupling remain poorly understood. Using informed grounded theory, this paper introduces the Augmented Cognitive Extension (ACE) model, which explains how sustained generative AI use reorganizes cognition through five regulatory mechanisms. These are attentional synchronization, epistemic calibration, affective attunement, risk construal, and narrative preservation. The findings reveal a three-stage developmental trajectory from Engagement through Calibration to Integration, with Calibration serving as the critical regulatory window shaping long-term cognitive coupling. Each mechanism can produce productive cognitive extension or problematic dependency. This is the stabilization–destabilization spectrum, and the same user may show stabilizing patterns in some domains and destabilizing patterns in others. ACE provides an empirically grounded, mechanism-level account of human-AI interaction and challenges binary views of AI augmentation by showing that cognitive extension and cognitive erosion emerge from the same regulatory mechanisms.

Comments

13_Human_technology_interaction

Download

Share

COinS
 
Dec 14th, 12:00 AM

Generative AI and the Reorganization of Human Cognition: The Augmented Cognitive Extension Model

Generative AI is becoming integrated into human cognition, yet the mechanisms shaping this cognitive coupling remain poorly understood. Using informed grounded theory, this paper introduces the Augmented Cognitive Extension (ACE) model, which explains how sustained generative AI use reorganizes cognition through five regulatory mechanisms. These are attentional synchronization, epistemic calibration, affective attunement, risk construal, and narrative preservation. The findings reveal a three-stage developmental trajectory from Engagement through Calibration to Integration, with Calibration serving as the critical regulatory window shaping long-term cognitive coupling. Each mechanism can produce productive cognitive extension or problematic dependency. This is the stabilization–destabilization spectrum, and the same user may show stabilizing patterns in some domains and destabilizing patterns in others. ACE provides an empirically grounded, mechanism-level account of human-AI interaction and challenges binary views of AI augmentation by showing that cognitive extension and cognitive erosion emerge from the same regulatory mechanisms.