Intact reinforcement learning in healthy ageing

Area of Science:

• Cognitive neuroscience
• Neuroscience of aging
• Behavioral neuroscience

Background:

• Reinforcement learning (RL) performance in older adults is inconsistent across studies.
• Previous research suggests task complexity may explain age-related differences in RL.
• Older adults may exhibit preserved cognitive functions in more complex scenarios.

Purpose of the Study:

• To investigate the impact of age on reinforcement learning in a complex navigation task.
• To determine if increased task complexity influences age-related performance differences in RL.
• To examine whether higher-level cognitive functions remain intact in older adults.

Main Methods:

• Employed a complex navigation task, distinct from simpler RL paradigms.
• Compared the performance of older adults to younger adults.
• Analyzed performance efficiency over a sufficient number of trials.

Main Results:

• Older adults initially demonstrated less efficient performance in the navigation task.
• With adequate trial experience, older adults' performance became comparable to that of younger adults.
• The findings suggest that age does not universally impair cognitive abilities.

Conclusions:

• Ageing does not necessarily lead to a decline in reinforcement learning capabilities, especially in complex tasks.
• Higher-level cognitive functions, such as those involved in navigation and learning, appear largely preserved in older age.
• Task complexity and learning opportunities are critical factors in understanding age-related cognitive performance.