Compatibility tests love to measure the wrong things. Personality quizzes, values checklists, love-language assessments — an entire industry built on the idea that finding someone who scores like you on the right instrument is the thing that makes a relationship work. A new study called “Harmonies of Taste: Shared Music Preferences and Compatibility in Romantic Relationships”, published today in Social Psychological and Personality Science, tested that assumption on an axis nobody usually checks: what’s actually in your playlist. And the axis that turned out to matter wasn’t the one self-help keeps measuring.

Julia Vigl, Felix Lene Ihrig, Miriam Greif, and Marcel Zentner — researchers with the Music, Emotion, Assessment, and Personality Lab at the University of Innsbruck — ran three studies covering 309 couples, mostly young adults in Germany and Austria in relationships averaging somewhere between 3.5 and 6.5 years. They measured music preference similarity two ways: by genre (rock, hip-hop, classical, the usual buckets) and by actual favorite and least-favorite artists, cross-referenced against online tags to catch stylistic overlap a genre label would miss. Then they measured the same couples on Big Five personality traits and personal values, and compared all three against real couples’ pseudo-couple counterparts — randomly paired participants who never dated, included specifically to rule out “everyone in this sample just likes the same stuff culturally.”

Quick Verdict

Question	Answer
Who ran the study?	Julia Vigl, Felix Lene Ihrig, Miriam Greif, Marcel Zentner (University of Innsbruck)
Sample	309 couples across three studies, mostly young adults, relationships averaging 3.5–6.5 years
What was compared?	Music preference similarity (genre and artist level) vs. Big Five personality similarity vs. personal values similarity
Which one was bigger?	Music, by a wide margin — personality similarity was close to negligible
Which one predicted relationship quality?	Music similarity, especially artist-level overlap. Personality and values similarity didn’t.
Effect size on relationship quality	Modest but significant — small, real, not decisive

Bottom line: Couples look alike on music in a way they don’t look alike on personality. And unlike personality matching, the music match actually tracks with how good the relationship is.

What the Study Actually Tested

This wasn’t one survey. It was three, plus a mini meta-analysis pooling all of them, which is the part that makes the finding hold up instead of reading like a fluke from one questionnaire. Each study asked couples to report their music preferences at both the genre and artist level, then ran the same couples through standard personality and values inventories. The pseudo-couple comparison is the detail that separates this from a lot of pop-science compatibility content: instead of just asking “are couples similar on music,” the researchers asked “are couples more similar on music than two random strangers from the same cultural pool would be by chance.” That’s a harder bar to clear, and music cleared it. Personality barely did.

Do Couples Actually Match on Music More Than Personality?

Yes — and the gap isn’t close. Three findings carry the result:

1. Music preference similarity between real couples was significantly greater than chance, at both the genre and artist level. Partners weren’t just both “into music broadly.” They converged on specific genres and, more strikingly, specific artists.
2. Personality similarity was negligible by comparison. Big Five traits — the framework most compatibility content leans on — showed couples pairing up close to randomly. Two people who love each other do not reliably share a personality profile.
3. Values similarity fell somewhere in between, but nowhere near the music effect. Shared beliefs about what matters in life didn’t distinguish real couples from randomly paired strangers nearly as well as shared taste in artists did.

That ordering — music way out front, personality trailing, values in the middle but still behind — is the opposite of what most relationship advice assumes. Personality and values are the stuff dating apps ask you to fill out. Music is the thing you put on in the car and never think of as a compatibility metric.

The Part That Should Bother Compatibility Culture: Music Also Predicted Quality

Being similar on something and that similarity actually mattering for the relationship are two different claims, and a lot of compatibility research collapses them into one. This study didn’t. Shared musical taste — again, especially at the artist level, not just “we both like pop” — showed a modest but statistically significant association with higher relationship quality. Personality similarity and values similarity did not show that same link. Couples can match closely on values and it tells you nothing reliable about how the relationship is doing. Couples who match on artists show a small real bump.

Small is the honest word here. This isn’t “swap your dating profile for a Spotify Wrapped comparison and solve your relationship.” The effect is modest, and the researchers frame it that way. But it’s a real, statistically significant effect sitting on top of a null result for personality — which is a strange thing for an industry built on personality-matching quizzes to sit with.

Why Music Might Be Doing Work Personality Doesn’t

The study’s authors point toward a plausible mechanism without overclaiming it: shared musical taste creates shared activity, not just shared preference. Concerts. Festivals. Making a playlist together, or fighting over the aux cord in a way that’s really just two people negotiating shared time. How to Feel Loved makes a related point about the love-languages framework — that compatibility is often less about matching internal traits and more about what two people actually do together, repeatedly, over time. Music preference converts into behavior in a way a Big Five score doesn’t. You don’t go do “conscientiousness” together on a Saturday night. You do go see a band.

That’s a mechanism worth taking seriously, but it’s still a hypothesis, not something this study directly tested. What the data shows is the correlation. The “why” is the researchers’ best explanation, not a proven pathway.

Where This Fits With What This Site Keeps Finding

This is the third time recently that a study has taken a piece of compatibility folklore and found the “match” variable doing less work than expected — or, in this case, doing more than expected on an axis nobody was checking. The touch-comfort research found overall comfort with physical affection beat matching love languages as a predictor of relationship well-being — matching barely helped. The mate-value meta-analysis found self-esteem tracks with felt romantic desirability more than any trait-matching exercise. Both of those studies found “similarity” underperforming as a compatibility mechanism. This one flips that pattern: on this specific axis, similarity is unusually strong and it’s one of the only similarity measures tested here that actually predicts relationship quality.

Put those three together and a shape emerges. Compatibility isn’t one variable working the same way everywhere. Personality similarity: doesn’t distinguish couples from strangers, doesn’t predict quality. Touch-comfort similarity: some effect, smaller than raw comfort. Music similarity: strong on distinguishing couples from strangers, and it’s the one that actually correlates with how the relationship is going. The research on why happy marriages still show up in bodies that age faster is a reminder from the same beat that “good relationship” and “good outcome on every metric” aren’t the same claim either — good marriages track with worse health habits, not just good feelings. Relationship science keeps finding that the intuitive, tidy story about what makes couples work isn’t the one the data supports.

What This Doesn’t Mean

• It doesn’t mean personality doesn’t matter at all. It means personality similarity specifically didn’t predict much here. A partner’s personality can matter enormously to a relationship without two partners needing to share the same one.
• It doesn’t prove music similarity causes better relationships. This is correlational, observational data. Couples with more compatible taste might also just be couples who spend more shared leisure time together for other reasons, and the music overlap is a symptom of that, not the cause.
• It doesn’t mean shared taste developed before the relationship. The study can’t say whether couples found each other because their taste already matched, or whether taste converged over years of shared listening. Both are plausible; this data doesn’t settle it.
• It doesn’t generalize past the sample with full confidence. Mostly young adults in Germany and Austria. Whether the same pattern holds for older couples, different countries, or very different music cultures is an open question.
• The effect size is modest. A significant correlation on relationship quality isn’t a strong one. Don’t restructure a relationship decision around a playlist comparison.

What Actually Holds Up, Given This

• Drop personality-matching quizzes as a compatibility signal. This data adds to a growing pile suggesting shared traits aren’t doing the predictive work relationship content assumes they are.
• Shared activity might matter more than shared internal profile. Music converts into things couples actually do — concerts, playlists, road trips. That’s worth more attention than another values questionnaire.
• Take “we have nothing in common” seriously if it’s about how you spend time, less seriously if it’s about a personality test result. This study gives some backing to the first concern and very little to the second.
• Don’t overweight a small effect. Modest and significant is still modest. This is one input among many, not a verdict on a relationship.

The Bottom Line

Three studies, 309 couples, and a finding that inverts the usual compatibility script: personality similarity was close to negligible, values similarity was middling, and music similarity — genre and especially artist level — was both the strongest form of matching in the data and the only one of the three that tracked with relationship quality. It’s a small effect, and it doesn’t prove causation. But it’s one more data point saying the traits self-help treats as the core of compatibility (personality, values) aren’t the ones the evidence keeps pointing to. What you both put on in the car might matter more than what you both scored on a personality test.

---

Vigl, J., Ihrig, F. L., Greif, M., & Zentner, M., “Harmonies of Taste: Shared Music Preferences and Compatibility in Romantic Relationships,” Social Psychological and Personality Science (2026). Additional coverage from PsyPost, September 19, 2026. Related reading: Comfort With Touch Predicts Relationship Happiness, Feeling Like a Catch Is Tied to Self-Esteem, Happy Marriage Speeds Aging With Bad Habits, and How to Feel Loved by Lyubomirsky & Reis Review.