The advice on how to become happier is endless, confident, and mostly built on the same assumption: that your happiness is a project, largely under your control, waiting for the right routine of gratitude lists and cold showers. The research is quieter and less flattering to that idea. It suggests happiness is harder to move than the self-help shelf implies, and that the things which do move it are not the ones being sold.
This is a reading of the evidence, not clinical or psychological advice. But it is worth knowing what the science actually supports before reorganising your mornings around it.
The part of happiness that barely budges
Start with an uncomfortable finding. In 1996, the behavioural geneticists David Lykken and Auke Tellegen studied several thousand middle-aged twins and reported that somewhere between 44 and 52 per cent of the variation in wellbeing was associated with genetic differences. More striking, income, education, marital status and religious commitment each accounted for less than three per cent. Lykken put it bluntly at the time: trying to be happier, he suggested, might be as futile as trying to be taller.
He later softened that line, and it is worth softening. Heritability describes variation across a population, not a fixed ceiling for any individual, and it says nothing about what happens when circumstances change for everyone at once.
But the underlying pattern is real and well replicated. People carry a broadly stable disposition, a personal baseline they tend to return to. Win the lottery or suffer a serious setback, and after the initial swing most people drift back towards where they started. Psychologists call this hedonic adaptation, and it is the reason the new job, the new city, and the new phone all stop delivering their lift sooner than expected.
The famous pie chart that does not hold up
If you have read anything about happiness, you have probably met the figure: 50 per cent genes, 10 per cent circumstances, and a hopeful 40 per cent left over for intentional activity. It is a tidy, motivating number, and it launched a great deal of positive psychology.
It also does not survive scrutiny. In a 2020 paper in the Journal of Happiness Studies, Nicholas Brown and Julia Rohrer re-examined the model and found little empirical basis for slicing happiness into those neat proportions. The original figures came from mixing different kinds of studies in ways that do not really licence the conclusion, and, crucially, a percentage of variance is not the same as a lever you can pull. The 40 per cent was never a measured share of achievable change. It was closer to a leftover.
None of this means effort is pointless.
It means the confident arithmetic behind so much happiness advice was never as solid as it sounded.
What actually has some support, and how modest it is
So what survives? A smaller, less glamorous set of findings, most of which point away from self-optimisation and towards other people.
One of the more durable results is that spending on others tends to feel better than spending on yourself. In a 2008 study in Science, Elizabeth Dunn, Lara Aknin and Michael Norton gave people a small sum and randomly assigned them to spend it on themselves or on someone else. Those who gave it away reported more happiness that evening. The original experiment was small, with 46 participants, and later replication efforts found the effect real but more modest than the headline suggested. That is roughly the shape of this whole field: genuine effects, smaller than advertised.
The broader theme is social. The behaviours that correlate most reliably with wellbeing tend to involve connection, attention to others, and being useful to someone, rather than any private routine performed alone in front of a mirror. Money helps too, though less than people expect once basic needs are met, and Silicon Canals has covered the careful recent work on income and happiness elsewhere.
What this does not mean
It would be easy to read all this as fatalism, and that would be the wrong lesson.
A stable baseline is not a locked one. Adaptation is a tendency, not a law, and some changes, particularly in health, relationships, and chronic circumstances like a bad commute or a corrosive job, seem to resist the drift back and matter more than the set-point story first implied. The evidence argues against quick fixes and against grand claims, not against the idea that a life can get better.
The honest version is less marketable than a morning routine. Much of your happiness is dispositional and will not be reasoned away. The formulas promising precise control oversold it. And the interventions with the best evidence are quiet and outward-facing: give something away, pay attention to someone, be of use. They will not transform you, because little does. They are simply the parts of the advice that survived being checked.
