What the data actually shows
The fluid-versus-crystallized distinction, developed by Raymond Cattell and John Horn, is one of the better-supported frameworks in the field. Across studies, fluid abilities (working memory, processing speed, novel reasoning) tend to peak in early adulthood and decline gradually, while crystallized abilities (knowledge, vocabulary, expertise) often keep rising into one's 60s or beyond. Different abilities peak at different ages, so there is no single age at which a person is simply 'smartest.'
The Flynn effect shows intelligence is not fixed at the population level either. Across much of the twentieth century, average IQ test scores rose substantially — on the order of roughly three points per decade in many countries — almost certainly faster than genes could change. This points to environmental drivers like better nutrition, education, and more abstract, cognitively demanding daily life. In some countries these gains have slowed or reversed in recent decades, which is itself an active area of study.
The weak spot is general transfer. Commercial brain-training programs reliably make you better at the trained task, but reviews and consensus statements have found little good evidence that this transfers to broader cognitive ability or everyday functioning. By contrast, building knowledge and specific skills clearly works — you genuinely know and can do more — it just does not act like a global upgrade to raw processing power.
Why this feels different from how it actually is
Intelligence feels fixed partly because it is often presented that way — as a single number you are born with and stuck with. IQ does show meaningful stability and a substantial genetic component, which is real, but stability across a population is not the same as being unchangeable, and it leaves a great deal of room for growth through learning and environment.
It also feels fixed because the two kinds of intelligence change in opposite directions, which is easy to misread. Noticing that you are slower to grasp something new than you were at 20 can feel like decline, even as your knowledge, judgment, and skill in familiar areas are still improving. The visible slowdown in one component masks the quiet growth in another.
And the brain-training industry sells the opposite illusion — that a few minutes of games will make you globally smarter. When the promised general boost does not materialise, it can reinforce a sense that intelligence is simply fixed, when the truer picture is that the right kind of effort (deep learning, not games) grows the kind of intelligence that actually compounds.
You can clearly get more knowledgeable, skilled, and capable across your whole life — which is most of what 'smart' means in practice.
What the research says to do about it
Keep learning real things deeply, because that is what reliably grows. Building knowledge and skills expands crystallized intelligence directly, and there is no clear ceiling to it across a normal lifespan. Sustained, effortful learning in areas you care about does more than any 'cognitive enhancement' product.
Look after the general conditions that support cognition. Physical activity, sleep, managing cardiovascular health, and staying socially and mentally engaged are associated in the research with better cognitive ageing. These are not magic, and effects are modest, but they are far better supported than brain-training games. (This is general educational information, not medical advice — see a qualified clinician for anything concerning.)
Aim your effort at depth rather than at raising a number. Because crystallized intelligence and expertise grow through focused practice and accumulated knowledge, the practical move is to go deep in specific domains rather than chasing a global IQ boost. That is the version of 'getting smarter' the evidence actually supports.
What the research says does not help
Brain-training apps and games are the clearest example of something that overpromises. They improve the trained task and little else; reviews have found weak evidence that the gains transfer to general intelligence or daily life. Expecting them to make you broadly smarter is not supported.
Treating your IQ as a fixed verdict on your potential is both inaccurate and self-limiting. Scores are meaningfully stable but not immutable, the Flynn effect shows whole populations shifting, and most of what makes someone capable in real life — knowledge, skill, judgment — keeps growing with learning. A single number is a narrow snapshot, not a ceiling.
Assuming that getting older means inevitable mental decline overstates the case. Some abilities (speed, novel reasoning) do tend to slip gradually, but others (vocabulary, expertise, accumulated knowledge) often keep improving for decades. The honest picture is a trade-off across abilities, not a simple downhill slide.
Stability across a population is not the same as being unchangeable, and it leaves a great deal of room for growth.
What this looks like in real life
Slower at 40, but knowing more
Someone notices they grasp a genuinely new idea more slowly than they did at 20 and reads it as decline. But the two kinds of intelligence move in opposite directions: the visible slowdown in fluid ability masks the quiet growth of knowledge, judgment, and skill in familiar areas. It is a trade-off across abilities, not a simple downhill slide.
The brain-training app that doesn't transfer
A few minutes of games each day reliably makes you better at the games and little else. When the promised general boost never materialises, it can reinforce a sense that intelligence is simply fixed — when the truer picture is that the right kind of effort, deep learning rather than games, grows the intelligence that compounds.
Real numbers in context
The Flynn effect is the headline number worth knowing: average measured IQ rose by roughly three points per decade across much of the twentieth century in many countries — far too fast to be genetic — which is strong evidence that environment and the demands of modern life shape measured intelligence. In several places these gains have plateaued or reversed recently, and why is still debated, so this is a real but evolving finding.
On the two kinds of intelligence, the pattern rather than a precise age is the useful part. Fluid abilities tend to peak in early adulthood and decline slowly thereafter, while crystallized abilities often keep rising into the 60s or later. The practical takeaway is that 'can you get smarter?' is best answered 'yes, in the knowledge-and-skill sense that matters most day to day — and that part has no obvious ceiling — while raw speed is harder to change.'