Skip to main content

Neurocomputational mechanisms underlying age-related performance changes in goal-directed decisions from experience

Article Category

Article available in the folowing languages:

Finding the key to making better decisions in old age

Many of us will be lucky enough to live longer than our grandparents, but the quality of our lives in old age will depend on successfully maintaining our decision-making capabilities. EU project AGERISK carried out fundamental research that could help societies help older people remain independent decision-makers.

Digital Economy

Older people often make worse goal-directed choices than their younger counterparts, a change that could affect key decisions about health and finances, found fundamental research on EU project AGERISK. The study could one day help societies implement action to boost the quality of our lives as we age. Researcher Dr Job Schepens, with support from the Marie Skłodowska-Curie programme, identified changes in the way older adults make choices depending on the complexity of the environment. “It seems that older adults rely longer on simpler learning strategies, the more demanding the choice environment becomes,” wrote Dr Schepens and colleagues in ‘Aging of the Exploring Mind: Older Adults Deviate more from Optimality in Complex Choice Environments’, a paper published on the study’s results. The authors compared 32 older volunteers and 29 younger adults as they tackled a series of bandit problems, a test based on participants having a slot machine with a number of arms and a limited number of trials in which to accumulate money by pulling on different arms. They had to decide which arm to pull: the arm that gave the highest reward in the past or another in the hope of getting a better one?

Entering bandit territory

The bandit test poses the ‘explore or exploit’ trade-off: ‘Should you choose what you know and get something close to what you expect (exploit) or choose something you aren’t sure about and possibly learn or acquire more (explore)?'. “Older adults chose the highest expected value option about 5 % less often than younger adults did,” found the researchers. “The exploration phase seems to happen sub-optimally in older adults and in a more varied way.” The researchers believe this may occur because we find it harder to construct rich imaginations of future rewards the older we are. Besides lab-based experiments, Dr Schepens is also gathering data at a larger scale online, drawing on his degree in artificial intelligence and his doctorate in computational linguistics. As part of the Marie Skłodowska-Curie fellowship, he also received training in computational models of reinforcement learning. The researchers have presented their findings at conferences in a number of different fields, including developmental cognitive neuroscience and neuroeconomics. “The results should be of interest to scholars in ageing, decision-making and neural plasticity and learning,” says Dr Schepens. “The project is raising new research questions about ways to enhance mental representations of complex choice environments.” Universities and educational companies trying to target mature students may be particularly interested in AGERISK’s insight into how older adults learn. More studies will be needed to build on the initial findings of the fundamental research, but the researchers believe this area is crucial for policy-makers as our populations age. “The insights the project offers should be able to help inform how older adults can remain independent decision-makers,” says Dr Schepens.


AGERISK, decision-making, ‘explore or exploit’ trade-off, bandit test, neuroscience, neuroeconomics

Discover other articles in the same domain of application