It has become one of the most familiar explanations for modern unhappiness.
We spend more time on our phones, scroll through more social media and compare ourselves with more people than previous generations ever could.
So it seems intuitive that using these technologies more should gradually make us less satisfied with our lives.
A new longitudinal study suggests the relationship may be considerably weaker than that simple story implies.
Researchers followed 1,966 adults in the United States across five separate survey waves, measuring how frequently they used ten common social technologies and how satisfied they felt with their lives.
The assessments were repeated every three months over approximately one year.
The researchers then asked a particularly important psychological question.
If the same person started using a technology more frequently than usual, did their life satisfaction subsequently decline?
For the overwhelming majority of the technologies examined, the answer was essentially no.
The study, published in Nature Human Behaviour, found little credible evidence that increases in social-technology use predicted meaningful changes in life satisfaction three months later.
That does not mean every online experience is harmless.
It suggests something more specific and potentially more important: how often someone uses a technology may tell us far less about their psychological wellbeing than we commonly assume.
Comparing two people can give a very different answer from following one person
One of the central problems in social-media research is surprisingly easy to understand.
Imagine that people who use TikTok frequently report lower life satisfaction than people who rarely use it.
That relationship might suggest that TikTok reduces happiness.
But it could also mean that people who are already less satisfied with their lives are more attracted to TikTok.
Or the difference could be explained by something else entirely.
Age, employment, personality, social isolation, health, income or hundreds of other characteristics could affect both technology use and wellbeing.
A simple comparison between heavy and light users cannot easily separate these possibilities.
The new study therefore focused heavily on what happened within the same person over time.
If a participant normally used a particular technology occasionally but later began using it several times a day, the researchers could examine whether that individual’s life satisfaction subsequently changed.
This is a much stronger test of the idea that technology use itself gradually changes wellbeing.
The researchers checked participants every three months
Kostadin Kushlev, Kibum Moon, Matt Motyl, Nathanael Fast and Juliana Schroeder analysed data from the Understanding America Study, a longitudinal panel maintained by the University of Southern California.
Participants completed five rounds of measurements.
At each wave, they reported their life satisfaction and how frequently they used ten different social technologies.
Responses ranged from not using a technology at all to using it multiple times each day.
The technologies included different forms of digital communication and social-media use, including activities such as texting, calling and video calling, as well as platforms including YouTube and TikTok.
The researchers then used both Bayesian and conventional statistical models specifically designed to distinguish stable differences between people from changes occurring within individual participants.
That distinction ended up being crucial.
At first glance, some platforms did appear connected to happiness
When the researchers simply compared different people, several familiar patterns appeared.
People who reported texting more frequently tended to report higher life satisfaction.
People who used YouTube and TikTok more frequently tended to report lower life satisfaction.
If the analysis ended there, it would be tempting to conclude that texting is psychologically beneficial while video-based social platforms make people less happy.
But the longitudinal results did not support that interpretation.
When researchers tracked changes within the same individuals, using these technologies more frequently generally did not predict how satisfied those people felt three months later.
This suggests that some of the apparent relationships observed between users may say more about who uses a platform than what the platform subsequently does to them.
TikTok users being less satisfied does not mean TikTok made them less satisfied
This is one of the most important distinctions in behavioural research.
A correlation between two characteristics does not tell us which one caused the other.
Suppose people experiencing loneliness spend more time watching online videos.
A survey could then find that heavy video users report lower wellbeing.
But reducing those people’s video use would not necessarily remove the loneliness that contributed to the behaviour in the first place.
The new study tested whether changes in usage reliably came before changes in life satisfaction.
Across the technologies examined, the researchers found scant evidence that they did.
The longitudinal relationships were generally small, uncertain or absent.
The researchers also tested the relationship in reverse
Perhaps digital technology does not determine happiness.
Perhaps happiness determines digital behaviour.
The researchers therefore reversed the question.
If a person’s life satisfaction increased, did they subsequently change how often they used social technologies?
Again, most effects were negligible.
There was some evidence that increases in life satisfaction predicted modest increases in voice or video calling among particular demographic groups.
But the pattern was not sufficiently broad to support a simple claim that happier people generally become more socially active online.
The direction of influence appears more complicated than either “phones make people unhappy” or “unhappy people use their phones more”.
Why does this differ from so many alarming headlines?
Partly because studies are often answering different questions.
Some research examines teenagers.
This study examined adults.
Some studies measure depression or anxiety.
This study focused on overall life satisfaction.
Some examine time spent online during one particular week.
This research measured the frequency of technology use at three-month intervals.
Some investigate specific behaviours such as cyberbullying, appearance comparison or compulsive use.
This study examined relatively broad patterns of platform use.
These are not interchangeable outcomes.
A social-media interaction could temporarily worsen someone’s mood without meaningfully changing how satisfied they feel with their life three months later.
Likewise, a harmful experience for a vulnerable teenager cannot automatically be generalised to every adult who uses the same platform.
Earlier large studies have already suggested that average effects may be small
The new finding does not appear in isolation.
In 2019, psychologists Amy Orben and Andrew Przybylski analysed three large datasets containing more than 355,000 adolescents.
They found a negative relationship between digital technology use and wellbeing, but the association explained no more than approximately 0.4% of the variation in wellbeing.
That does not mean the relationship was nonexistent.
It means that technology use alone provided very little information about why one teenager reported better wellbeing than another.
More recent meta-analytic work has similarly questioned whether time spent on social media, by itself, provides a reliable explanation for mental-health differences.
The accumulating evidence increasingly suggests that asking only how much technology people use may be too crude.
An hour of social media is not one psychological experience
Consider two people who both spend an hour using the same social network.
One spends that time messaging a close friend who recently moved overseas.
The other scrolls through hundreds of carefully edited photographs of strangers while repeatedly comparing their appearance, wealth and lifestyle with their own.
The stopwatch records the same hour.
Psychologically, almost nothing else about the experiences is equivalent.
This is one reason researchers have increasingly questioned the usefulness of treating all screen time as a single exposure.
The platform itself may matter less than what someone is actually doing there.
Active and passive use may affect people differently
Digital communication can provide genuine social benefits.
Messaging allows people to maintain relationships over distance.
Video calls can preserve contact between families living in different countries.
Online communities can connect individuals who would otherwise struggle to find people with similar experiences or interests.
At the same time, some online behaviours may encourage comparison, conflict or compulsive checking.
These possibilities can coexist.
Trying to estimate the psychological effect of “social media” without considering what someone actually does on it may be similar to asking whether “talking to people” is good or bad for mental health.
The answer depends heavily on who the person is talking to and what happens during the conversation.
Why texting was associated with greater life satisfaction
The between-person association involving texting is particularly interesting.
People who texted more frequently tended to report greater life satisfaction.
One possible interpretation is that frequent texting reflects stronger social relationships.
Someone who regularly communicates with friends, partners and family may simply have a larger or more active social network.
But again, the longitudinal analysis provides an important warning.
Increasing how often an individual texted did not reliably produce a subsequent improvement in life satisfaction.
Texting may therefore function partly as an indicator of someone’s existing social life rather than being the cause of their happiness.
The same logic may apply to TikTok and YouTube
Participants who used YouTube and TikTok more frequently tended to have lower life satisfaction when compared with other people.
That relationship is worth studying.
But because increases in an individual’s use did not reliably predict later declines in life satisfaction, it cannot simply be interpreted as evidence that these platforms reduced wellbeing.
People with different lifestyles may use platforms differently.
Age could matter.
People with more free time might consume more video content.
Someone who feels socially disconnected may turn to entertainment more frequently.
The content itself could also differ enormously between users.
One person may spend time watching educational material.
Another may consume highly emotional or appearance-focused content.
Frequency alone does not capture those differences.
The study does not prove social media is harmless
This is the most important limitation.
The absence of a strong average relationship between technology-use frequency and life satisfaction does not show that nobody is harmed by social media.
An average can hide important differences between individuals.
A platform may have almost no effect on most people while having substantial effects on a smaller vulnerable group.
People experiencing eating disorders, bullying, addiction-like use patterns or severe social comparison may respond very differently from the average participant.
Specific content could also produce effects that disappear when thousands of different online behaviours are collapsed into one usage-frequency variable.
The researchers measured frequency rather than actual minutes
The study also relied on self-reported use.
Participants indicated whether they used technologies never, infrequently, daily or multiple times per day.
That is different from obtaining objective logs directly from smartphones.
Someone who opens an application five times for a total of ten minutes and someone who opens it five times for three hours could receive similar frequency classifications.
People are also imperfect at remembering their own digital behaviour.
Future studies using passive device measurements could provide much greater precision.
Three months may be too long and too short at the same time
The researchers examined whether changes in technology use predicted life satisfaction approximately three months later.
That timeframe makes sense for studying relatively durable changes in wellbeing.
But psychological effects may operate on very different timescales.
A hostile online interaction could change someone’s mood for the next hour without affecting them three months later.
Alternatively, effects could accumulate gradually over several years and remain difficult to detect across a one-year study.
The researchers therefore cannot rule out relationships occurring faster or slower than the intervals they measured.
People also did not change their technology habits very much
Another limitation arose from the stability of participants’ behaviour.
Most people did not move dramatically between being very light and very heavy users over the study period.
That makes it statistically harder to determine what would happen if someone’s digital life changed radically.
The findings therefore describe the relatively ordinary fluctuations that occurred naturally among adults.
They do not tell us precisely what would happen if a person who never used social media suddenly began spending six hours a day on it.
Life satisfaction is not the same as mental health
The wording matters here too.
Life satisfaction is a broad evaluation of how someone feels about their life overall.
It is related to psychological wellbeing but is not identical to depression, anxiety, loneliness, self-esteem or moment-to-moment mood.
A technology could plausibly influence one of those outcomes without producing a detectable change in overall life satisfaction.
The new study therefore answers an important question, but it does not answer every question about digital mental health.
The results are particularly relevant to digital detox claims
Digital detoxes are frequently promoted on the assumption that simply reducing screen or social-media exposure will make people happier.
The latest findings suggest that expectation may be too strong, at least for many adults.
If a person’s dissatisfaction is primarily driven by work, finances, relationships, health or loneliness, deleting an application does not automatically address the underlying problem.
For some people, reducing a particular form of technology use may still be beneficial.
But the reason probably matters.
Someone who stops compulsively comparing themselves with influencers may experience a different outcome from someone who removes the messaging application they use to communicate with close friends.
The better question may be “what are you doing online?”
The debate around technology and wellbeing has often concentrated on quantity.
How many hours?
How many notifications?
How many times did someone open an application?
Those numbers are easy to measure.
Psychologically, they may not be the most informative ones.
Researchers may need to know whether use is active or passive, social or solitary, supportive or hostile, intentional or compulsive.
They may need to know what content the person sees, who they interact with and what they could have been doing instead.
Two hours spent maintaining relationships is different from two hours spent doomscrolling.
A single number labelled “screen time” cannot easily distinguish the two.
The study also offers a useful lesson about correlation
Psychology frequently studies relationships that cannot easily be manipulated experimentally.
Researchers cannot realistically assign thousands of people to use TikTok heavily for several years simply to see whether their lives become worse.
This means much digital-wellbeing research depends on observational data.
The new study demonstrates why the distinction between between-person and within-person relationships matters so much.
Different people can show a strong association even when changing the behaviour within one person produces little measurable change.
That is not merely a statistical technicality.
It changes the story.
“People who use TikTok more are less satisfied” and “using TikTok more makes people less satisfied” may sound similar.
Scientifically, they are very different claims.
The psychology of technology may be more individual than universal
The desire for a simple answer is understandable.
People want to know whether social media is good or bad for them.
Parents want rules.
Policymakers want thresholds.
Technology companies want evidence that their products are harmless.
Critics want evidence that they are not.
Psychology increasingly suggests that the reality may resist this binary.
Technology can connect people and isolate them.
It can entertain and distract.
It can provide support and expose users to comparison or abuse.
The same platform can do all of these things to the same person on the same day.
The most interesting finding may be how little the needle moved
The researchers followed nearly 2,000 adults through five measurements and tested multiple forms of digital communication.
They looked in both directions.
Did technology use predict later life satisfaction?
Did life satisfaction predict later technology use?
Across most analyses, very little happened.
That null result is scientifically valuable.
It does not absolve technology companies of responsibility for harmful design or dangerous content.
Nor does it invalidate people who feel that particular online experiences have affected their mental health.
Instead, it challenges the assumption that frequency of use alone is a powerful psychological force capable of explaining why adults become more or less satisfied with their lives.
The question may no longer be simply whether we spend too much time online.
It may be what happens to us while we are there.
Source Information
Primary Study: Social technology use and life satisfaction in a five-wave panel study of US adults
Authors: Kostadin Kushlev, Kibum Moon, Matt Motyl, Nathanael J. Fast and Juliana Schroeder
Journal: Nature Human Behaviour
Published: 28 August 2026
Participants: 1,966 US adults
Design: Five-wave longitudinal panel study
Measurement interval: Every three months
Technologies: Ten common social technologies
Main finding: Little credible evidence that within-person increases in social-technology use predicted subsequent life satisfaction
DOI: 10.1038/s41562-026-02564-8
Supporting Study: The association between adolescent well-being and digital technology use
Authors: Amy Orben and Andrew K. Przybylski
Journal: Nature Human Behaviour
Published: 2019
Participants: 355,358 adolescents across three large datasets
Key finding: Digital technology use showed a small negative association with wellbeing, explaining at most approximately 0.4% of variation
DOI: 10.1038/s41562-018-0506-1
Supporting Meta-analysis: Do Social Media Experiments Prove a Link With Mental Health: A Methodological and Meta-Analytic Review
Author: Christopher J. Ferguson
Journal: Psychology of Popular Media
Volume: 14
Pages: 201–206
Published: 2025
DOI: 10.1037/ppm0000541
