About
About the Research
The Judgmentalism Project is an independent research initiative and the permanent home for the theoretical work developed under its name. It addresses a core underlying problem across three interconnected lines of enquiry: how individuals and artificial intelligence systems process, misattribute, and ultimately judge ambiguous social and emotional data.
The first line, Trait Judgmentalism & Evaluative Rigidity, explores how individuals form and maintain rigidly negative social evaluations. This rigidity is driven by Contextual Moderation Failure (CMF) and the Evaluative Gating Model, a cognitive disruption where mitigating context is fully comprehended but assigned negligible weight in the final judgment. To capture this empirically, this branch includes the development of the Judgmentalism Assessment Scale (JAS), an instrument designed to bypass the traditional “Expression Assumption” to measure latent, internalised punitive attitudes rather than just overt behaviour.
The second line focuses on Psychometrics and Affective Perception, specifically within ability emotional intelligence. It investigates the “Perception-Comprehension Gap,” which is the cognitive disconnect between accurately seeing an emotional signal and accurately understanding its meaning. To solve the measurement confounding inherent in these layered processes, the project developed Branch-Residualised Interpretation (BRI). This regression-based statistical framework mathematically disentangles upstream perception variance from downstream reasoning variance, allowing researchers to isolate specific cognitive misfires, such as those modeled in the Branch Dissociation Typology.
The third line extends these theoretical and measurement frameworks into computational territory. Utilizing Bayesian modeling and the newly introduced Violation Construction Index (VCI), this branch audits classification-stage and persona-driven bias in Large Language Models. It tests whether AI systems objectively detect moral violations or if they systematically construct them based on prompted social identities, mimicking human evaluative rigidity.
Full open-access preprints and R pipelines are available in Research. The project is entirely self-funded and operates as a non-commercial initiative, with no external investors, sponsorships, or financial interests.
About the Researcher
Emile Boullineau is a British–South African professional in global risk management with over 25 years of international experience in institutional governance, holding Chartered Financial Analyst (CFA) and Financial Risk Manager (FRM) designations. He holds a Bachelor of Arts in Applied Psychology from the University of South Africa and a Postgraduate Diploma in Accounting from the University of Natal. He is currently completing an MSc in Psychology (Distance Learning) at Northumbria University.
His independent research investigates the cognitive architecture of human judgment, spanning psychometric methodology, cross-cultural test validation, and algorithmic fairness in AI models. Although his extensive professional background in structural risk and governance has informed his approach to auditing cognitive and systemic bias, The Judgmentalism Project is an entirely separate, independent academic initiative.
Professional Disclaimer
This work is a private academic project, developed entirely outside the scope of my professional corporate role and institutional duties. The research, preprints, and theories hosted here are independent scholarly pursuits and do not represent the views, positions, or intellectual property of any employer or corporate affiliation.