Topics / Relationships
Loneliness and social connection
Published November 20, 2026. 4 sources at the foot of the page.
What is loneliness?
Loneliness is the feeling of having less connection with other people than a person wants, whatever the number of people around them. It is about the gap between the company somebody has and the company they are looking for.
Where you might meet it
- Somebody moves to a new city for a job, is busy all day among friendly people, and finds the evenings longer than they used to be.
- A man with a full calendar and hundreds of contacts notices that nobody on it knows how his week actually went.
- A woman who lives alone by choice, and likes it, cannot think of anyone she is lonely for.
Cacioppo and Patrick (2008) describe loneliness as a signal rather than a flaw. Hunger says the body needs food, thirst says it needs water, and on their account loneliness says that a social animal has drifted further from others than it can afford.
The signal is uncomfortable for the same reason hunger is: discomfort is what gets a creature to move. Read that way, feeling lonely is not evidence that something is wrong with a person, it is evidence that the person is one.
Hawkley and Cacioppo (2010) put the definition the same way, as the distance between the relationships a person wants and the ones they have. That makes loneliness a matter of fit rather than of count, which is where the next chapter starts.
Two things that are easy to confuse
Being alone is a count: how many people are in the room. Being lonely is a feeling about fit. The two come apart in both directions, and the research depends on keeping them apart.
Weiss (1973) separated two kinds of loneliness from the accounts people gave him. Emotional isolation is the absence of one close attachment, the person somebody would call first. Social isolation is the absence of a wider circle, the people who would notice if somebody stopped turning up.
Weiss reports that the two do not stand in for each other. A person with a close partner and no friends can be lonely for the circle, and a person with a full circle and nobody close can be lonely for the one.
Hawkley and Cacioppo (2010) report that how many people somebody sees predicts loneliness only weakly, and that how those relationships feel to the person predicts it much better. Somebody in a crowded office can be lonely and somebody in a cabin need not be.
That is why counting contacts, meetings or followers says so little. The signal is about the relationships a person wants and does not have, and only the person can say which those are.
A signal that can work against itself
Cacioppo and Patrick (2008) describe a paradox at the center of the feeling. Loneliness is meant to send a person back toward others, and it also makes them warier of others while it lasts.
Hawkley and Cacioppo (2010) review the studies and report that lonely people attend more to negative social information, read ambiguous faces and remarks as less friendly, and expect rejection more readily than people who are not lonely.
The wariness has a logic. For a social animal on its own, a stranger is a danger before it is a friend, so the signal turns up the watch for threat at the same time as it turns up the wish for company.
The cost is a loop. Somebody who expects a cold reception reads a neutral one as cold, holds back, and gets a cooler one next time, and the reading looks confirmed. Hawkley and Cacioppo describe the pattern as one that can hold loneliness in place once it has started.
None of this is a fault of character. It is what the signal does, and the reason the authors describe it as a signal is that it behaves like one: it is loud, it is uncomfortable, and it shapes what gets noticed.
What Holt-Lunstad, Smith and Layton gathered
Holt-Lunstad, Smith and Layton (2010) collected every study they could find that had measured people’s social relationships and then followed them for years to see who was still alive. That came to 148 studies and more than three hundred thousand people, followed for about seven and a half years on average.
Across those studies, people with stronger social relationships were more likely to be alive at follow-up than people with weaker ones. The authors put the size of that association alongside well known risks such as smoking, and above physical inactivity and obesity.
The relationships were measured in many ways, from whether somebody lived alone to how supported they said they felt, and the studies that used richer measures of a person’s ties found the stronger association.
What a review like this can and cannot say
Every study in it watched people rather than assigning them to anything, so the review reports an association. Nobody was made lonely or made connected and then followed, and the authors say so.
Holt-Lunstad and colleagues report that the association held when they took account of age, sex, health at the start and cause of death, which rules out some of the easy explanations without ruling out all of them.
What is left is a large, consistent association between having people and staying alive, described here in words because that is what the evidence supports: connection travels with health, across many studies, in a way the authors think deserves the same attention as the risks everybody already knows.
Limits, and two places this connects
Loneliness is common, it comes and goes, and on the account here it is a signal rather than a verdict on a person. This course describes the signal and what the research reports about it, and it hands nobody a plan.
- A signal model is a model. Cacioppo and Patrick built it from what is known about social animals and from studies of people, and it is a way of reading the feeling rather than a proof of where it came from.
- The review of relationships and health is an association across studies that measured different things. It does not say that any one person’s ties will change what happens to them.
- Much of the research on loneliness comes from wealthy countries and from people willing to answer questionnaires about it, and loneliness is not the same everywhere.
Nothing here promises anybody company, and nothing here says what any reader should change. Wanting more connection than one has is, on this account, ordinary equipment.
Where to go next on the path
The Attachment unit in Know yourself is about the close bonds Weiss had in mind when he wrote about emotional isolation, and about how a person tends to hold them.
Self-disclosure and closeness, on this track, describes how closeness is built between two people a layer at a time, and What predicts a relationship lasting describes what researchers saw when couples who had it argued.
Sources
- Cacioppo, J. T., & Patrick, W. (2008). Loneliness: Human Nature and the Need for Social Connection. Norton.
- Hawkley, L. C., & Cacioppo, J. T. (2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218-227.
- Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316.
- Weiss, R. S. (1973). Loneliness: The Experience of Emotional and Social Isolation. MIT Press.
Practice it on Learn
Learn teaches loneliness in short games and lessons, with the same research as this page.
Practice lonelinessSee what Learn is
Your first lesson is free and needs no account. For adults 18 and over.
More topics
- Self-disclosure and the 36 questions
Self-disclosure is the process of telling another person something true about oneself that they could not have known otherwise.
- What predicts a relationship lasting
A conflict pattern is a habit of arguing that a couple falls back into whenever the same disagreement comes around.
- Effect sizes and p values, explained
An effect size is the amount of a difference or a relationship a study found, stated in original units or as a number of standard deviations.