The Sardinian Secret: Why Face-to-Face Contact Is a Health Intervention

TLDR:

• Face-to-face interaction has distinctive biological effects, including oxytocin release and cortisol regulation, that some forms of digital contact, such as text messaging, may not replicate.
• Strong social relationships are associated with substantially higher odds of survival, an effect that some research says is comparable with the health impacts of quitting smoking.
• Digital contact creates an illusion of connection that can substitute for physical presence, but which may not provide the same biological benefits as actual face-to-face interaction. This is a substitution fallacy with real consequences for health.
• The effort that sustains relationships (the travelling, reaching out, forgiving, compromising, taking risks, being there on someone else’s agenda) benefits other people and our future selves, so we often undervalue it (classic miswanting).
• Consequently, we’ve been busy designing it out of our lives, something we think is a considerable mistake.
• Like much of what we talk about, we think this is an example of acting without considering externalities. In this instance, we’ve been working hard to alleviate minor present discomfort at the expense of significant future positive externalities.

The secret to living past 100 in Sardinia is not the diet. It is who comes to visit.

The Barbagia region of Sardinia is rugged, isolated mountain country in the island’s interior, cut off from the coastal towns by roads that wind steeply through rock and scrub. It is not wealthy. Its healthcare is not exceptional. Its diet is broadly Mediterranean, much like the rest of the island. And yet its villages are reported to have the highest concentration of male centenarians anywhere in the world. The men here live, consistently and reliably, far longer than men in comparable populations. They also live far longer than men in wealthy, technologically advanced countries with excellent healthcare systems.

This is the puzzle at the heart of Susan Pinker’s 2014 book The Village Effect. Her answer is that the men of Barbagia aren’t biologically special and don’t eat anything unusual. What sets them apart is their daily social life: how dense it is, how regular, and how physical. They live inside networks of face-to-face contact that modern, urban, digitally connected life has largely dismantled. Pinker argues that this isn’t incidental to their longevity. She thinks it is the main cause.

The Sardinian Puzzle

Sardinia’s status as a “Blue Zone” (one of a handful of regions worldwide with unusually high concentrations of people who live past 100) has attracted considerable research attention. There has even been a Netflix documentary of Blue Zones. The data behind the Blue Zones are contested, though: the UCL demographer Saul Newman has argued that records of extreme old age are often unreliable, while Blue Zone researchers say they verified each centenarian’s dates. It is worth treating the exact figures with some caution. Much research has focused on the Mediterranean diet or similar, physical activity patterns, and genetic factors. These are not irrelevant. But Pinker thinks the diet-focused research underweights something else: how social life in Barbagia is organised.

The typical centenarian in Barbagia doesn’t live alone or in a care home. They live close to their children, grandchildren, siblings and lifelong friends, and people visit them daily. Contact isn’t a special occasion; it’s a daily rhythm. The shop, the church, the piazza, the neighbour’s kitchen: none of these are destinations that need planning. They’re simply where the day happens.

Pinker contrasts this with modern urban and suburban life in wealthy countries, where family and close friends are increasingly far away, where people may have hundreds of digital connections and very few regular face-to-face relationships, and where social life has shifted from being in the same room to being in touch. The contrast isn’t subtle, and she argues the health consequences are just as stark.

What Face-to-Face Contact Actually Does

The biological case for the particular importance of physical presence is substantial. It rests on mechanisms that some forms of digital contact, text messaging especially, may not activate. Face-to-face interaction triggers the release of oxytocin, the hormone linked to bonding and trust. Touch plays a part, and so does eye contact. So does the automatic reading of micro-expressions and posture that goes on below conscious awareness whenever we are physically with another person.

Cortisol, the main stress hormone, is reliably reduced by face-to-face contact with people we trust. Having a friend or family member with you during a stressful experience measurably changes your cortisol response. Whether other kinds of contact can do the same is less clear. In a small study by Seltzer and colleagues, girls aged roughly 7 to 12 who phoned their mothers after a stressful task had stress-hormone and oxytocin responses similar to those who were with their mothers in person, while girls who instant messaged them did not, which suggests that the sound of a familiar voice matters. The pathways involve the vagus nerve (a major conduit for bidirectional communication between the brain and body) and the autonomic nervous system, and being physically together may add to the effect: reading the other person’s body, being read by them, and the subtle attunement that happens between people in the same room.

The case also seems to extend to gene expression (the process our cells use to turn the instructions written in our DNA into a working product, usually a protein). Face-to-face contact affects how genes involved in immune function and inflammation are expressed, and social isolation is associated with patterns that leave us more vulnerable to infection, inflammation and a range of chronic conditions. In effect, the body treats being with other people, or not, as information about its environment, and adjusts accordingly.

The Relationships That Matter

One finding stands out: how strongly our relationships predict how long we live. Julianne Holt-Lunstad and colleagues’ 2010 meta-analysis of 148 studies and more than 300,000 people found that those with stronger social relationships had about 50% higher odds of survival, and those scoring high on composite measures of social integration about 91% higher. The authors compare the size of the effect to quitting smoking, and say it exceeds that of obesity and physical inactivity. The studies are observational, so causation is hard to establish, but the pattern held across age, sex, initial health and cause of death. In her TED talk, Pinker presents close relationships and everyday social contact as the strongest predictors on Holt-Lunstad’s list, ahead of items such as smoking, exercise and weight.

This cuts against how we normally think about health. We have put a huge amount into promoting physical health behaviours as the way to live longer. Nobody contests that smoking kills or that exercise extends life, and the public health machinery around both is large. But the evidence suggests that social relationships predict how long we live about as strongly as smoking does, and they get a fraction of the attention. Or at least they used to, though we think that is changing.

Note what kind of relationships we are talking about here. Pinker names two things near the top of her list. One is close relationships: a handful of people who are genuinely close, involve regular physical contact, and bring mutual knowledge and care. The other is social integration, the everyday contact with other people as you move through the day. Deep social support is not easy to maintain, and it becomes harder as people age, move cities, change jobs, and lose the institutional structures such as school, university and the workplace that once scaffolded their social lives. Writing this in my mid-forties while working for myself, I’m painfully aware of all of the social scaffolding that these organisations used to provide to me.

Why Video Isn’t Enough

The natural response to the evidence about face-to-face contact is to ask whether video contact provides an adequate substitute. This is a particularly salient question given the huge jump to digital connection we saw during the pandemic. Unfortunately though, the answer is no. Video contact does not provide an adequate substitute for face-to-face contact. Understanding why it does not is important for anyone making decisions about how and where people work and connect. It’s actually also important for anyone trying to shape a good life for themselves or others.

Video calls reproduce some of what happens face to face. You can see the other person’s face and read some of their expressions, and it’s closer to a real conversation than a phone call or a text exchange. But a lot is missing. There is no touch. There is no smell, a sense that plays a surprisingly significant role in social bonding and emotional communication. Eye contact over video is close to impossible: looking into someone’s eyes means looking at the camera rather than the screen, so it feels performed rather than natural. The micro-expressions and posture cues our nervous systems read automatically in the same room are degraded or absent, and the vagal and autonomic responses that physical proximity seems to trigger don’t appear to switch on.

Pinker’s argument is that video keeps a relationship alive better than no contact does, but doesn’t deliver the health benefits of being together. That’s no reason to despair about technology. It’s a reason to be precise about what it can and can’t replace.

The Substitution Problem

The danger Pinker identifies is what we might call the substitution fallacy: treating digital contact as equivalent to physical contact because it feels like connection, meets some of our immediate social needs, and is so much more convenient. We video-call instead of visiting and message instead of meeting (I’m 100% guilty of this). We maintain the sense of connection, at least the knowledge that someone is there, the exchange of information and humour and affection, all without the physical presence that appears to drive many of the health effects. In doing this we’ve removed the “friction” in these relationships, without realising that the “friction” is where the benefit often lives.

The substitution is seductive because the costs don’t show. Someone with a rich digital social life and a thin physical one doesn’t feel the health consequences straight away. Any damage is gradual, cumulative and hard to trace to its cause. Take someone who moves to a new city and keeps their friendships going over video calls and group chats. They have conversations, they laugh, they feel cared for. What they are missing, the regular face-to-face contact with close others that the evidence associates with measurably better health outcomes, is not an obvious absence. It’s a low-level deficit that builds over years. It’s perhaps like swapping regular soda for diet soda: you get the taste, without the calories.

Social media is the purest version of the swap: the sugar without the substance. A “like” takes a fraction of a second. Scrolling through a friend’s holiday photos feels like keeping up with them, and watching strangers live their lives can feel a lot like company. None of it asks anything of us; there’s no effort, which is both the appeal and the problem. It takes the edge off the itch to feel connected, so the itch to go and actually see someone quietens down.

The research on social media and wellbeing is considered mixed, though I very much believe that social media has a material net negative impact. One specific study found that receiving personal messages from close friends went with improvements in wellbeing, while one-click reactions and passive scrolling didn’t. The effortful kind of contact was the kind that helped.

I think most people already know this and feel it instinctively, and many people already sense the negative impact social media has on them and others. In fact, a study of more than 1,000 US college students found that the average user would need to be paid around $50 to stay off TikTok or Instagram for four weeks, yet would on average pay $24 (TikTok) or $6 (Instagram) for all the participating students at their university, themselves included, to come off.

Unfortunately, the substitution of online for face-to-face connection is also no longer just a personal choice. Geographic mobility, remote working, the decline of local community structures and the loss of third places all shrink the opportunities for face-to-face contact, and technology, by making the substitute feel adequate, takes the pressure off fixing the loss. Of course, the addictive nature of social media also draws us to that saccharine digital world.

The Hidden Value of Friction

Almost everything that makes close relationships valuable also makes them a bit of a hassle. Staying in touch takes effort. Seeing someone means travelling, rearranging a week and giving up an evening we would rather spend on the sofa. Friendship involves risk, forgiveness and sacrifice, none of which are comfortable or convenient. That is all friction, and we are very good at spotting it and very good at avoiding it. Two things seem to be going on that I want to touch on: externalities and “miswanting”.

Firstly, externalities. A positive externality is a benefit that lands materially on someone other than the person paying the cost, so it doesn’t really show up in an actor’s decision-making process. Relationships are full of them. When I turn up, reach out or forgive, the other person benefits, the relationship strengthens and the wider group becomes a slightly better place to belong to. I feel little of that while deciding whether to bother. Even my own payoffs arrive late: the health benefits of strong relationships build over years, while the cost of this evening is due tonight. Immediate, personal costs and delayed, shared benefits mean people under-invest. It’s the worry behind Robert Putnam’s Bowling Alone too: connection behaves like a public good, and public goods often get under-provided.

Secondly, there is miswanting, the term psychologists Daniel Gilbert and Timothy Wilson use for the gap between what we think will make us happy and what actually does. It shows up most when we choose between effort and ease. The phone is right there; going out into the cold to meet people is not. And our forecasts of the social option are consistently too gloomy. In experiments by Nicholas Epley and Juliana Schroeder, Chicago commuters predicted that talking to a stranger would be the worst way to travel. Those who did it reported the most pleasant journey. Peggy Liu and colleagues found something similar with reaching out: we brace for awkwardness, but the other person is usually more pleased than we expect (we’ve written more about this in You Should Talk to Strangers. Science Says So.) So we skip the effort because we overrate its costs and underrate its rewards, and we are not even the only ones who lose out.

Effort Is the Message

Take a birthday card. A Moonpig card is perfectly nice: personalised, delivered, no stamp needed. But the value of a card was never really the card. It was the effort that went into it: remembering the date, choosing something that suits the person, writing something that sounds like you, finding a postbox. Take the effort away and what’s left is closer to a notification.

Online dating is the same story at scale. Apps have taken a great deal of friction out of meeting people, which is genuinely useful. But presenting people as profiles to be compared tips us into an assessment mindset, where we evaluate attributes rather than get to know someone, and the attributes on a profile are poor guides to whether we’ll click in person. Removing the friction doesn’t remove the need for slow, repeated time together to find out who someone really is.

What We Can Learn From Villages

In one sense the Sardinian finding is obvious. We evolved in small, densely connected groups where face-to-face contact was unavoidable and relationships lasted a lifetime, and the biology that makes physical presence protective was shaped over millions of years of that being the default. See Dunbar’s number, and so on. What has changed is not our need for this social connection, but how optional, inconvenient and structurally unlikely modern life has made it. Again – we’ve treated it as friction to be optimised away.

None of this is about romanticising village life or turning down the benefits of urban modernity. It’s about noticing that modern life has quietly stripped out conditions the body seems to treat as essential, with health costs that are real and measurable. For public health, it means that approaches focused almost entirely on diet, exercise and clinical risk factors, and largely ignoring the social side of health, are missing something big.

For organisations, the question is what the office used to do for people. The coffee chat, the corridor conversation and the shared lunch were quietly keeping face-to-face contact going. Where those have gone, the cost hasn’t disappeared. It has moved somewhere that doesn’t show up on a dashboard. That isn’t an argument for five days a week in the office, just for replacing what has been lost on purpose rather than hoping it looks after itself.

Rebuilding a Bit of the Village

Nobody is suggesting we all move to the mountains of Barbagia, but the village does offer a design lesson. People there didn’t have to decide each week whether to put the effort into holding their relationships together, because the structure of life and society made doing so the default. The effort of friendship was always there. The organising effort, and the self-regulating effort of choosing to put the effort in, wasn’t. Research suggests it takes around 50 hours together to go from acquaintance to casual friend, about 90 to become a friend, and over 200 to become a close friend. Nobody gets there through a single coffee. You’ve gotta put the effort in.

Take parkrun. It’s an amazing thing. Every Saturday morning thousands of people turn up to a free, volunteer-run 5k that they can run, jog, walk, marshal or volunteer at. It has the same time and place every week, familiar faces, a low barrier to entry and a role for everyone, including people who would rather not run. There’s still some friction, because you have to get out of bed on a cold Saturday, but the organising is taken care of and the commitment stays.

Meetup groups, choirs, walking groups and volunteer rotas work the same way. (As an aside, you might enjoy this episode of the BBC’s “Desert Island Discs” with Paul Sinton-Hewitt, founder of parkrun.)

Smaller ties count too. Gillian Sandstrom and Elizabeth Dunn found that students tended to feel happier and more connected on days when they interacted with more acquaintances. The benefit sits on the far side of the effort, so the aim is to make the effort easier to start without taking it away altogether. I also think that we can all learn to link the benefits of effort to the payoffs of connection (to change our predictions / predictive coding), but we’re not focusing on that here.

Is Anyone Prescribing This?

Is anyone in the health system doing anything about this? Ten years ago, barely. Today, a bit.

Since 2019 the NHS in England has funded social prescribing link workers in primary care. If a GP decides that someone’s real problem is loneliness rather than anything medical, they can refer them to a link worker, who works out what matters to that person and puts them in touch with local groups: walking clubs, choirs, gardening projects, befriending schemes, volunteering. There are around 3,300 link workers. A study in The Lancet Public Health estimated that about 1.3 million people were referred in 2023 alone, well above the NHS target of 900,000 by 2023/24. Similar models exist in Canada, Australia, the Netherlands and Singapore, among others.

Part of this is parkrun. Since 2018 the Royal College of GPs and parkrun UK have run the parkrun practice initiative, in which GP surgeries link up with their local parkrun and encourage patients and staff to run, jog, walk or volunteer. The 2,000th practice registered in March 2025. Volunteering is part of the pitch: the National Academy for Social Prescribing suggests that patients struggling with isolation or low mood could be encouraged to volunteer, because it gives them a role and a reason to turn up. So a doctor suggesting that someone who is feeling low or isolated goes and volunteers at parkrun is not a hypothetical. That is what the guidance describes.

There are two caveats. First, in practice it’s mostly encouragement and signposting rather than a formal clinical pathway, so “prescribing” is a metaphor. Second, the evidence is thinner than the enthusiasm. A 2022 systematic review in BMJ Open of link-worker social prescribing (eight controlled studies, about 6,500 people) found no consistent improvement in mental health, quality of life or social support, and concluded that the evidence was insufficient to judge effectiveness or cost-effectiveness. The reported benefits of parkrun practices rest mostly on perceptions, not measured outcomes. That doesn’t mean it doesn’t work: the studies are few and the schemes vary a lot. It does fit the theme, though. Pointing someone at a group removes the friction of finding it, but not the friction of turning up week after week until it becomes a habit.

Another aside – you might enjoy this BBC podcast episode: “What’s up docs? – Are friends good for your health?”.

Learning More

To go deeper, explore these resources on people-shift.com:

• Wellbeing
• Subjective Wellbeing
• Employee Engagement
• Bowling Alone
• Dunbar’s Number
• Miswanting
• You Should Talk to Strangers. Science Says So.
• Making Friends After 30 Is Harder Than It Should Be
• Loneliness Is a Health Crisis, Not a Feeling

People Shift View: We very much feel that social connection hugely matters and that it has been eroded in recent years. There’s evidence out there to support both of these claims. In fact, we think that this is one of the most important issues of the start of the 21st century, and something more people should focus on rectifying.

Things are moving and action is being taken. Our relationship with digital tech is changing. Potentially, how people choose to spend their time is changing. Social prescribing is a bit of a thing. This is good news, but there is more to do.

For us as individuals, the takeaway from all this is pretty simple. The hassle of building relationships, travelling to maintain them, reaching out, forgiving and investing in our connections with others is exactly what we are tempted to engineer out of our lives, and exactly what the evidence says we should protect. Pick one regular, in-person thing with the same people, and keep turning up.