A reduced capacity to want, pursue, or enjoy things has its own name in the clinical literature, its own measurement scales, and, in the research framework used by the National Institute of Mental Health, its own domain. The name is anhedonia. It is filed as a thing in itself rather than as the absence of good mood.

That distinction sits underneath a claim now circulating in shortened form: that people who cannot simply decide to feel better are not failing at effort or gratitude. The evidence supports a narrower version of the claim than the phrasing implies. Treatment helps most people, including with this symptom. What the research describes is a symptom that tends to improve more slowly and less completely than the sadness it accompanies, and that shows up on separate measurement instruments.

We are writers, not clinicians. What follows is a reading of the research, not medical advice.

Where the word came from

The French psychologist Théodule Ribot coined anhedonia in 1896, using it for a total loss of both physical and psychic pleasure in melancholia. It entered formal psychiatric diagnosis in the third edition of the Diagnostic and Statistical Manual, published in 1980, and survives in DSM-5. A major depressive episode there requires five or more symptoms, at least one of which must be either depressed mood or loss of interest and pleasure. Neither is individually mandatory. Either can serve as the gateway.

That structure is decades old, and it is the part most summaries drop.

Researchers have since broken the term apart. Kent Berridge and Terry Robinson’s work on reward established a separation between wanting and liking, between the pull toward something and the pleasure taken in it. Der-Avakian and Markou, reviewing that literature in Trends in Neurosciences in 2012, set out reward processing as motivation, consumption, and learning. Michael Treadway and David Zald, in a 2011 paper in Neuroscience & Biobehavioral Reviews, argued that psychiatry’s definition of anhedonia was too loose to be useful because it failed to separate consummatory from motivational deficits. Someone can lose one and keep the other, still enjoying a meal once it is in front of them while finding they cannot generate any pull toward arranging it.

Why it is measured separately from low mood

The separation began with questionnaires rather than scanners, and it is older than the neuroscience.

In 1988, David Watson, Lee Anna Clark, and Auke Tellegen published the Positive and Negative Affect Schedule in the Journal of Personality and Social Psychology: two ten-item mood scales. Across their development samples the two scales correlated between minus 0.12 and minus 0.23, which the authors called invariably low. In a psychiatric inpatient sample of 61 people the correlation reached minus 0.27. Low, then, but not zero. Reporting little positive feeling and reporting a great deal of negative feeling turn out to be largely, though not completely, separate things to report.

Those are facts about self-report items, not about brain systems, and the two get collapsed constantly in secondhand accounts.

The NIMH framework makes the separation structural. Its Research Domain Criteria list Positive Valence Systems and Negative Valence Systems as separate domains, with reward responsiveness, reward learning, and reward valuation as constructs beneath the first, and effort as a subconstruct beneath reward valuation. That framework is a hypothesis about how to carve the problem up, adopted by a funding agency because the older diagnostic categories were not producing results. It organizes research. It does not settle anything.

The clinical literature supports the bet without closing it. A 2025 review in Translational Psychiatry by Congchong Wu and colleagues at Zhejiang University School of Medicine gathers evidence that people with major depression and prominent anhedonia differ from those without it in symptom pattern, in reported cognitive difficulty, and in treatment response. The review repeats a background figure from a 2019 systematic review by Cao and colleagues in Progress in Neuro-Psychopharmacology and Biological Psychiatry, putting the share of people with major depressive disorder showing clinical features of anhedonia at roughly 70 percent. That figure is a summary statement rather than a derived prevalence, with no instrument or interval attached to it.

The effort study, and the numbers inside it

The best-known behavioral evidence comes from a task built to measure willingness to work for a reward rather than to ask people how much they enjoy things. In 2009, Treadway and colleagues published the Effort Expenditure for Rewards Task in PLOS ONE. Participants choose repeatedly between an easy button-pressing task for a small reward and a harder one for a larger reward, at varying odds of payout.

The sample was 61 people, 64 percent of them female, recruited through Vanderbilt University and the surrounding community and drawn from a pre-screened pool of 324 undergraduates. It was not a clinical sample.

The result is usually passed along as an inverse relationship between anhedonia and willingness to expend effort. The numbers are more mixed than that. On the Chapman anhedonia scales, the correlation with choosing the hard task reached minus 0.28 at the 50 percent probability level, and was not significant at either the 88 percent or the 12 percent levels. The Snaith-Hamilton Pleasure Scale, which other researchers treat as the field’s standard instrument, showed no significant correlation with hard-task choices at any probability level. The authors note that with 60 participants the study had 80 percent power to detect correlations of around 0.36, which is larger than what they found.

This is one study, not settled consensus. The task has been widely adopted since, and measuring choices is a real advance over measuring self-reported enjoyment. An effect of that size in 61 undergraduates is still a modest foundation for any claim about what a particular person can do.

Both therapies helped, and anhedonia still lagged

Nothing in this section is a reason to delay or stop treatment. In the largest relevant trial, both treatments improved both measures.

Alsayednasser and colleagues published a secondary analysis of the COBRA trial in Behaviour Research and Therapy in 2022. COBRA randomized 440 adults with current major depression to cognitive behavioral therapy or behavioral activation. Anhedonia improved in both arms across the six months of acute treatment, with no significant difference between them. The trial had no untreated comparison arm, so it can show that the two therapies performed similarly to each other; it cannot separate their effect from the passage of time.

What the analysis adds is a pattern in what was left over. Participants remained above healthy population averages for anhedonia at six months, with no further significant improvement at 12 or 18 months. Across both arms, anhedonia repaired less completely than depression did. Greater anhedonia at the start predicted worse depression outcomes and fewer depression-free days.

The medication literature points a similar way without being conclusive. The Cao review found the benefit of some selective serotonin reuptake inhibitors on hedonic impairment to be limited, with several other agents reported as more effective. Those conclusions are drawn across heterogeneous studies, and none of it is a reason for anyone to change or stop a medication.

A 2020 secondary analysis by Gabriela Khazanov and colleagues, also in Behaviour Research and Therapy, complicates the picture in a useful direction. Higher baseline distress and anhedonia predicted longer times to remission and recovery, which agrees with COBRA. But those same patients gained from having cognitive therapy added to antidepressant medication, while patients with lower distress and anhedonia showed no added benefit from the combination. On that reading, prominent anhedonia marks a need for more treatment rather than for less.

What none of this establishes

There is no scan and no blood test that identifies anhedonia. The Wu review collects imaging studies reporting group-level differences in reward-related regions between depressed people with and without prominent anhedonia, including striatal and caudate volume. Those are averages across modest samples, and a difference between group means does not diagnose an individual. Reward research also sits in an area with well-documented replication problems, and this literature is not exempt.

The instruments do not fully agree with one another either. A 2016 review by Rizvi and colleagues in Neuroscience & Biobehavioral Reviews describes the Snaith-Hamilton scale as the field’s standard, and it asks mainly about pleasure taken in the moment. The Temporal Experience of Pleasure Scale was built to pull anticipatory and consummatory pleasure apart. Studies using different scales are not always measuring the same construct, which is part of why the effort-task results split the way they did.

The willpower framing deserves care in both directions. Nothing here shows that effort is irrelevant. What it shows is that effort-based decision making is itself among the things being measured, which makes “just try harder” a description of the difficulty rather than an instruction that resolves it.

Anyone who recognizes a persistent version of this in themselves is better served by a conversation with a clinician than by a reading of the literature, including this one.

The open question is whether anhedonia is a severity marker for depression or a distinct condition that often travels with it. Settling that would take trials designed with anhedonia as the primary outcome rather than as a secondary measure, in samples large enough to separate the two. There are still few of those.

Featured image is representative and does not depict any person described in the research. Photo by Patricia Bozan via Pexels.

Editor’s note: This article was updated on September 14, 2026, to align its headline and wording more closely with the evidence and sources.