Attachment styles, honestly
Attachment theory has a real research foundation — but the version circulating online (four tidy types, from a ten-question quiz, that explain your whole love life) is far stronger than the evidence supports. Attachment is better described as two continuous dimensions — anxiety and avoidance — than as four boxes, it is not fixed, and it varies across relationships.
The Attachment Space
Place yourself on the two dimensions — this tool will refuse to give you a type
What the theory actually claims
The original work described how infants use a caregiver as a secure base, and later researchers extended it to adult romantic bonds. That extension found something real: people do differ reliably in how much they worry about abandonment (attachment anxiety) and how much they suppress closeness and dependence (attachment avoidance). Those two dimensions predict relationship behaviour meaningfully.
Where the internet version goes wrong
- They are not types. The evidence favours continuous dimensions; most people sit in the middle, not in a corner. "I'm an anxious-avoidant" is a horoscope-shaped answer to a bell-curve-shaped question.
- They are not fixed. Attachment can and does change — with time, with therapy, and with a different partner. Anyone telling you this is your permanent romantic operating system is overselling.
- They are not one-per-person. You can be secure with one person and anxious with another. It is a property of a relationship at least as much as of a person.
- The childhood link is weaker than claimed. Infant attachment predicts adult romantic attachment only modestly. Your mother is not the reason your last relationship ended.
Because it does something powerful: it turns "I am broken" into "I have a pattern, and patterns can change." That reframe is genuinely valuable — and you can keep it without pretending the science is more settled than it is.
The useful residue
Strip the mysticism and something true remains: people differ in how they handle closeness and the threat of losing it — and naming your pattern out loud, to someone who cares about you, is often the beginning of it shifting. That is the part worth keeping.
The five love languages are a useful conversation-starter with weak empirical support — studies have generally failed to show that matching partners' "languages" improves relationships. Talking to your partner about what makes them feel loved is great. The framework is not why it works.
Keep going
More from this wing: all 8 pages · related: jealousy, grief, manipulation tactics.
Frequently asked
Are attachment styles scientifically valid?
Can your attachment style change?
Do love languages work?
Sources & further reading
- Attachment theory originates with Bowlby & Ainsworth; modern work treats adult attachment dimensionally (anxiety × avoidance) rather than as fixed “types.” Attachment, dimensional view ↗
- Fraley & Waller: adult attachment is better modelled dimensionally; categorical models manufacture spurious “change” near the thresholds. Fraley & Waller, author copy (ResearchGate) ↗
- Chopik and colleagues: attachment shows moderate stability but shifts across the lifespan and through relationships (“earned security”). Longitudinal cohort study ↗
- Impett, Park & Muise (2024): the “five love languages” framework lacks strong empirical support — people value all five, and “matching” doesn’t reliably predict satisfaction. Impett et al., 2024 ↗
Educational content, not therapy or professional advice. Last reviewed July 2026.