The short answer: yes, a phone that keeps pulling attention away from your partner is linked to lower relationship satisfaction, but the link is modest, it comes mostly from studies that cannot prove cause and effect, and it seems to depend less on how much you use your phone than on when you use it. A 2025 meta-analysis pooling 52 datasets puts the average correlation between being “phubbed” and relationship satisfaction at about −0.22. That is real, and it is a long way from the verdict that smartphones are wrecking couples.
Most arguments about this subject are conducted with anecdotes: the dinner where both people scroll, the partner who answers one more email in bed. The research is more interesting than the anecdotes because it keeps contradicting the easy story. Our own work on ignored bids for connection already suggested that small, repeated moments of turning toward or away from a partner matter more than grand gestures. The phone is simply the most efficient device ever invented for turning away.
Two words for one habit
Two terms dominate the literature, and they are not interchangeable. Technoference refers to everyday interruptions of couple time by technology: a notification during a conversation, a laptop open during a shared evening, a television that stays on at dinner. It is about the interruption, whoever causes it. Partner phubbing (a blend of “phone” and “snubbing”) is narrower and more personal: ignoring the person you are with in favor of your phone, as the partner experiences it.
The distinction matters for how studies are built. Technoference is usually measured from the point of view of one person in the couple, asking how often devices intrude on shared activities such as leisure time, conversations and meals. Phubbing scales ask about specific behaviors, like a partner glancing at the phone during a conversation or keeping it in hand while eating together. Both rely heavily on what people report, and what people report is filtered by how they already feel about the relationship. Keep that in mind for everything that follows.
What the founding studies found
The study that popularized the word technoference was published in 2016 in Psychology of Popular Media Culture by Brandon McDaniel and Sarah Coyne. It surveyed 143 married or cohabiting women. Women who reported more technoference in their relationships also reported more conflict over technology use, lower relationship satisfaction, more depressive symptoms and lower life satisfaction. The authors tested a chain: technology interruptions lead to conflict about technology, conflict lowers relationship satisfaction, and lower satisfaction feeds depression and dissatisfaction with life.
The same year, James Roberts and Meredith David reported in Computers in Human Behavior that being phubbed by a partner was associated with more conflict, lower relationship satisfaction and more depressive symptoms. Their paper introduced a scale for partner phubbing that many later studies reused, which is partly why the field piled up so quickly.
A third piece of the foundation concerns conversation quality rather than marriages. In an experiment published in Environment and Behavior, Shalini Misra and colleagues assigned 100 pairs of people to discuss either a casual or a meaningful topic for ten minutes. A trained observer, watching from a distance, noted whether either person put a mobile device on the table or held one. Conversations held without a device were rated significantly better than those held in the presence of one, above and beyond age, gender, ethnicity and mood, and people without a device nearby reported more empathic concern. Among pairs who were close to each other, the presence of a device went with lower empathy than among pairs who were less close. The paired people were not necessarily couples, so the study speaks to conversation in general, but its lesson transfers neatly: a phone does not need to ring to change a conversation.
The numbers across dozens of studies
Ten years of follow-up papers made it possible to ask what the whole pile says. A 2025 meta-analysis in Frontiers in Psychology by Nie Ni, Seyedali Ahrari and colleagues gathered 52 datasets (58 samples, 19,698 participants). For relationship satisfaction, based on 30 studies and 9,040 people, the pooled correlation with partner phubbing was −0.219 (95 percent confidence interval from −0.26 to −0.175). For marital satisfaction specifically, five studies covering 1,878 people gave −0.264, with a much wider interval. The link to overall life satisfaction was not statistically significant.
Two details deserve more attention than the headline. First, culture appeared to matter: the negative association was stronger in Western samples (−0.250) than in Eastern samples (−0.176). Second, relationship length moderated the effect on satisfaction, though the authors report a small coefficient. A correlation of −0.22 means phubbing accounts for a small slice of the differences in satisfaction between couples. It is a risk factor to take seriously, not a diagnosis. The authors also stress the limit that frames everything here: most studies are cross-sectional, so they cannot tell whether phubbing damages relationships or whether a relationship in trouble produces more phubbing.
| Study | Design | What it found | The caveat |
|---|---|---|---|
| McDaniel & Coyne (2016) | Survey, 143 women in couples | More technoference went with more conflict, lower satisfaction, more depressive symptoms | One snapshot, women only |
| Roberts & David (2016) | Survey on partner phubbing | Being phubbed went with more conflict and lower satisfaction | Self-report, no timeline |
| Misra et al. (2016) | Field experiment, 100 pairs | Conversations without a device rated better, with more empathic concern | Pairs, not only couples; short conversations |
| McDaniel & Drouin (2019) | 14-day diary, 173 couples | Days with more interruptions than usual felt worse | Parents of young children, single-item measure |
| Ni, Ahrari et al. (2025) | Meta-analysis, 52 datasets | Average correlation of about −0.22 with satisfaction | Mostly cross-sectional |
| McDaniel et al. (2025) | 8 days of phone tracking, 247 parents | Phone use around the partner, not total use, tracked relationship quality | Parents of infants, proximity self-reported |
A cleaner test: the same couple, day after day
The way out of the cause-and-effect problem is to compare people with themselves. In a daily diary study in Computers in Human Behavior, Brandon McDaniel and Michelle Drouin followed 173 couples for 14 consecutive days (173 women and 171 men completed the daily surveys). On days when a person perceived more technoference than their own usual level, they felt worse about the relationship, perceived more conflict over technology use, rated face-to-face interactions as less positive and reported a more negative mood. Over the two weeks, 56.1 percent of participants experienced technoference at least a few days.
This design removes one explanation: the person is not a “chronically unhappy partner”, since they are being compared with their own ordinary days. It leaves others standing. The couples were U.S. residents with children under five, all heterosexual, and technoference was measured with a single item. A bad day at work could plausibly raise both phone use and irritability. The finding is a stronger hint than a survey, not a proof. It also has to compete with everything else that moves satisfaction up and down, which the research on relationship satisfaction describes at length: a phone effect is one voice in a crowded room.
- A cross-sectional survey compares different couples at one moment and cannot separate cause from consequence.
- A diary study compares a person with themselves and is much better at timing, but it still cannot rule out a third factor such as stress.
- An experiment can isolate the phone, which is why the Misra design is so useful, but it covers a ten-minute conversation, not a marriage.
- A meta-analysis reveals what is consistent across studies and says nothing new about causation if the underlying studies cannot.
It may be about when, not how much
The most useful recent evidence replaces self-report with tracking. In a study in Human Behavior and Emerging Technologies, McDaniel, Uva, Cornet and Drouin followed 247 U.S. adults, 75 percent of them women, all parents of infants. Participants installed tracking apps (RescueTime on iPhone, Chronicle on Android) for eight days and filled in nightly diaries about when their partner was around, with a baseline and a follow-up survey.
On average, people used their phone during 27.75 percent of their time around their partner, and 86 percent used it at least a little around their partner every single day. So the behavior is nearly universal, which already suggests that “uses a phone near a partner” cannot be the dividing line between happy and unhappy couples. What separated outcomes was the share of partner time spent on the phone. Phone use around the partner predicted lower relationship satisfaction and lower coparenting quality, a pattern that reached significance for women only. Total daily phone use, on its own, did not predict relationship quality. Both measures predicted more depression and lower life satisfaction, with stronger effects for women.
Read carefully, that is a surprisingly practical result. The problem is not owning a phone, or spending hours on it at work, or even a heavy evening of scrolling alone. It is the moments when someone is physically with their partner and mentally elsewhere. The authors flag the usual limits: the sample is parents of infants, proximity to the partner was self-reported in diaries, and the design is correlational. A couple that has drifted apart may reach for the phone more, not the other way around.
Why a glance can feel like rejection
None of the numbers explains why the effect would exist. The survey of this literature by Michelle Drouin and Brandon McDaniel in the Oxford Handbook of Cyberpsychology (2018) gives the likeliest answer: interruptions by technology can spur negative feelings such as sadness, rejection and anger, and negative behaviors such as conflict and disengagement, in both romantic partners and children. What a person feels in that moment resembles being unseen.
This is where the literature on perceived partner responsiveness becomes relevant. In that framework, the feeling of being understood, validated and cared for does the heavy lifting in relationships. A partner who looks at a screen while you describe your day has not said anything hostile. They have failed, in a small and repeated way, to signal that you are being heard. A single lapse is invisible. A habit tells the other person where they rank, even if no one intended it.
One plausible reading of the paradox in the data, that phubbing is ubiquitous yet only weakly related to satisfaction, is that couples differ in how they absorb it. Some share the habit and share the norms. What probably hurts is the asymmetry: one person feeling ignored while the other does not notice, and the silence that follows when the feeling is never named. That is an interpretation, not a finding.
What this does not show
Three limits keep the headline honest. The first is direction. Cross-sectional correlations fit three stories equally well: phones hurt couples, distance drives people to their phones, or stress does both. The diary and tracking studies narrow the field without closing it.
The second is who was studied. Several of the key studies recruited parents of young children in the United States, and the earliest technoference survey recruited only women. Families with a newborn are tired, interrupted and under strain, so reliance on a phone may mean something different there than for a couple without children. The meta-analysis notes a near absence of research on vulnerable populations.
The third is meaning. Using a phone together, checking a message from a sick relative or looking up a restaurant for the couple’s evening are not the same act as scrolling away from someone. The measures cannot always tell them apart, and a ratio of time on a device says nothing about its function.
Using the evidence without turning it into a rule
The studies do not test interventions, so what follows is reasoning from their findings, not research results. If the problem is concentrated in the moments of shared attention, the most useful change is not a general phone limit but a few protected moments. Many couples find that the hardest places to keep are the first minutes after reuniting at the end of the day, meals, and the minutes before sleep.
- Name the moments rather than the device: decide which situations are for each other (a meal, a walk, the first ten minutes at home) and let the rest go.
- Say what you are doing when you do pick up the phone ("one second, I'm checking when the train leaves") so it is not read as withdrawal.
- Treat a partner's complaint as information about their experience, not an attack on your character. Nothing in the data suggests the person who feels phubbed is exaggerating.
- If the conversation about phones turns into a fight in its own right, start from the feeling ("I felt left out") rather than the accusation ("you're always on that thing").
A dispute about phones is a conflict like any other, and what the research on managing couple conflict says about repair applies to it as well: it matters less that the argument happens than how it is resolved.
The aim is not surveillance or a household ban. It is that, at the moments when it counts, the other person gets the sense of being chosen over the screen. That is a modest goal, and it fits what the better studies suggest. For a broader view of how couples keep each other’s attention in everyday talk, our guide to communication in a couple is a good place to continue.
Key takeaways in plain terms
The evidence supports a measured claim: more phone use during shared time goes with lower satisfaction, a modest effect that appears in surveys, a daily diary study and tracking data. It does not support a claim that smartphones cause breakups, and it says little about couples without children, same-sex couples or older couples. It also says that the quantity of screen time matters less than where it lands in a couple’s day. That is both less alarming and more useful than the common narrative.