Open Loop Lab is an independent research studio studying how people think, act, evaluate evidence, and build knowledge under conditions of uncertainty.
The work combines cognitive psychology, social science, data analysis, and human-AI collaboration. AI participates across problem framing, evidence synthesis, methodological design, coding, analysis, and drafting. Human judgment remains responsible for the questions, the assumptions, the interpretation, and the final claims.
Current projects examine interpersonal violence and conflict dynamics, AI-generated citation error in scholarly writing, and evaluation frameworks across education and AI. The common thread is evidence: how it is produced, organized, distorted, validated, and argued from.
The loop stays open. Questions are revised as evidence accumulates. Assumptions are made visible. Findings are treated as arguments to inspect, not outputs to accept.
Projects in violence and conflict modeling, hallucinated citation detection, and evaluation methodology.
Reproducible workflows for evidence synthesis, modeling, human-AI collaboration, and transparent research communication.
Open Loop Lab is led by Joe Stafura, a cognitive psychologist working across research, data, and applied knowledge systems.