A life containing only positive feelings sounds like an easy choice until the choice is made concrete. In a 2026 experiment, most participants rejected it. They preferred an emotional life that still contained some sadness, worry, anger and other unpleasant states, even when an almost entirely pleasant alternative was available.
The same paper found something quieter in a second sample. People reported the highest life satisfaction when the emotions they experienced were close to the emotional mix they wanted. Those with the most positive feelings did not automatically report the greatest satisfaction.
The study, led by Jinrui Liu and published in the Journal of Happiness Studies, offers an interesting correction to the idea that happiness is simply a score to maximise. It does not show that negative feelings make people happier, and it does not supply an ideal emotional formula. This is one paper, not settled consensus.
The experiment offered an emotionally perfect life
The first study recruited 352 adults in the United States through Prolific. Participants were aged 19 to 76. The researchers asked them to choose between lives containing different proportions of positive and negative emotional moments.
One option held positive emotion at either 100 or 99 per cent, or negative emotion at zero or one per cent. The alternative began slightly farther from that theoretical maximum and changed in steps as participants continued choosing. This forced people to do more than agree vaguely that sadness has a place in life. They had to decide how close to emotional perfection they actually wanted to go.
Most did not choose the maximum. Across the study’s different measures, participants tended to describe an ideal life containing roughly 66 to 80 per cent positive affect and 10 to 13 per cent negative affect. “Affect” here simply means felt emotional experience, not a diagnosis or a judgement about personality.
Those percentages belong to this experiment. They are not targets, recommendations or universal thresholds. The forced-choice design may also have encouraged some people to avoid an extreme answer simply because it looked unrealistic. The authors recognised that possibility, although participants’ written explanations suggested that response style was not the whole explanation.
The second study compared ideals with experience
The researchers then combined the first sample with a separate international, English-speaking sample, producing a total of 638 participants. They compared the emotions people said they wanted with the emotions they said they experienced, then examined how that gap related to life satisfaction.
Positive emotion still mattered. Across the analysis, more experienced positive affect was associated with higher life satisfaction. More negative affect than a person wanted was associated with lower satisfaction.
But the pattern did not continue without limit. Participants whose emotional experience matched their ideal reported the highest life satisfaction. Once people were already experiencing less negative emotion than they said they wanted, reducing it further was no longer clearly linked with greater satisfaction.
The design was cross-sectional, meaning the measurements were taken at one period rather than following people through a change. It cannot establish whether emotional alignment raises life satisfaction, whether satisfied people judge their emotions as more aligned, or whether another factor influences both.
This is about fit, not a defence of suffering
The result is easy to misread. It does not say that distress is desirable, that people should seek painful experiences, or that too much happiness is a problem. Participants strongly preferred a life dominated by positive rather than negative feelings.
The narrower finding is that maximum pleasantness was not the only thing they valued. Many appeared to want emotions that fitted events, relationships and personal standards. Sadness after a loss, concern about a real problem or anger at unfair treatment may feel unpleasant while still seeming appropriate to the person experiencing it.
That interpretation remains an interpretation. The study measured preferences and self-reports, not the function of each emotion in a person’s life. It also grouped specific feelings into broad positive and negative categories, which can hide meaningful differences. Anger, fear and sadness are all unpleasant, but they do not feel interchangeable and do not arise in the same situations.
An older international study found a similar pattern
The idea that emotional fit matters did not begin with the 2026 paper. In 2017, Maya Tamir and colleagues studied 2,324 university students across eight countries: the United States, Brazil, China, Germany, Ghana, Israel, Poland and Singapore.
Participants reported the emotions they wanted, the emotions they experienced and several measures of wellbeing. Across the samples, people who more often experienced the emotions they wanted also reported greater wellbeing. That association appeared even among participants who wanted less pleasant or more unpleasant emotion than they currently experienced.
The 2017 study had limits of its own. Its participants were university students with an average age a little above 22, so it did not represent entire national populations. Like the alignment analysis in the newer paper, it relied on self-report and could not settle the direction of cause.
Still, two independent research teams arrived at a related distinction. Feeling good and feeling emotionally right overlap, but they may not be identical.
People do not define happiness in one way
The word itself creates another problem. Researchers use “happiness” for immediate positive emotion, an overall judgement of life satisfaction, or some combination of pleasure, meaning and functioning. Ordinary language is just as varied.
A 2016 study of 2,799 adults in 12 countries asked people to define happiness in their own words. Inner harmony was the most common psychological definition, while family and social relationships were prominent in descriptions of life circumstances. The answers varied, but they often reached beyond the idea of feeling intensely good.
I explored a neighbouring distinction in an earlier article about pleasure and meaning. The new study asks a different question inside the emotional side of that divide: even if pleasant feeling is the goal, do people truly want as much of it as possible?
For most participants, the answer was no.
A study of happiness can only measure the question it asks
Happiness research often produces contradictions because one paper measures yesterday’s mood, another asks for a judgement of life as a whole, and a third measures whether experience feels meaningful. Those are related questions, but they are not substitutes for one another.
The 2026 study is strongest when read at its actual scale. In one American experiment, most people preferred a mixed emotional life to a perfectly positive one. In a combined sample, the match between desired and experienced emotion was associated with life satisfaction. Neither result tells every person which feelings belong in a good life.
What it does show is that many people do not imagine happiness as uninterrupted pleasure when given a careful choice. They leave room for emotional experiences that fit the life being lived, including some they would never describe as pleasant.