(September 2026) Hobbies like doing your daily crossword, tending the garden, dancing, or singing in a choir have all been linked with keeping a sharp mind in later life. But most research into hobbies and brain health stops at one question: do hobbies delay decline? 

A new study published in the European Journal of Ageing, using data from the Survey of Health, Ageing and Retirement in Europe (SHARE) and our three sister surveys in England, the US, and China, went a step further by asking whether hobbies can help once decline has already begun.

Hobbies and the theory of cognitive reserve 

Let’s start with why hobbies might make a difference to brain health in the first place.

One explanation is the theory of cognitive reserve. The idea is that a stimulating environment helps the brain build stronger, more efficient connections, creating a reserve that can stave off decline and illness, much like a strong muscle in your leg won't waste as quickly as a weaker one. 

Hobbies, viewed very generally, tap into three areas that are particularly brain healthy: they tend to require some kind of mental effort, they are often done with other people, and they can be a good way to let off steam. 

Hobbies and cognition in over 120,000 people aged 50+ across 31 countries  

Yunjia Cai, Yi Sun and Fan Yang from Fudan University in Shanghai combined data from four major ageing surveys: SHARE (Waves 5 to 8, covering 2013 to 2019), the English Longitudinal Study of Ageing (ELSA), the US Health and Retirement Study (HRS) and the China Health and Retirement Longitudinal Study (CHARLS). SHARE made up the largest share of the sample, with almost 78,000 participants from 28 countries. Harmonised versions of the surveys can be accessed through the Gateway to Global Aging Data, which provides matching variable definitions across the international family of ageing studies.  

Hobbies were defined broadly, covering activities such as reading, games, arts and crafts, sport, gardening, volunteering and taking part in social clubs. The US and English surveys just ask a binary question, whereas SHARE and our Chinese counterpart ask about a whole list of activities covering volunteering, social clubs, reading, games and educational courses. The researchers turned these into a matching yes/no measure so everyone could be compared on the same terms. 

Cognitive function was measured with tests of word recall, orientation to date and time, and executive function (counting backwards from 100 in sevens). Based on their scores, participants were placed into one of three groups: normal cognition, mild cognitive decline or severe cognitive decline. Importantly, these groups are based on test scores, not clinical diagnoses of dementia. 

Death was included as a fourth group. This matters because people who die during follow-up can no longer decline or recover, and ignoring them can distort the results. 

Tracking the dynamic nature of hobbies longitudinally 

A lot of previous research in this area has asked whether someone has a hobby at a single point in time, and then looked at how their brain health fared further down the line. Anyone who has a hobby or two will know this is not ideal. A single snapshot can't tell apart someone who has just picked up a new pastime from someone for whom hobbies have been embedded into their daily life for years.

Fortunately, because all four surveys follow the same people over many years, the researchers were able to track hobby engagement over time and then see how changes in it related to changes in brain health. 

The team used a sophisticated method that other longitudinal researchers may well be interested in, called a continuous-time multi-state Markov model. A simple way to think about it is as a map with four stations (normal, mild, severe and death), where the model estimates how often people travel between each pair of stations and whether having a hobby speeds up or slows down each journey. 

This approach has two big advantages for longitudinal survey data. First, it copes with the fact that the time between interviews varies from person to person and survey to survey. Second, it allows people to move backwards as well as forwards, from mild decline back to normal cognition for example. Many earlier studies only looked at whether people got worse. 

The researchers also took account of a wide range of other factors that could affect both hobbies and brain health, including age, sex, education, income, employment, marital status, smoking and whether people lived with their children. 

Is cognitive decline a one-way street? 

Cai and colleagues first looked at whether hobbies were linked to a lower risk of decline. In Europe, people with hobbies were 40% less likely to move from normal cognition into mild decline, and similar results were found in China (34%) and England (20%). In the USA, the association pointed in the same direction but was not statistically significant. In China and Europe, hobbies were also linked to a lower risk of moving on from mild to severe decline. 

The recovery finding is more unusual. Among people who had already slipped into mild cognitive decline, those with hobbies were more likely to return to normal cognitive function: 37% more likely in England, 21% in Europe and 15% in the USA. In Europe, hobbies were also linked to a higher chance of moving back from severe to mild decline. 

It is worth noting that “recovery” from mild decline, as per cognitive tests, was surprisingly common overall. Over five years, around two thirds of people in England and Europe who were in mild decline were estimated to return to normal cognition, compared with around half in the USA and China. While the results offer some tentative optimism, some "recovery" may also reflect the normal ups and downs of test scores from one survey wave to the next – for example some days participants might just not be feeling as sharp. This may be the case especially for people whose scores sit close to the line between two groups. 

Hobby participation varies widely around the world 

The ability to take part in hobbies is in no way down to individual factors alone. If you look within one country, you can examine how differences between people, like in terms of their wealth or mobility, affect whether they take up hobbies. Cross-national surveys like ours also let you see how much further-reaching differences between places matter, like leisure infrastructure available to the community, affordability, family structures and societal expectations. 

This is illustrated perfectly in the study. Around three quarters of English participants reported having a hobby, compared with only about a quarter of those in China, where many older people are still working (almost six in ten) or living with their adult children (over half), which may leave less time for leisure. Differences like these offer a useful natural comparison of how cognition varies between populations, while taking into account a whole host of other factors that might affect brain health. 

A note of caution 

The study is observational, so it cannot prove that hobbies cause better cognitive outcomes. People whose memory is starting to slip may give up their hobbies long before memory changes are detected on tests, which could make the link appear stronger than it is. The four surveys also asked about hobbies in different ways, which makes direct comparisons between countries harder. 

That said, the researchers put their results through a series of extra checks. They took account of chronic illness and depressive symptoms, repeated the analysis for people aged 65 and over, and adjusted for participants who dropped out of the surveys over time. The main findings held up throughout. 

Where could research go next? 

The study leaves plenty of room for others to build on. Because SHARE asks about specific activities and how often people take part in them, researchers could look beyond a simple yes/no measure to see whether some types of hobby, or more regular participation, matter more than others. The study also grouped all SHARE countries together as "Europe". With 28 countries in the survey, there is lots of scope to explore how pension systems, community resources and cultural habits shape the link between leisure and brain health from one country to the next. 

Why does this matter? 

Over 55 million people worldwide live with dementia, and that number is expected to rise sharply as populations age. The results fit a growing body of evidence that staying engaged in enjoyable activities is good for the ageing brain. The authors argue that making hobbies accessible and affordable, particularly in countries where participation is low, should be part of public health efforts under the WHO Decade of Healthy Ageing. 

Interested in conducting research into cognitive ageing with SHARE data? CLICK HERE