N-gram uptake analysis for human–AI co-writing

Research Method · All research methods
Evidence · 2 claims (2 for) · 1 study (1 associational), q3 · 0 of 1 report an effect size · 2 claims rest on one study

Description

The article proposes analyzing textual uptake in AI-supported writing via recurring n-grams between ChatGPT output and the writer's final text, extending Bereiter and Scardamalia's knowledge telling/knowledge transforming distinction to human–AI interaction. The authors write that "Overlaps in n-grams can thus serve as indicators of uptake" and that high verbatim overlap reflects adoption "analogous to knowledge-telling activities" while reformulated use points toward knowledge transforming. The three clusters are interpreted as positions on a spectrum from passive uptake to increasingly active and integrative collaboration, aligned with the ICAP framework.

Accounts

• N-gram uptake analysis: mapping human–AI co-writing onto knowledge telling and knowledge transforming: The article proposes analyzing textual uptake in AI-supported writing via recurring n-grams between ChatGPT output and the writer's final text, extending Bereiter and Scardamalia's knowledge telling/knowledge transforming distinction to human–AI interaction. The authors write that "Overlaps in n-grams can thus serve as indicators of uptake" and that high verbatim overlap reflects adoption "analogous to knowledge-telling activities" while reformulated use points toward knowledge transforming. The three clusters are interpreted as positions on a spectrum from passive uptake to increasingly active and integrative collaboration, aligned with the ICAP framework. (Teresa Luther · Kate Belcher · Joachim Kimmerle · Detmar Meurers · Ulrike Cress (2026))

Claims

• AI-supported argumentative writing yields three distinct uptake clusters (text reproducers, text integrators, text reconstructors) based on recurring n-gram overlap [+M]
• Text reproducers copied nearly all of their final text from ChatGPT with long verbatim sequences, while text integrators showed fragment-level micro-uptake [+M]

Related Research Methods

-

Key Sources

• Teresa Luther · Kate Belcher · Joachim Kimmerle · Detmar Meurers · Ulrike Cress (2026). Collaborative knowledge construction with generative AI: Exploring argumentative co-writing processes through n-gram and cluster analysis. Intern. J. Comput.-Support. Collab. Learn. https://doi.org/10.1007/s11412-026-09470-1

Page metadata

Field	Value
type	research-method
id	n-gram-uptake-analysis-for-humanai-co-writing
title	N-gram uptake analysis for human–AI co-writing
description	The article proposes analyzing textual uptake in AI-supported writing via recurring n-grams between ChatGPT output and the writer's final text, extending Bereiter and Scardamalia's knowledge telling/knowledge transforming distinction to human–AI interaction.
status	review
generated	by process:settle-candidates on 2026-10-18
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.