---
title: "Modelling Experiments in Scientific Discovery"
authors:
  - "Peter C.-H. Cheng"
doi: ""
url: "https://ijcai.org/proceedings/1991/108"
pdf: "https://ijcai.org/proceedings/1991/0108.pdf"
venue: "Proceedings of the Twelfth International Joint Conference on Artificial Intelligence"
year: 1991
pages: "739-744"
track: "Learning Be Knowledge Acquisition"
keywords:
---

# Modelling Experiments in Scientific Discovery

Peter C.-H. Cheng

*Proceedings of the Twelfth International Joint Conference on Artificial Intelligence*, pages 739-744.

## Abstract

Investigating the character of scientific discovery using computational models is a growing area in Artificial Intelligence and Cognitive Science. Scientific discovery involves both theory and experiments, but existing discovery systems have mainly considered the formation and modification of theories. This paper focuses on the modelling of experiments. A general characterization of the nature of experiments is given and more specifically Galileo's motion experiments are examined. The SlERN scientific discovery system has been used to model Galileo's investigations of free fall, and is introduced here. The system has an extensive representation for experiments and uses experiments to: (i) confirm existing hypotheses; (ii) find new hypotheses; (ii) enhance its own performance; and, (iv) make intractable hypotheses tractable.