We write about attachment psychology for a general audience, which means being careful in two directions at once: staying grounded in real research, and never overstating what a quiz can tell you about a person. This page explains how we try to do both.
Scoring is rules-based, not AI-guessed
Your attachment anxiety and avoidance scores come from a deterministic scoring model applied to your answers — the same inputs always produce the same result. AI is only used afterward, to explain a result that's already been calculated, and it is never allowed to change the score or invent a classification of its own.
Claims are pre-approved, not generated on the fly
When AI writes a paid report, it can only draw on a fixed database of research-backed statements that we've reviewed and approved in advance, each with its own source and an honest note on how strong the underlying evidence actually is. The AI can't cite a study or make a claim that isn't already in that approved set.
Every report is checked before it's sent
Paid reports go through a second, independent AI review after they're written, specifically looking for diagnostic language, unsupported claims, or advice that isn't appropriate for a non-clinical tool. Nothing is delivered until it passes that check.
Language we commit to
We write “your responses are most consistent with…”, never “you are…”. We assume nothing about your family structure, gender, sexuality, or relationship history, and every question that doesn't apply to everyone includes a way to say so.
Sourcing
Our starting point is Bowlby and Ainsworth's foundational attachment research, cross-referenced against the current literature on PubMed and Google Scholar. We'd rather understate a finding than overstate it.
Corrections
If something on this site is inaccurate or out of date, we want to know. Email contact@myattachmentquiz.com and we'll review it.