The Psychology of Online Dating — What It Changes and Its Limits
Last reviewed by the Men Women Psychology editorial team.
The evidence
What the research actually shows
The most comprehensive review of this field, by Finkel, Eastwick, Karney, Reis and Sprecher (2012), concluded that online dating differs from conventional dating in three main ways: it offers access to far more potential partners, it provides communication before meeting, and it uses matching algorithms. The authors found strong evidence for the benefit of access, but they were skeptical that algorithms can predict long-term compatibility from the kind of data profiles typically capture.
Their analysis suggests that the very features designed to help can sometimes hinder. Browsing many profiles can encourage an evaluative, almost shopping-like mindset, and an abundance of options may make people more likely to keep searching than to invest. The research indicates these effects vary between individuals, and that more choice is not always experienced as better.
Work on attraction more broadly reinforces the limits of prediction. Eastwick and Finkel (2008) found that people's stated partner preferences often do not match who they are actually drawn to in person, and Montoya and Horton's (2013) meta-analysis on similarity suggests attraction depends heavily on perceived responsiveness and interaction, which are hard to capture from a static profile.
The mechanism
Why this happens
Profiles compress a person into a few photos and lines of text, which favors easily described traits over the qualities that actually drive connection. Finkel and colleagues argue that compatibility tends to emerge from how two people interact, something a profile cannot show. This helps explain why a promising match on paper can fall flat in person, and why an unremarkable profile can lead to real chemistry.
The psychology of choice also shapes the experience. When options feel nearly unlimited, people can become more comparative and less committed to any one prospect. The research suggests this 'choice overload' can shift attention toward what a different match might offer rather than toward investing in the conversation at hand, though individuals differ a lot in how strongly they feel this.
Attraction itself relies on cues that are difficult to convey digitally. Eastwick and Finkel's findings suggest people are often poor at predicting whom they will be drawn to, and Montoya and Horton's work highlights that feeling understood in real interaction matters more than matching on listed attributes, both of which are easier to experience face to face than to forecast online.
In practice
What this looks like in real life
Someone might match with a person who seems ideal on paper, then feel little spark when they meet, while feeling unexpected chemistry with someone whose profile they almost skipped. Research on the gap between stated preferences and actual attraction helps explain why this is common rather than a personal failing.
A person scrolling through dozens of profiles in one sitting may notice they start evaluating people like options to compare, finding it harder to focus on any single conversation. This evaluative mindset is one of the effects the literature flags as a downside of abundant choice.
On the positive side, online platforms regularly connect people who would never have crossed paths otherwise — different social circles, neighborhoods, or schedules. The research is clear that this expanded access is the technology's most reliable and genuine advantage.
Myth vs. evidence
What most people get wrong about this
A common misconception is that a sophisticated algorithm can identify a soulmate from profile data. The major review by Finkel and colleagues found little evidence that current matching algorithms predict long-term success better than chance would, largely because compatibility depends on interaction that profiles do not capture.
Another error is treating early text exchanges as a reliable read on chemistry. Research on attraction suggests that responsiveness and connection are best assessed in person, so extended messaging can build expectations that an in-person meeting may or may not confirm. The platform is generally better at introductions than at prediction.
Why it matters
What this means for relationships
Because profiles and algorithms predict chemistry poorly, the research broadly supports treating online dating as a tool for introductions rather than a verdict on compatibility. Meeting in person relatively early tends to give a more accurate read than prolonged messaging, since the qualities that drive attraction emerge through interaction.
Being aware of choice overload can also help. Approaching matches with curiosity and a willingness to invest, rather than endless comparison, aligns with what the research suggests actually builds connection. This applies across genders, even though the experience of online dating can differ in some respects between groups.
Where it varies
The nuance
Online dating experiences vary widely, and average patterns overlap heavily across people. Janet Hyde's gender similarities hypothesis (2005) is a useful reminder that men and women are far more alike than different on most psychological measures, so broad generalizations about how each 'uses' dating apps tend to overstate real differences.
Platforms and norms also keep changing, and the research describes tendencies rather than universal outcomes. Many people form lasting relationships online, and many find it frustrating; individual circumstances, expectations, and how someone uses the tools shape the experience as much as the technology itself.
Questions people ask about this
Can dating algorithms really predict compatibility?
The major review by Finkel and colleagues found little evidence that current matching algorithms predict long-term success better than chance. Profiles tend to miss the interactional qualities that drive real compatibility. Algorithms are generally better at expanding access to potential partners than at forecasting who will click.
Why do good matches on paper sometimes fall flat in person?
Research by Eastwick and Finkel suggests people are often poor at predicting whom they will actually be drawn to, and attraction depends heavily on in-person interaction. A profile captures listed traits but not chemistry, so a strong-looking match can feel different once you meet.
Does having more options make dating better?
Not always. Studies suggest an abundance of choices can encourage a comparative, evaluative mindset and make it harder to invest in any one person. This effect varies between individuals, but more options are not reliably experienced as better, according to the research.
Is it better to text a lot first or meet sooner?
Research on attraction suggests chemistry is best assessed in person, since responsiveness and connection are hard to gauge through text. Extended messaging can build expectations that meeting may not confirm. Many findings point toward meeting relatively early as a more accurate read than prolonged texting.
Is online dating worse than meeting in real life?
The research suggests it is mainly different, not simply better or worse. Its clearest benefit is expanded access to people you would not otherwise meet. Its limits involve prediction and choice overload. Outcomes vary widely by individual, and many people form lasting relationships this way.
Do men and women use dating apps very differently?
There are some reported differences, but individual variation tends to outweigh average gender differences, and Hyde's gender similarities hypothesis reminds us the sexes are largely alike on most measures. Broad generalizations about how each gender 'uses' apps often overstate the actual differences the evidence supports.
Research sources
These references point to the published research and established frameworks behind this page. They are provided for further reading; see our research methodology for how sources are selected.
- Finkel, E. J., Eastwick, P. W., Karney, B. R., Reis, H. T., & Sprecher, S. (2012). Online dating: A critical analysis. Psychological Science in the Public Interest, 13(1), 3–66.
- Eastwick, P. W., & Finkel, E. J. (2008). Sex differences in mate preferences revisited. Journal of Personality and Social Psychology, 94(2), 245–264.
- Montoya, R. M., & Horton, R. S. (2013). A meta-analytic investigation of the processes underlying the similarity-attraction effect. Journal of Social and Personal Relationships, 30(1), 64–94.
- Hyde, J. S. (2005). The gender similarities hypothesis. American Psychologist, 60(6), 581–592.