Trying to feel happier can become a task of its own. A person has to notice the present mood, compare it with the mood they want, choose something that might change it and keep checking whether the effort worked. The pursuit may feel emotional, but it also involves attention, choice and control.
Aekyoung Kim and Sam Maglio asked whether that work leaves people less able to regulate the behavior that comes next. Their 2025 paper, Happiness depletes me: Seeking happiness impairs limited resources and self-regulation, reported four studies in Applied Psychology: Health and Well-Being. Maglio is affiliated with the University of Toronto Scarborough and its Rotman School of Management; Kim was at Jeonbuk National University when the article appeared.
ScienceBlog recently examined the paper’s four-study design and its immediate limitations. This follow-up takes the narrower claim in the new title seriously: what does it mean to say happiness pursuit consumes the “same mental resources” as self-control, and did the researchers actually show that everyday habits suffer?
This is one paper, not settled consensus. Its short experiments support a claim about subsequent behavior under controlled conditions. They did not follow sleep, exercise, study routines, saving, household work or relationships across months. The idea that happiness maximization could undermine the habits that sustain a good life is a hypothesis built from the findings, not an outcome the studies directly observed.
Four studies built a chain from association to experiment
Study 1 asked people about their tendency to seek happiness and about failures of self-regulation in daily life. Stronger happiness seeking was associated with more reported lapses. That establishes a relationship between answers, but it cannot establish direction.
Someone who already finds restraint or persistence difficult might become more concerned with improving how they feel. Stress, personality, current mood or another unmeasured factor could influence both answers. Self-report also asks participants to summarize complicated patterns they may not remember precisely.
Study 2 added a behavioral outcome. Participants ranked lists of objects, and those who scored higher on happiness seeking spent less time on the task. The result moved beyond a second questionnaire, although the happiness measure was still observed rather than manipulated.
The final two studies changed the sequence. Researchers prompted one group to pursue happiness and then watched what happened on a later task. This gives the paper a stronger basis for saying that happiness-focused effort came before a difference in behavior.
The chocolate result depends on what counts as temptation
In Study 3, 36 participants saw advertising that either emphasized happiness or served as a comparison. They were then given a large bowl of chocolates, told they could eat as many as they wanted and asked to rate the taste. The happiness-primed group ate more.
The University of Toronto’s account of the paper describes this as greater susceptibility to temptation after trying to become happier. That interpretation is understandable, but it contains an assumption worth making visible.
Eating fewer chocolates counts as successful self-control only when restraint is the person’s active goal. The experiment did not establish that every participant wanted to eat less or that eating another piece represented a failure in their own life. It measured a short-term response to available candy under a standardized condition.
The sample was also small. Thirty-six people can produce a useful experimental clue, particularly when the result sits inside a wider set of studies, but it is not a stable foundation for claims about how happiness seekers generally eat, spend or handle temptation. Replication with larger and more varied samples matters.
The persistence experiment did not measure a daily habit
Study 4 used 188 participants and a different design. People made repeated choices between ordinary products. One group was instructed to choose the option that would improve happiness, while a comparison group focused on making accurate choices. Participants then moved to a challenging mental task.
The happiness-focused group stopped sooner. Because the instruction preceded the persistence measure, this is the paper’s clearest evidence that deliberately maximizing happiness can carry over into what someone does next.
It is still a measure of persistence on one artificial task. Stopping earlier does not tell us whether someone abandons a savings plan, misses a workout, leaves dishes in the sink or gives up on a difficult conversation. It also does not show that quitting was irrational; the happiness frame may simply have made a tedious follow-up seem less worth the effort.
Habits introduce another difference. A well-established routine can reduce the need for a fresh decision each time a cue appears. The 2025 studies did not measure automaticity, track repeated behavior or test whether happiness pursuit eroded a routine that already existed. “Less persistence now” and “weaker habits over time” are related ideas, not interchangeable results.
The shared mental resource was inferred, not measured
Kim and Maglio interpret the pattern through limited regulatory resources. On this account, managing emotion and exercising self-control rely on some of the same finite capacity. Effort spent trying to feel happier leaves less available for restraint or persistence a few minutes later.
The studies did not place a meter on that capacity. They recorded survey answers, time spent ranking objects, chocolate consumption and persistence after a choice exercise. They did not measure a biological fuel, a neural reserve or another direct marker that fell between the happiness manipulation and the later behavior.
This does not make “resource” meaningless. Psychology often uses latent concepts to explain patterns that cannot be observed directly. It does mean that depletion is the authors’ model of the sequence, not a separate empirical finding alongside the sequence.
The distinction matters most in the final step of the title. A shared resource could connect happiness monitoring with everyday self-control. But the paper did not show that the resource remained reduced, that repeated pursuit created cumulative depletion or that beneficial habits deteriorated as a result.
The simple depletion model has a difficult replication history
The wider idea is usually called ego depletion: exerting control on one task is said to reduce performance on a second, unrelated task. The model became influential because it offered a single explanation for why restraint can feel harder after effort.
Its empirical history is disputed. A 2015 series of bias-sensitive meta-analytic tests found very little evidence that the commonly used laboratory depletion effect reflected a real limited resource. The authors argued that publication and small-study biases could explain much of the apparent pattern.
A 2016 preregistered replication across 23 laboratories tested 2,141 people using a standardized protocol. Its estimated effect was very small, and the 95 percent confidence interval included zero.
A second major project published in 2021 used 36 laboratories and 3,531 participants. The confirmatory analysis again found no statistically reliable depletion effect. An exploratory analysis that ignored planned exclusions found a small effect, which is a reminder that design and analysis choices can change the answer.
Those results do not prove that prior effort never changes later self-control. Some studies find small effects under particular conditions, and researchers continue to disagree about the best tasks and necessary intensity. They do make the metaphor of a simple, universal willpower tank difficult to treat as established fact.
Attention and motivation can produce the same outward behavior
An alternative process account asks what people attend to and want after an effortful task. Following work, they may shift attention towards reward, become less interested in another demand or decide that conserving effort is reasonable. Performance falls, but no consumable mental substance has to be empty.
Michael Inzlicht and Brandon Schmeichel outlined this attention-and-motivation model of depletion more than a decade ago. It replaces one broad resource with a set of changes that can, in principle, be measured separately.
The happiness studies fit more than one possibility. Choosing repeatedly for maximum happiness may require effort. It may also heighten the appeal of immediate pleasure, make a dull task feel inconsistent with the active goal or shift the standard used to decide whether persistence is worthwhile.
These alternatives do not erase the observed chocolate or persistence differences. They change what those differences mean. If motivation explains the result, “unable to persist” may be too strong. The participants may have been less willing to persist under that particular framing, which is behaviorally similar but theoretically different.
The target is constant maximization, not happiness itself
The paper does not show that feeling happy drains self-control. Its target is active happiness seeking, defined as trying to experience as much positive emotion and as little negative emotion as possible. Enjoying something and managing enjoyment as a maximization project are not the same activity.
A person can spend time with a friend, cook, exercise, read or finish a demanding piece of work because the activity is valued. Happiness may follow without being scored after every step. The proposed cost appears when emotion becomes a target that must be produced, monitored and increased.
This fits a distinction ScienceBlog has covered before. A separate line of research found downsides tied more closely to evaluating happiness than to valuing it. Wanting a good life is broad; repeatedly asking whether one is happy enough can keep drawing attention to the gap between the current state and the desired one.
Nor does the study establish that deliberate mood change is always counterproductive. Context matters. Seeking relief from a bad afternoon, changing an unsafe situation and treating every ordinary moment as a failed happiness target are very different acts. The experiments isolate a short bout of maximization; they do not rank every way people try to improve their lives.
Showing damaged habits would require a longer bridge
The title’s final claim is best understood as a mechanism the authors propose. If happiness management repeatedly reduces subsequent regulation, and if useful behaviors repeatedly require that regulation, then habits might be harder to establish or maintain. Each link is plausible. The four studies did not observe the complete chain.
A stronger test would follow actual behavior across days or months. Researchers could measure happiness monitoring, established routines and later lapses while accounting for current mood, stress, prior self-control and the stability of each habit. Passive sensing, diaries or objective records could reduce reliance on memory.
Preregistered experiments could also compare mechanisms. If changing incentives restores persistence, motivation becomes more plausible. If attention shifts towards rewards, that supports a process account. If the effect remains after those routes are tested, and a specific shared capacity can be measured, the resource interpretation becomes stronger.
The chocolate result especially deserves a larger replication with a clear personal restraint goal. The persistence result would benefit from varied tasks and cultures. Both would be more informative if the research tested whether the same people later failed at behaviors they themselves considered important.
A careful reading preserves the paradox without turning it into advice
The paper provides evidence for a short sequence. People prompted to pursue happiness ate more chocolate in one small study and stopped a difficult task sooner in a larger one. Earlier correlational studies pointed in the same direction.
It does not establish that happiness is costly, that every deliberate effort to improve a mood backfires or that long-term habits inevitably weaken. It also does not justify ignoring persistent unhappiness or circumstances that need to change.
“Chase happiness too deliberately” is the defensible part of the warning when it refers to effortful, repeated maximization. “You may undermine the habits that sustain it” remains a possible consequence that needs direct longitudinal evidence.
The most useful question is therefore not whether happiness should be pursued. It is whether turning happiness into a constantly audited performance target makes attention, restraint and persistence harder to spend on the life from which happiness might otherwise emerge.