The tragedy of loneliness

10 May 2022

We know that about 10% of the population feel lonely at any one time (Department for Culture, 2020). While it is accepted that older people are at high risk of social isolation, loneliness and, thus, poor mental and physical health, there is now more recognition of the impact of loneliness on children and young people too. Young people, from adolescence to early adulthood, attach particularly high value to close friendships and romantic relationships. Loneliness is particularly common at this stage of life. 

During lockdown, young people aged 11 to 16 years old who reported higher levels of loneliness also had more mental health problems (Cooper et al., 2021). Primary school children who lacked social connectedness, and those children without access to the internet, showed the most emotional problems (Skripkauskaite et al., 2021) (Speight et al., 2021). The UCL Covid survey of 55,000 people also pointed to loneliness levels being higher amongst younger adults and it was strongly linked to mental health concerns (Henderson et al., 2020). How common loneliness is varies with age, but at all ages it is associated with mental health problems. Indeed, in one study it is linked to at least 90% more depression, 20% more anxiety, and 30% more suicidal ideation (Beutel et al., 2017). 

But what is it about loneliness that hurts, and even kills? What is it that makes it so dangerous to our physical and mental health?

An adaptive response?

Our species evolved for social living, which needs larger brains to develop after passing through the birth canal – which, in turn, means that we are brought into this world quite helpless and relatively immature. As infants, we are all dependent on care for longer than any other primate. Birth itself is painful and dangerous. It all adds up to the need for help from kin. Therefore, the feeling of loneliness may be a helpful and adaptive response to being socially disconnected and it provides us with a drive to re-join with social groups. 

Of course, when these predispositions evolved, we were hunter-gatherers. This was perhaps 100,000 years ago, and before the development of agriculture dramatically altered human lifestyle and priorities. Typical groups of early humans were relatively small, maybe up to 150 individuals. Familiarity and reciprocal relationships characterised the group which – in addition to close relatives – was responsible for protecting each young human (Dunbar, 2016). Relatives and other group members were present 24/7. Children were able to roam freely, to seek contact and comfort, and to play with whoever they chose. No hunter-gatherer walked alone.

Social isolation is not quite the same as loneliness. Social isolation describes an objective situation in which social connections are limited or absent. It is about there being friends, or just one special person, who is around when needed. Loneliness, by contrast, is a subjective feeling of distress – when an individual feels the potential for connections available to them are inadequate or unfulfilling. Loneliness is about feeling that one lacks companionship, or feeling left out, feeling isolated from others, feeling alone, or feeling we’re missing out. While isolation and loneliness often co-occur, importantly, they can be experienced independently of one another (Matthews et al., 2016). 

Individuals who are frequently socially isolated do not always feel lonely. More importantly, the availability of ample social opportunities does not preclude experiencing acute loneliness. Data from the age-18 wave of the Environmental Risk Longitudinal Twin Study of over 2,200 young people showed that depression was significantly correlated with both social isolation and loneliness, but the feeling of loneliness explained 70% of the effect of social isolation (Matthews et al., 2016). From this and other studies, it appears that young adults’ feelings of loneliness are more strongly associated with their experience of depressive symptoms than their experience of social isolation.

Seeing oneself as socially isolated (i.e., loneliness) triggers feelings of vulnerability which are perhaps linked to a sense of being cut off from one’s social groups. It provokes a range of physiological, behavioural and cognitive responses geared towards self-protection involving both brain and mind. Multiple brain areas are associated with perceived social isolation, including areas concerned with mentalizing, identifying thoughts and feelings that might explain someone’s actions (the temporoparietal junction), and areas indicating enjoyment and reward (ventral striatum) (Cacioppo, Capitanio, & Cacioppo, 2014).  When lonely people look at negative pictures of people, they avoid thinking about how they might be feeling, and appear to focus on their visual features and not engage their capacity to think about social experience. 

Lonely people appear to be more likely to activate reward brain areas for pleasant images of non-human physical objects than pleasant images of people. Their body indicates the typical impact of long-term stress, with higher blood pressure and activation of flight/fight emergency response, but also lower inflammatory control and immune response. These physical reactions might explain the 26% increased mortality associated with loneliness and the 29% associated with social isolation. Loneliness and social isolation are particularly impactful at a younger age (under 65, they are highly predictive of premature death) (Holt-Lunstad et al., 2015). One study which followed 40,000 fathers over a period of 15 years – and adjusted for other risk factors (age, ethnicity, education, etc) – observed that 63% of single fathers died over the period, compared with about 35% of partnered fathers (Chiu et al., 2018). 

The years ahead

Social isolation in childhood may also be critical for physical and mental health in the years ahead. A study of health harming behaviours such as smoking, poor diet and alcohol consumption found that adverse childhood experiences greatly increased the risk of such behaviours, especially in groups experiencing high levels of social deprivation (Holt-Lunstad et al., 2010). However, the presence of ‘a trusted and always available adult’ during childhood cut the impact of such adversity by 66% (Bellis et al., 2017). Perhaps childhood adversity is especially likely to turn into trauma when the child feels alone.  

But the cognitive changes are equally important. Lonely individuals are more likely to lack trust, to look at those around them with suspicion, to see people more negatively, to be less hopeful about the outcomes of social interactions, and to prioritise self-protection by being defensive or even hostile upon entering a social encounter. These strategies may be understandable as ways of minimising potential risks from social interactions, but they bring with them two major life-limiting consequences. 

First, obviously these strategies are likely to ensure that nothing changes, that positive social relationships continue to evade them. There is a self-fulfilling quality in this anxious and pessimistic stance towards social relationships. Loneliness becomes the cause of further isolation, as opportunities to develop positive social relationships are lost. Each of these missed opportunities provide further evidence for what the person experiences as the hopelessness of ever arriving at a positive social exchange. It has the potential of creating the deepest of grooves, the walls of which ultimately defy scaling.

The second aspect is perhaps even more pernicious, but harder to explain. The social nature of our beings has made humans unique amongst species in efficiently passing knowledge from one generation to the next, enabling us to accumulate a store of knowledge called culture. Each generation can learn from the last. Young humans need to be able to learn and benefit from their social context if they are to thrive (Wilson et al., 2014). But in a social network as complex as ours, vigilance is advisable (Sperber et al., 2010). We need to be able to work out quickly who to trust so they can help us learn the ropes. How does this happen? To protect the naive learner from acquiring inaccurate information, evolution has equipped us with a gating mechanism, a capacity to open our minds to learning only when learning new knowledge is safe. This is called epistemic trust. 

Epistemic trust means a person’s ability to trust appropriately in social sources of new knowledge (Fonagy, Luyten, & Allison, 2015). There is a need for vigilance to ensure that we are only open to learning when the source of knowledge is trustworthy. Humans – in order to learn about the world, about themselves and about what they need to know – need to trust the knowledge which is being communicated to them. It’s a person’s openness to the social context that enables change and adaptation to circumstance, namely to learn. Human communities, families, schools, workplaces need to create an environment that ensures the openness that epistemic trust affords. Of course, the experience of adversity is bound to undermine epistemic trust and may even be a helpful adaptation to a hostile environment (Fonagy et al., 2021).

Jointly seeing to it…

So, what opens the gate in our minds so that we are ready for social learning? Human sociality is explained by the unique capacity to share the mental states of others. When people are poised to interact, they achieve interpersonal awareness through a ‘meeting’ of minds. Mental states are assumed by individuals in the social system to be joint or shared by everyone. The Finnish philosopher, Raimo Tuomela, has named this category jointly seeing to it (jstit) (Tuomela, 2005). Feeling of we-ness initiates social collaboration. It is a moment of understanding each other and creating a sense of being part of a set of thoughts and feelings that are beyond one’s own, in an irreducibly collective mode of cognition that neuroscientists have called the we-mode. 

The we-mode opens the gates of the mind for social learning – we-ness triggers epistemic trust. The tragedy of a sense of loneliness is the inaccessibility of the we-mode experience, and consequently an inability to change to update how one sees oneself and how one sees others. We would argue that the close link between the feeling of loneliness and mental ill health arises because, when we feel isolated (whether we actually are or not), we find ourselves creating social experiences of failed collaboration. These experiences serve as self-fulfilling prophecies, and so we are deprived of the experience of epistemic trust which is necessary to open our minds to fresh understandings, new ways of looking at things. 

What does all this mean for policy and practice for young people’s mental health? These observations suggest that we should focus on interventions to decrease feelings of loneliness that can, in turn, improve mental health. As loneliness can be experienced even without social isolation, simply increasing individuals’ amount of social contact may be insufficient for improving outcomes. Just providing social support and increasing social contacts, even creating superficial friendships, will not address loneliness. 

We need to be more subtle and sophisticated in our approaches. We need to address the issues of mistrust and suspicion that can engulf the thinking of a young person. We need to understand that they may be feeling misunderstood and isolated because, when they experienced adversity, there was no one around to provide support. Interventions that focus on relationships, such as interpersonal therapy, are highly effective and much needed in reducing depressive symptoms in children and young people.

- Professor Peter Fonagy is Chief Executive of the Anna Freud Centre. Find more from him in our archive; plus more on loneliness.


Bellis, M. A., Hardcastle, K., Ford, K., Hughes, K., Ashton, K., Quigg, Z., & Butler, N. (2017). Does continuous trusted adult support in childhood impart life-course resilience against adverse childhood experiences - a retrospective study on adult health-harming behaviours and mental well-being. BMC Psychiatry, 17(1), 110. 

Beutel, M. E., Klein, E. M., Brahler, E., Reiner, I., Junger, C., Michal, M., . . . Tibubos, A. N. (2017). Loneliness in the general population: prevalence, determinants and relations to mental health. BMC Psychiatry, 17(1), 97. 

Cacioppo, S., Capitanio, J. P., & Cacioppo, J. T. (2014). Toward a neurology of loneliness. Psychological Bulletin, 140(6), 1464–1504. 

Chiu, M., Rahman, F., Vigod, S., Lau, C., Cairney, J., & Kurdyak, P. (2018). Mortality in single fathers compared with single mothers and partnered parents: A population-based cohort study. Lancet Public Health, 3(3), e115-e123. 

Cooper, K., Hards, E., Moltrecht, B., Reynolds, S., Shum, A., McElroy, E., & Loades, M. (2021). Loneliness, social relationships, and mental health in adolescents during the COVID-19 pandemic. Journal of Affective Disorders, 289, 98-104. 

Department for Digital, Culture, Media and Sport. (2020). Wellbeing and loneliness: Community life survey 2019/20.

Dunbar, R. (2016). The social brain hypothesis and human evolution. In Oxford Research Encyclopedia for Neuroscience (pp. 1-33). Oxford: Oxford University Press. 

Fonagy, P., Campbell, C., Constantinou, M., Higgitt, A., Allison, E., & Luyten, P. (2021). Culture and Psychopathology. Development and Psychopathology, 1-16. 

Fonagy, P., Luyten, P., & Allison, E. (2015). Epistemic petrification and the restoration of epistemic trust: A new conceptualization of borderline personality disorder and its psychosocial treatment. Journal of Personality Disorders, 29(5), 575-609. 

Henderson, M., Fitzsimons, E., Ploubidis, G., Richards, M., & Patalay, P. (2020). Mental health during lockdown: Evidence from four generations

Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science, 10(2), 227-237. 

Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: a meta-analytic review. PLoS Med, 7(7), e1000316. 

Matthews, T., Danese, A., Wertz, J., Odgers, C. L., Ambler, A., Moffitt, T. E., & Arseneault, L. (2016). Social isolation, loneliness and depression in young adulthood: a behavioural genetic analysis. Social psychiatry and psychiatric epidemiology, 51(3), 339-348. 

Skripkauskaite, S., Shum, A., Pearcey, S., Raw, J., Waite, P., & Creswell, C. (2021). Changes in children’s and young people’s mental health symptoms: March 2020 to January 2021

Speight, S., Taylor, I., Taylor, B., Kolbas, V., Smith, N., Bristow, T., . . . NatCen Social Research. (2021). Study of Early Education and Development (SEED): Findings from the Coronavirus (COVID-19) follow-up

Sperber, D., Clement, F., Heintz, C., Mascaro, O., Mercier, H., Origgi, G., & Wilson, D. (2010). Epistemic vigilance. Mind & Language, 25(4), 359-393. 

Tuomela, R. (2005). We-intentions revisited. Philosophical Studies, 125(3), 327-369. 

Wilson, D. S., Hayes, S. C., Biglan, A., & Embry, D. D. (2014). Evolving the future: Toward a science of intentional change. Behavioral and Brain Sciences, 37(4), 395-416.