Similarity Isn’t Destiny: Why “Birds of a Feather” Might Be a Red Herring in Long-Term Love

Tuesday, July 1, 2025.

You know the cliché: happy couples finish each other’s sentences, order the same sushi, and secretly share a Spotify playlist full of Fleetwood Mac.

Compatibility, we’re told, is about similarity—same interests, same values, same neurotic love for seasonal throw pillows.

But a massive new review just dropped a wet towel on that fantasy.

According to a scoping review of 339 studies published between 1937 and 2024, actual similarity between long-term romantic partners has only a modest and inconsistent connection with relationship satisfaction or longevity (From et al., 2024).

Let that sink in. Hundreds of studies. Eighty-seven years of data. And the results? Meh.

Perceived vs. Actual Similarity

The researchers, led by University of Michigan doctoral candidate Annika From, weren’t asking a small question.

They wanted to know: Do couples who actually share the same personality traits, values, hobbies, or demographics do better over time? Or is the feeling of similarity—however illusory—doing all the heavy lifting?

Turns out, it’s the latter.

In study after study, perceived similarity (“I think we’re alike”) predicted greater satisfaction and stability far more consistently than actual similarity (as measured by matched questionnaires). It’s not who your partner really is—it’s who you believe they are that may matter most.

The Evidence Was… Mixed at Best

This wasn’t a cherry-picked sample.

The team combed four major academic databases and winnowed over 7,600 results down to 339 relevant studies on long-term couples.

Most participants were heterosexual and married, and while the studies covered six domains of similarity—values, lifestyle, personality, appearance, romance habits, and demographics—the trend was surprisingly consistent: actual similarity didn’t correlate strongly with better outcomes (From et al., 2024).

The one modest exception? Demographics like race and education.

These showed more predictive value than other domains—but even those effects were inconsistent across studies.

How Methodology Messes With Meaning

Here’s where it gets spicier: the type of statistical method used mattered a lot.

Simpler studies using basic score-matching tended to find stronger links between similarity and happiness.

But when researchers used more robust statistical controls—the kind that adjust for personality bias and base-rate matching—the effects all but disappeared.

In other words, much of the old evidence for compatibility may be an artifact of lazy math.

The authors point out that studies rarely adjust for something called “positivity bias”—the tendency to rate both yourself and your partner highly (especially early on), which makes you feel similar even if you aren’t. Once that’s corrected for, similarity’s romantic sheen fades.

What Does This Mean for Couples Therapy?

As a couples therapist, I’ve seen this play out in session after session. One partner says, “We’re just too different.” But when we dig deeper, the pain isn’t about values or tastes—it’s about feeling unseen. Misattuned. Misunderstood. The mismatch isn’t about brunch preferences or Myers-Briggs types. It’s about the narrative: “You’re not like me, and that makes me alone.”

This study adds fuel to an important insight: feeling emotionally aligned matters more than factual similarity.

If your partner gets you, they don’t have to be you.

The Power—and Fragility—of Perception

So what does shape our perception of similarity?

Memory. Mood. Confirmation bias. Relationship stage.

We see more likeness in the honeymoon phase than we do after someone leaves wet towels on the bed for the 112th time. But perception, like attraction, is malleable.

And if perception drives satisfaction, then the therapeutic focus should be less on “fixing differences” and more on building empathy, shared meaning, and secure connection.

Because we fall in love with who we think someone is. The trick is learning to stay in love as they change—and as we do, too.

Compatibility in the Age of Algorithms

The findings also cast a long shadow on algorithm-based dating platforms, many of which still push the compatibility myth. These platforms match users on preferences, traits, and demographics as if shared love for Thai food and indie music guarantees relational bliss.

It really doesn’t.

As the authors note, even the most well-designed studies were underpowered to detect small but real effects. Which means there might be slight benefits to actual similarity—but they’re far too subtle to bet a marriage on.

So What Can Couples Do? Build the Feeling.

If actual similarity doesn’t do the trick, perceived similarity is still gold. And unlike personality traits or birth charts, that’s something you can build. Here’s how:

Practice Narrative Alignment

Tell your story together. Couples who co-construct a “we” story—how they met, what they’ve survived, what they believe—build a deeper sense of perceived commonality, even across different life experiences.

Mirror Back Values

Noticing and reflecting your partner’s values (“You really care about fairness, don’t you?”) helps them feel known, which increases their sense that you’re similar—even if you vote differently.

Create Shared Rituals

Repeated experiences—Sunday walks, bedtime questions, weird inside jokes—forge a sense of cohesion. Rituals don’t just bond us; they make us believe we’re on the same page.

Be Curious, Not Corrective

Similarity perception rises when partners feel safe being themselves. Curiosity—not correction—creates emotional mirroring that builds a stronger mental model of “we.”

Use Perspective-Taking Language

Studies show that partners who say “we,” “us,” and “our” rather than “you” and “I” report stronger perceived alignment. It’s not magic, just neuroscience dressed up as grammar.

Conclusion: Same-Same, But Who Cares?

We think we want a partner just like us. But the research suggests we don’t need one.

What we seem to crave more than similarity is resonance—a sense that who we are is understood and respected, even if it’s not mirrored. That’s not sameness. That’s attunement.

And maybe that’s the real story behind “birds of a feather.” It’s not that they’re identical. It’s that they know how to fly in formation.

Be Well, Stay Kind, and Godspeed.

REFERENCES:

From, A., Diamond, E., Kafaee, N., Reynaga, M., Edelstein, R. S., & Gordon, A. M. (2024). Does similarity matter? A scoping review of perceived and actual similarity in romantic couples. Journal of Social and Personal Relationships, 41(6), 1152–1175. https://doi.org/10.1177/02654075241250476