Relationships in Aging, Cognitive Processes, and Contingency Learning

This study investigated the influence of age, processing speed, working memory,and associative processes on the acquisition of contingency information. Young and older adults completed positive (+.65) and negative (-.65) contingency tasks that measured their ability to discover the relationship between a symptom (e.g., FEVER) and a fictional disease (e.g., OLYALGIA). Both d\u27 scores, i.e., contingency learning, and contingency estimates, i.e., contingency judgment, were examined. Participants were also asked to complete cognitive tasks that measure the constructs of processing speed, working memory resources, associative memory, and associative learning. Structural equation modeling was used to examine the direct and indirect relationships between processing speed, working memory resources, associative memory, associative learning, and positive and negative contingency learning and judgment for young and older adult groups. Young adults outperformed older adults on the cognitive tasks and on contingency learning and judgment tasks. However, age differences were smaller for the positive contingency than for negative contingency. A comparison of the structural equation models for young and older adults showed no relationship between any cognitive construct and negative contingency learning. However, young adults\u27 judgment for the negative contingency was directly influenced by associative learning, while their learning and judgment for the positive contingency was directly influenced by associative memory. For older adults, working memory executive function directly influenced their judgment for the negative contingency and their learning and judgment for the positive contingency. Processing speed had an indirect effect on older adults\u27 contingency learning and judgment that was mediated by working memory executive functioning. The differences in the young adults\u27 models as well as the difference between the young and older adults\u27 models for positive and negative contingencies suggest that while associative processing is important, it may not account for all of the variation in contingency learning and judgment. The young adults\u27 models for the negative contingency task indicates that higher level processes, such as inductive reasoning, maybe involved in negative contingency judgment because the associative learning task required some level of hypothesis testing. In contrast, positive contingency learning and judgment could rely primarily on more basic associative processes. The present findings therefore suggest that an overall model of contingency learning must include both associative processes and inductive reasoning processes. Older adults\u27 general contingency performance was most directly related to their working memory executive functioning, suggesting that the decline in their working memory has the strongest effect on their ability to acquire and use information about contingencies. In fact, the age related decline in working memory seems to affect older adults\u27 ability to acquire both positive and negative contingencies. The similarities across the older adult models for positive and negative contingencies indicate that the underlying deficit in older adults\u27 working memory executive functioning that affects their overall contingency learning and judgment performance. This basic working memory executive functioning deficit for older adults also explains why their models for positive and negative contingency did not exhibit direct relationships between associative tasks and contingency learning as observed for the young adult models