Characterizing and optimizing the performance of younger and older adults in paired associate tasks: A Markov modeling approach
Developing a theory of learning and a strategy for training paired associates (i.e., associations between stimuli and responses) is important since proficiency in many real-world tasks critically depends on memorizing paired associates. In the first part of this dissertation, I propose a theory for the learning of paired associates that can be mathematically expressed by a Markov model. In a first experiment, this Markov model is found to account nicely for paired associate data of younger and older adults. It is also formally shown that this proposed Markov model is the minimally complex Markov model that can account for critical findings in the paired associate learning literature. In the second part of this dissertation, I investigate if the performance of younger and older adults in paired associate tasks is affected In the same way when different training strategies are employed and I propose a strategy for training paired associates. Specifically, in a second experiment, the performance of younger and older adults is found to be affected in the same way when the training strategy is varied. Then, an optimal training strategy according to the proposed Markov learning model is developed. This strategy is tested in a third experiment and is found to lead to significant improvements over standard practices of training paired associates.
0633: Cognitive therapy
0796: Operations research
0620: Developmental psychology