The claim resurfaces every year, always phrased the same way: at 33, you stop listening to new music. It is clean, it is specific, and it sticks. There is just one problem. That number appears nowhere in the analysis everyone credits it to.
We went back to the source. It is real, it is competent, and it says something quite different from what it has been made to say.
Where the number actually came from
The origin is a 2015 post by Ajay Kalia, then at Spotify, titled "Music was better back then." Kalia took the 2014 listening histories of U.S. Spotify users and paired them with artist popularity rankings from The Echo Nest. For each listener he computed the median popularity rank of the artists they streamed, then took the median across all listeners of a given age.
His conclusion, in his own words: the share of very popular music drops steadily through people's twenties, and "by their mid-30s, their tastes have matured." He writes about the early thirties and the mid-thirties.
He never writes 33. The number does not appear once in his text. It was read off a chart and rounded into a headline by the music press. NME reported in 2015 that "people stop listening to new music at 33," and called the analysis a study. It was not a study in the academic sense: no peer review, no journal. It was an engineer's blog post — transparent about its method, but a blog post.
So two things slipped in a single sentence. An approximation became a precise figure. And "popular music" became "new music." The second slip is the damaging one.
What the curve is really measuring
Look again at what Kalia measured: the average popularity of the artists you stream. Not their newness. Not how many of them there are. Not your curiosity.
A 35-year-old who discovers three obscure bands every month will watch that number fall — exactly like someone who has stopped listening to anything unfamiliar. Both behaviors produce the same curve, because the curve only tracks your distance from the charts.
Kalia says as much himself. He names two forces behind the drift: first, listeners "discover less-familiar music genres that they didn't hear on FM radio as early teens, from artists with a lower popularity rank"; second, they return to the music that was popular when they were coming of age, which has since dropped off the charts.
In other words, the leading cause of this supposed taste freeze is discovery. People do not stop searching. They search somewhere other than the top 50. The version that circulates online flatly inverts its own source's explanation.
The other number, the one from Deezer
A second figure circulates alongside it: "musical paralysis" supposedly arrives at 30 years and 6 months. That one comes from a 2018 survey commissioned by Deezer of 1,000 people in Britain, which also placed peak discovery at 24 years and 5 months. The barriers respondents named: feeling overwhelmed by choice (19%), a demanding job (16%), and caring for young children (11%).
Interesting — but self-reported. People were asked to recall their habits, not measured as they listened. And the press coverage moved the number around: some outlets reported 28 rather than 30 and a half.
Two different figures, produced by two incomparable methods, both commissioned by streaming services, and neither published in a scientific journal. That is the actual evidence base behind a claim people repeat as settled fact.
What the research does say
Academic work on this exists, and it is more careful. The reference study is Arielle Bonneville-Roussy and colleagues, published in 2013 in the Journal of Personality and Social Psychology, drawing on more than a quarter of a million people. Their findings: the importance people assign to music declines with age, and young people listen far more often than middle-aged adults. But crucially, preferences shift — intense and contemporary dimensions recede, while sophisticated and unpretentious ones grow.
Shifting is not freezing. A taste that changes direction is still a taste in motion.
A separate body of work explains why adolescence carries so much weight. The "reminiscence bump" describes how music from our younger years stays loaded with memory in a way later music does not. Carol Lynne Krumhansl and Justin Zupnick showed in 2013, in research published by the Association for Psychological Science, that young adults held strong memories not only for the music of their own teens but for their parents' generation as well — cascading bumps. A 2024 review in Frontiers in Psychology places that bump between ages 10 and 30, peaking around 14, and attributes it to the adolescent brain's heightened sensitivity to musical reward.
None of this describes a switch that flips at 33. It describes a gentle slope, a durable teenage imprint, and preferences that keep evolving.
The only statistic that concerns you is your own
Every one of these values is a median. A median describes nobody in particular; it describes the middle of a population. Kalia measured one variable that makes the point nicely: becoming a parent moves the indicator about as much as aging four years. That is not biology. That is a calendar.
Which points to the useful question. Not "at what age will my taste freeze," but "what am I actually listening to right now." Three concrete markers beat one fateful birthday:
- Distinct artists played over a month. This is the closest thing to what we loosely call curiosity.
- Share of first-time plays — tracks you have never played before, regardless of release date. Finding a 1974 album is a discovery.
- Rotation size: how many tracks make up the bulk of your week. A rotation that narrows tells you more than an age ever will.
All three already sit in your listening history. Connecting your Spotify account to Kantise lets you follow them over time and set them alongside the rest of your daily life, rather than trusting an average computed from strangers. That is the point of a quantified self approach to wellbeing: look at your own data before accepting anyone else's.
And if your rotation genuinely is tightening, that says nothing about your age. The Deezer survey names the real reasons, and they are mundane: workload, children, too much choice. Those are life constraints, not neurological fate. They loosen about as easily as they set in — a discovery playlist left running while you do the dishes is often enough to restart the engine.
This echoes what we found looking at the tracks people play on repeat without tiring of them: repetition is not impoverishment, it is one listening mode among several. And as we noted while exploring what listening data reveals about mood, a history always reads better over months than in a single snapshot.
What to take away
The "taste freeze at 33" is not a pure myth. It rests on a genuine analysis of genuine data. But that analysis measures your distance from the charts, not your appetite for discovery. The precise number was manufactured in the retelling, and the phrasing that circulates contradicts the explanation its own author gave.
You do not have a deadline. You have a rotation, and it can be looked at.
FAQ
Is the age-33 figure simply wrong?
It is mostly invented. The Spotify analysis it comes from refers to the "early" and "mid" thirties and never mentions 33. That precise number was added by press coverage in 2015.
Do people really listen to less new music as they age?
What is documented is that they listen to less very popular music. That is a different thing. Part of the decline is caused by discovering lesser-known artists, as the analysis itself states.
Why does music from adolescence stay with us?
It is the reminiscence bump. The 2024 review in Frontiers in Psychology places it between ages 10 and 30, peaking near 14, and links it to how strongly the adolescent brain responds to musical reward.
Does having children actually change what you listen to?
According to the 2015 analysis, becoming a parent shifts the listening indicator about as much as aging four years. The 2018 Deezer survey backs this up: 11% of respondents named childcare as a barrier to discovery.
How can I tell if my own curiosity is fading?
Track three things in your history: distinct artists per month, the share of tracks you have never played before, and the size of your weekly rotation. Those describe you. A population median does not.
Does finding an old album count as discovery?
Yes — and that is exactly what popularity metrics cannot see. A record from 1974 you found this week is new to you, even though it pushes your "current music" score down.
Ready to look at what you actually play?
Instead of asking at what age your taste is supposed to freeze, watch your real rotation across a few months. It is the only curve that is about you.
Sources
- Kalia, A. (2015). "Music was better back then": When do we stop keeping up with popular music? Skynet & Ebert. Read
- NME (2015). People stop listening to new music at 33, study shows. Read
- Billboard (2018). Over 30? Deezer Survey Says You May Suffer From "Musical Paralysis." Read
- Bonneville-Roussy, A., Rentfrow, P. J., Xu, M. K., & Potter, J. (2013). Music through the ages: Trends in musical engagement and preferences from adolescence through middle adulthood. Journal of Personality and Social Psychology. Read
- Krumhansl, C. L., & Zupnick, J. A. (2013). Cascading Reminiscence Bumps in Popular Music. Psychological Science / APS. Read
- Kudaravalli, S., Kathios, N., Loui, P., & Davidow, J. (2024). Revisiting the musical reminiscence bump. Frontiers in Psychology. Read
- Parris, D. When Do We Stop Finding New Music? A Statistical Analysis. StatSignificant. Read
Kantise is an observation tool, not a medical device. The information in this article is provided for general purposes and does not replace advice from a healthcare professional.