evidoria

← Back to browse

Good practice Imported

AI Custom Chatbots Cut Cognitive Load but Show No Clear Learning Gain — A University of Cologne Physics-Education RCT

Germany · Cologne · See the Germany profile

Evidence: Randomised controlled trial Top 89% 27/100 · Ask Evidence Copilot about this practice

A randomized study at the University of Cologne found AI-generated custom chatbot explanations in physics lessons significantly cut cognitive load and raised interest and self-efficacy, but found no clear advantage in objective learning outcomes over standard textbook material.

Details

Promoter
University of Cologne, Faculty of Mathematics and Natural Sciences
Period
2025
Region (NUTS)
DEA2
Keywords
secondary/tertiary education, physics education, generative AI chatbots

Context

Researchers in the Faculty of Mathematics and Natural Sciences at the University of Cologne, Germany, ran a randomized controlled study comparing AI-generated custom chatbot explanations against standard textbook material as supplementary support for physics learning.

Results

The AI-chatbot condition produced statistically significant increases in positive-activating emotions, situational interest and self-efficacy, and significant reductions in both intrinsic and extraneous cognitive load, compared with the textbook condition. However, the trial found no clear advantage for the chatbot condition on objective learning-outcome (test performance) measures.

Conclusions

The study is an honest example of an AI tool improving how students feel about learning without yet demonstrating it improves what they learn; it was conducted in a single research setting and does not address data-protection or teacher-training questions.

Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
University of Cologne, Faculty of Mathematics and Natural Sciences research team

Conditions for success

  • AI-generated chatbot explanations tailored to physics topics as a textbook supplement

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Replication kit

Reusable artefacts from this practice — as published by their sources.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful