What is it about?
Many areas of psychology lack well-developed theories that can make precise predictions. This paper provides a step-by-step guide for building such theories using probabilistic network models: mathematical tools originally from physics that describe how interconnected elements (like symptoms, beliefs, or cognitive abilities) influence each other. We present six starting-point models that researchers can choose from depending on their topic, along with free software tools for exploring how these models behave. We also show how researcher can extend and adapt the models to their context. As an example, we show how to build a model explaining why some people can stay neutral on issues they care deeply about (like judges deliberating a case) something existing attitude theories couldn't account for.
Featured Image
Photo by Alina Grubnyak on Unsplash
Why is it important?
Psychology faces atheory crisis: many of its core constructs lack formal, testable theories. Probabilistic network models have already produced promising theories of attitudes, depression, and intelligence, but until now there has been no systematic guide for developing them. This paper fills that gap by synthesizing models scattered across physics, biology, and social science into an accessible framework with practical tools.
Perspectives
Writing this paper was a great but rewarding challenge, as it doesn't fit into standard article formats. We had to invent a new format to make technical mathematical models useful for a broad audience.
Adam Finnemann
Universiteit van Amsterdam
Read the Original
This page is a summary of: A theory-construction methodology for network theories in psychology., Psychological Methods, April 2026, American Psychological Association (APA),
DOI: 10.1037/met0000829.
You can read the full text:
Contributors
The following have contributed to this page
