Unsupervised Machine Learning for Social Cognition

Using unsupervised machine learning to reveal hidden patterns in human beliefs and attitudes.

A tutorial and applied methods paper where I show how K-means, DBSCAN, PCA, and market basket analysis can reveal structure in large-scale attitude data.

Open paper Open PDF Back to projects

doi.org

From Data to Discovery: Unsupervised Machine Learning’s Role in Social Cognition

Social Cognition2025 | Published

From Data to Discovery: Unsupervised Machine Learning’s Role in Social Cognition

Social Cognition methods article with DOI and public PDF link.

A tutorial showing how unsupervised machine learning can help social cognition researchers discover hidden patterns in large-scale attitude and belief data.

Open paper

PublishedUnsupervised machine learning

How the method moved

From material to interpretation.

Starts with

large-scale attitude data, Project Implicit Race IAT worked examples

What the method does

compare possible clusters, dimensions, and co-occurring patterns

What comes out

possible clusters, dimensions, and co-occurring response patterns that researchers can interpret with theory.

What to read for

patterns are candidates for interpretation, not answers by themselves.

Open paper page Open PDF Publications

Case study

Inside this project.

A short, scannable view of the question, the design, my role, the method I chose, and where to stop short of overclaiming.

Question: How can social-cognition researchers find hidden structure in attitude data without deciding the answer in advance?
Data / design: Methods tutorial using Project Implicit Race IAT data and worked examples.
My role: Lead author.
Method / approach: K-means, DBSCAN, PCA, and market basket analysis, used as discovery tools that still need theory-based interpretation.
What it contributes: A tutorial and applied methods paper where I show how K-means, DBSCAN, PCA, and market basket analysis can reveal structure in large-scale attitude data.
Venue / status: Published, Social Cognition, 2025. Open paper
What this does not show: I treat clusters and dimensions as patterns to interpret with theory, not as answers by themselves.

Concrete details

Project Implicit Race IAT data
K-means, DBSCAN, PCA, and market basket analysis
Theory check between model output and social-cognition constructs
Published in Social Cognition