A span of 20 years is short in the life of a science and long in the life of a field still working out what it is for. In 2005, the Gallup World Poll began asking ordinary people in dozens of countries a deceptively simple question — how is your life going? — and the answers have since accumulated into one of the largest records of human wellbeing ever assembled. Over the same period, laboratories, longitudinal surveys, smartphone apps, and a handful of unusually patient research teams have produced a steady stream of findings about what raises, lowers, and sustains the experience of a good life. Some of those findings have since been overturned or quietly shelved. Others have survived replication, refinement, and argument, and now form the working foundation of the field.
This essay selects ten of the survivors. The choice is a judgement, and a defensible list could differ — but each finding below was chosen for the same three reasons: it has been tested more than once, it has changed how researchers frame their questions, and it carries a practical consequence for how a person, a university, or a society might organise its efforts. Read together, they sketch a portrait of happiness that is considerably less glamorous than the popular version and considerably more useful.
If the field had to keep a single result, this would be it. The Harvard Study of Adult Development began in 1938 with a group of Harvard sophomores and, a little later, a second group of boys from some of Boston’s poorest neighbourhoods, and it has followed them — and now their children — for more than 80 years. George Vaillant directed the study for decades; Robert Waldinger, a psychiatrist and Zen priest, has directed it since 2005, alongside Marc Schulz. The conclusion Waldinger has returned to again and again is plain: the quality of a person’s close relationships in midlife predicts health and wellbeing in later life more reliably than wealth, fame, or even cholesterol levels. Participants who were most satisfied in their relationships at 50 were, on average, the healthiest at 80.
What makes the finding persuasive is not its sentiment but its source. The study’s data were gathered prospectively — that is, people were measured as their lives unfolded, not asked to reconstruct them afterwards — which spares the result from the distortions of hindsight. It also resists the reading that happy people simply find it easier to make friends; Vaillant’s team tracked how relationships shaped later outcomes across decades of divorce, illness, success, and failure. The practical lesson, as Waldinger has put it, is that relationships are not a pleasant supplement to a good life but one of its load-bearing walls.
A few years into this twenty-year period, Julianne Holt-Lunstad, Timothy Smith, and Bradley Layton pooled the results of 148 studies, covering 308,849 participants followed for an average of seven and a half years, and asked whether the strength of a person’s social relationships was associated with their likelihood of survival. It was — by about 50 per cent. The authors compared the effect with other established risk factors and found that the influence of social relationships on mortality was comparable to that of quitting smoking and exceeded that of obesity and physical inactivity. The paper, published in PLoS Medicine in 2010, has been among the most cited in the social-health literature ever since.
It is important to be exact about what this does and does not say. The result is an association across many observational studies, and it cannot by itself prove that loneliness causes early death — though later work on biological pathways, from stress hormones to inflammation, has strengthened the plausibility of a causal story. What it does establish is that social connection deserves to be treated as a public-health variable, in the same breath as diet and exercise. The Indian philosophical tradition, which has long held that a person is constituted by relation rather than standing apart from it, would find the conclusion unsurprising; the phrase Vasudhaiva Kutumbakam — the world is one family — is a moral claim in the Upanishads, but the evidence suggests it is also a statement about physiology.
For much of the late twentieth century, the dominant picture of happiness was the hedonic treadmill: the idea that people return swiftly to a stable baseline after good or bad events, so that neither a lottery win nor a serious injury leaves a lasting mark. The most influential early evidence came from a small 1978 study by Philip Brickman and colleagues comparing lottery winners and accident victims. In the period this essay covers, that picture was substantially revised. Ed Diener, Richard Lucas, and Christie Napa Scollon argued in 2006 that adaptation is real but partial, varies by event, and differs sharply between people. Using large, nationally representative panel data — notably the German Socio-Economic Panel — Lucas and colleagues showed that some experiences leave lasting changes in wellbeing: long-term unemployment, the death of a spouse, and the onset of a serious disability all produced shifts that did not fully recover on average, even years later.
The revision matters for two reasons. It corrects a fatalistic reading of the treadmill — the notion that nothing a person or a society does can durably raise wellbeing — and it directs attention to the circumstances that do not fade, such as unemployment and chronic illness, which are therefore worth preventing rather than merely enduring. The Yoga Sutras’ concept of santosha, contentment, was never a counsel of passivity; it describes a trained steadiness of mind that operates alongside effort rather than instead of it. The modern evidence arrives at a similar balance: adaptation is a resource, not a verdict.
No question in the economics of happiness has generated more heat. In 2010, Daniel Kahneman and Angus Deaton reported that emotional wellbeing in a large American sample rose with income up to about $75,000 and then levelled off, while life evaluation kept rising. The finding was widely taken to mean that money stops buying happiness at a modest threshold. In 2021, Matthew Killingsworth used real-time experience-sampling data collected through a smartphone application and found no such plateau: experienced wellbeing continued to rise with income for most people. The two camps agreed to settle the dispute, and in 2023 Killingsworth, Kahneman, and Barbara Mellers published a joint adversarial collaboration in PNAS. Their reconciliation was subtle: for roughly the least happy fifth of people, happiness rose with income up to about $100,000 and then stopped; for everyone else, it kept rising.
The lesson is twofold. First, the older slogan that money cannot buy happiness is too crude to survive the data; income is associated with greater wellbeing across most of its range. Second, the association is not uniform, and for people already carrying heavy unhappiness, additional income past a certain point does not address what is actually wrong — an observation that echoes Richard Easterlin’s 1974 insight that wealth is an incomplete explanation for wellbeing. The field’s willingness to revise its own best-known result in public, using better instruments and a structured collaboration between rivals, is itself one of the more reassuring developments of the past two decades.
In 2008, Elizabeth Dunn, Lara Aknin, and Michael Norton published a short paper in Science with a long reach. In a national survey, in a study of employees receiving profit-sharing bonuses, and in an experiment in which participants were handed either $5 or $20 and told to spend it either on themselves or on someone else, the same pattern appeared: spending on others made people happier than spending on themselves. It did not matter much how much money was involved. What mattered was the direction of the spending.
Aknin and colleagues later tested whether this was a peculiarity of wealthy Western populations. Using Gallup World Poll data from 136 countries, they found that prosocial spending was associated with greater happiness in the large majority of them — a result the authors described as a psychological universal. Aknin, Hamlin, and Dunn also found that toddlers in their second year of life showed more happiness when giving treats away than when receiving them, which suggests the effect is not a product of cultural instruction alone. The finding is the empirical cousin of seva, the Indian ethic of selfless service, and of the Bhagavad Gita’s teaching that action performed without grasping at its fruits is itself a source of steadiness. A person seeking to feel better, it appears, is often better advised to attend to someone else.
In 2010, Matthew Killingsworth and Daniel Gilbert at Harvard built a smartphone application that contacted around 2,250 adults at random moments in their day and asked what they were doing, what they were thinking about, and how they felt. The results, published in Science under the title “A Wandering Mind Is an Unhappy Mind,” were striking. People’s minds were wandering away from what they were doing nearly 47 per cent of the time, and they were less happy when they were wandering than when they were not — including when the activity at hand was unpleasant, and including when the thoughts they wandered towards were pleasant. The timing analysis suggested that mind-wandering preceded unhappiness rather than following it.
The study was important in part for its method, which opened a new way of measuring happiness moment by moment, and in part for its implication. It gave empirical weight to an observation that contemplative traditions have made for millennia. The Buddhist practice of sati, usually translated as mindfulness, and the Yoga Sutras’ account of dharana and dhyana — concentration and sustained meditative attention — rest on the premise that where attention goes, experience follows. Modern research has not yet resolved how much training can change a person’s habits of attention, but it has moved the question from the margins of psychology to its centre.
A recurring criticism of positive psychology is that its exercises — gratitude letters, acts of kindness, the identification of personal strengths — amount to self-help dressed in a laboratory coat. Nancy Sin and Sonja Lyubomirsky addressed the criticism directly in a 2009 meta-analysis of 51 studies involving 4,266 participants. Positive psychology interventions, they found, produced significant improvements in wellbeing (a correlation of about .29) and reduced depressive symptoms (about .31). The effects were moderate, not miraculous — and they were larger for people who were motivated and who chose activities that suited them.
The result has been refined since, and some later meta-analyses have suggested the effects may be smaller once publication bias is taken into account. A fair summary is that deliberate practices help, that they help more when they fit the person, and that their effect is real but not enormous. This is consistent with the traditional Indian view of abhyasa, steady practice, as the means by which a quality of mind is cultivated: no single exercise transforms a life, but sustained practice shifts the ground on which the next day is lived. The Rekhi Foundation for Happiness builds its Science of Happiness course on this premise, treating wellbeing as a set of teachable skills rather than a fixed temperament, and teaching those skills within a credit-bearing academic framework at more than 90 institutions across six countries.
Patrick Hill and Nicholas Turiano used data from the Midlife in the United States study, following more than 6,000 adults for roughly 14 years, to ask whether a sense of purpose predicted survival. Their 2014 paper in Psychological Science found that it did: people with a greater sense of purpose in life lived longer, and the protective association held across the adult lifespan, from the young to the old, and after accounting for other aspects of positive emotion and for factors such as age, gender, and emotional distress. Earlier work by Patricia Boyle and colleagues at Rush University had linked a stronger sense of purpose among older adults to a lower risk of Alzheimer’s disease and cognitive decline.
The finding fits the theoretical lineage of Viktor Frankl, who argued from his experience in the concentration camps that the search for meaning is the primary human motivation, and of Carol Ryff, whose model of psychological wellbeing treats purpose as a core dimension. It also finds a precise echo in two cultural traditions. The Japanese concept of ikigai — a reason for getting up in the morning — and the Indian concept of svadharma, one’s own rightful duty and path, both locate wellbeing in alignment between a person and a purpose larger than momentary comfort. Purpose, the research suggests, is not a luxury of the already contented; it is a resource that appears to protect the body as well as the spirit.
Related to purpose, but broader, is the evidence that positive wellbeing is associated with physical health. Yoichi Chida and Andrew Steptoe’s 2008 meta-analysis of prospective studies in Psychosomatic Medicine found that positive psychological wellbeing was associated with reduced mortality in initially healthy populations, with an effect of roughly 18 per cent lower risk. Ed Diener and Micaela Chan reviewed the broader literature in 2011 and concluded, in a paper titled “Happy People Live Longer,” that subjective wellbeing is associated with health and longevity across multiple kinds of evidence, including long-term studies, experimental work, and research on the biological processes involved.
The qualification, as always, is the difference between association and cause. People who are healthy tend to feel better, and some of the correlation certainly runs in that direction. But the prospective designs of the best studies — which measure wellbeing in people who are healthy at the start and follow them forward — make a purely reverse reading difficult to sustain. The finding gives weight to an idea that wellbeing education is more than a courtesy: if emotional wellbeing is associated with health outcomes at the population level, then teaching it to young adults at scale belongs on the same list of institutional responsibilities as nutrition and physical education.
The tenth finding is one that cannot be reduced to individual psychology. Since the first World Happiness Report appeared in 2012, edited by John Helliwell, Richard Layard, and Jeffrey Sachs, the annual rankings have documented large and persistent differences in wellbeing between nations — differences not well explained by national income alone. The factors the report examines include social support, healthy life expectancy, freedom to make life choices, generosity, and perceptions of corruption, alongside income. Finland has topped the ranking for eight consecutive years through the 2025 edition, a record that researchers have attributed in significant part to high levels of social trust and strong institutions rather than to wealth.
The implication is that happiness is partly a public good. A person’s wellbeing depends on their own choices and relationships, but also on whether the people around them can be trusted, whether institutions treat them fairly, and whether they have someone to call in a crisis. This is the point at which several of the preceding findings converge: relationships, generosity, purpose, and trust are not four separate topics but aspects of one social fabric. The Indian ideal of Vasudhaiva Kutumbakam names that fabric; the World Happiness Report measures it.
Read in sequence, the findings share a quiet pattern. The strongest influences on a good life, according to the evidence, are other people, a sense of purpose, a trained capacity for attention, and a willingness to give — none of which can be bought, and all of which can be practised. Money and circumstance matter, but unevenly and with diminishing returns for those who are already unhappy. And the traditions that human beings have developed over millennia for thinking about flourishing, from the Upanishads to Aristotle’s eudaimonia to Japanese ikigai, turn out to have anticipated much of what the instruments of the past twenty years have since measured.
The 20-year record also carries a sober note. Several once-celebrated claims in psychology have not survived replication, and the ten findings above are best understood as well-supported rather than final; effect sizes shrink under scrutiny, and the field’s most careful practitioners say so. But a science that corrects itself — as the income debate shows it can — is a science that is working.
Two decades of data have not produced a formula for happiness. They have produced something more modest and more durable: a map of where it tends to be found — in people, in purpose, in attention, in generosity, and in the quality of the places we share.
If one finding had to be chosen, most researchers would point to the primacy of close relationships. The Harvard Study of Adult Development, directed since 2005 by Robert Waldinger, has followed participants for more than 80 years and found that the quality of relationships in midlife predicts health and wellbeing in later life more reliably than wealth or social standing. Julianne Holt-Lunstad's meta-analysis of 148 studies reinforced this, finding that stronger social relationships were associated with roughly a 50 per cent greater likelihood of survival. The result matters because it is consistent across methods: prospective cohort studies, large-scale meta-analyses, and cross-cultural surveys all point the same way. It also has an unusually practical consequence, since relationships are among the few determinants of wellbeing that a person can deliberately invest in at any age. The Rekhi Foundation for Happiness reflects this in its Science of Happiness course, which treats connection, gratitude practice, and emotional wellbeing as skills that can be taught and strengthened, not temperaments a person either has or lacks.
The answer from the past two decades is more nuanced than either "yes" or "no." In 2010, Daniel Kahneman and Angus Deaton reported that daily emotional wellbeing stopped rising with income at about $75,000, though life evaluation continued to rise. In 2021, Matthew Killingsworth, using real-time data, found that experienced wellbeing kept rising with income for most people. A 2023 collaboration between Killingsworth, Kahneman, and Barbara Mellers resolved the conflict: for roughly the unhappiest fifth of people, happiness stopped rising with income at around $100,000, while for everyone else it continued to climb. In practice, this means money is associated with greater wellbeing across most of its range, but it does not solve the problems of people who are unhappy for reasons money does not address. It also sits alongside Richard Easterlin's older observation that national wealth alone is an incomplete explanation for happiness, which remains a central theme in happiness research and in emotional wellbeing education.
The evidence suggests it can, within limits. Nancy Sin and Sonja Lyubomirsky's 2009 meta-analysis of 51 studies found that positive psychology interventions, such as practising gratitude or performing acts of kindness, produced moderate improvements in wellbeing and reductions in depressive symptoms. Effects were larger for people who chose practices that suited them. Later reviews have suggested that the average effect may be somewhat smaller once publication bias is considered, so a fair summary is that deliberate practice helps, but modestly. Research on adaptation by Ed Diener and Richard Lucas also indicates that wellbeing is not fixed: some life events produce lasting changes. For students and institutions, this supports the case for formal teaching of wellbeing skills. A credit-bearing Science of Happiness course, of the kind the Rekhi Foundation delivers across more than 90 institutions, applies this logic by treating happiness research as an academic subject with practical exercises, not simply as advice.
The World Happiness Report, first published in 2012 and edited by John Helliwell, Richard Layard, and Jeffrey Sachs, ranks countries using the Cantril Ladder, in which respondents rate their own lives on a scale from zero to ten. Researchers then analyse how much of the variation can be explained by factors including income per person, social support, healthy life expectancy, freedom to make life choices, generosity, and perceptions of corruption. Finland has topped the ranking for eight consecutive years through the 2025 edition, and researchers attribute much of this to high social trust, strong institutions, and low inequality. The broader finding is that happiness is partly a property of places: it depends on whether people can rely on one another and on their institutions. This is one reason human flourishing is increasingly treated as a policy question and not merely a private matter, and why the Indian ideal of Vasudhaiva Kutumbakam, the world as one family, has found unexpected empirical support.
It is a fair question, and the field has taken it seriously. Several once-prominent findings in social psychology have not replicated, and happiness research has not been immune: effect sizes in intervention studies tend to shrink under scrutiny, and some earlier claims, such as strong fixed "set point" estimates, have been revised. The findings highlighted in this essay have been selected partly because they have held up across multiple methods and samples. The Harvard Study of Adult Development is prospective and long-running; the Holt-Lunstad analysis pooled 148 studies; Dunn, Aknin, and Norton's finding on giving was extended by Aknin and colleagues to 136 countries; and the income debate was resolved through a deliberate adversarial collaboration between researchers who had disagreed. A reasonable approach is to treat happiness research as a self-correcting body of work, and to treat any single striking study with caution until it has been replicated. Resources such as the World Happiness Report and the Harvard Study offer a transparent starting point for readers who wish to look at the evidence directly.
The Rekhi Foundation for Happiness is committed to advancing the science of happiness through education, research, and community engagement. Learn more at rekhifoundation.com
Rekhi Foundation, founded in 2016, promotes Happiness Science via university centers, collaborating globally across six countries.
Table of Contents
6/79, S/F, Gurudwara Road, Karol Bagh, New Delhi – 110005, INDIA
2051, Last Chance Court, Gold River, CA 95670, USA
