Brief 2: The Withdrawal Study

THE WITHDRAWAL STUDY PROTOCOL

A Research Design for Measuring AI Companion Dependency

Trinket Soul Framework — Research Brief No. 2

Michael S. Moniz

February 2026

A companion brief to The True Economy Audit (Volume III)

Creative Commons Attribution-NonCommercial-ShareAlike 4.0

ABSTRACT

This brief proposes a pre-registered longitudinal study design to test the Trinket Soul Framework’s core applied prediction: that AI companion applications failing more of the six structural tests (as defined in The True Economy Audit, Volume III) will produce greater user dependency, higher withdrawal distress, and more measurable erosion of human relational capacity than applications passing more tests.

1. BACKGROUND AND RATIONALE

1.1 The Knowledge Gap

AI companion applications have reached millions of users globally. Despite this scale, almost no peer-reviewed longitudinal research examines what happens when regular users stop using these applications.

1.2 The Theoretical Prediction

The Trinket Soul Framework generates a specific, testable prediction: AI companion applications that fail more of the six structural tests for genuine relational capacity will produce higher withdrawal distress in users, not lower.

1.3 Hypotheses

H1 (Primary): Users of AI companion applications that fail more of the six structural tests will report significantly higher withdrawal distress during a 30-day abstinence period.

H2: Withdrawal distress will be positively correlated with pre-abstinence usage intensity and negatively correlated with the number of structural tests passed.

H3: Users of applications failing Test 6 (attachment-sensitive calibration) specifically will show higher withdrawal distress.

H4 (Exploratory): Users with anxious attachment styles will show higher withdrawal distress, moderated by the application’s structural test profile.

H5 (Exploratory): Pre-to-post changes in human relational investment will differ between users of structurally sound versus structurally deficient applications.

2. STUDY DESIGN

The study uses a naturalistic quasi-experimental design with a 30-day abstinence intervention. The independent variable is the structural test profile of the participant’s primary AI companion application. The dependent variables are withdrawal distress, human relational investment, and psychological wellbeing.

3. ANALYSIS PLAN

Between-groups comparison of withdrawal scores across application structural tiers using mixed-effects models with time as a within-subjects factor and structural tier as a between-subjects factor.

4. LIMITATIONS AND ETHICAL CONSIDERATIONS

The study asks people to stop using a tool they may rely on for emotional support. The safety protocol is essential, not optional.

5. CALL FOR COLLABORATION

This study protocol is published as an open design. We invite academic researchers, institutional review boards, and funding bodies to adopt, adapt, and execute it.

Addendum: Withdrawal Taxonomy

THE WITHDRAWAL TAXONOMY

Trinket Soul Framework — Addendum to Brief No. 2

Michael S. Moniz · February 2026

Pathway-Specific Hypotheses for AI Companion Withdrawal Studies

CONTEXT

Brief No. 2 proposes an empirical investigation into the effects of AI companion removal. Volume IV establishes that individuals carry fundamentally different relational architectures, organized around four degradation pathways: extraction, template distortion, atrophy, and catastrophic loss. This addendum proposes that the study design must account for pathway-dependent variation in withdrawal response.

Pathway 1: Extraction-Degraded Users — Removal prediction: neutral to positive. Withdrawal distress: low.

Pathway 2: Template-Degraded Users — Removal prediction: acutely negative. The user loses their only experience of relational safety. Withdrawal distress: high and specific.

Pathway 3: Atrophy-Degraded Users — Removal prediction: initially negative, potentially positive with re-engagement support. Withdrawal distress: moderate and generalized.

Pathway 4: Catastrophically-Degraded Users — Removal prediction: potentially dangerous. The user loses their maintenance tool during acute vulnerability. Withdrawal distress: high and qualitatively different.

STUDY DESIGN RECOMMENDATIONS

Pre-withdrawal architectural assessment. Pathway-specific hypotheses. Ethical constraints for Pathway 4. Graduated withdrawal option. Longitudinal follow-up.

This addendum does not change Brief No. 2’s core proposal. It sharpens it.