Task Learning as a Mechanism of Transfer in Cognitive Intervention: Neuro-cognitive Predictors and Outcomes of Early, Middle and Late Stages of Task Learning

Investigation into methods of addressing cognitive loss exhibited later in life is of paramount importance to the field of cognitive aging. The passive protective factors of cognitive reserve and continued education, as well as the active factor of cognitive intervention, have all been found to ameliorate expected declines in cognition in adults aged sixty-five and up, and all three of these factors heavily involve the learning process. This dissertation presents three studies derived from a longitudinal cognitive intervention, each designed to illuminate factors which influence the learning process and in turn how that process bolsters cognition. The cognitive intervention in question was a working-memory-based video-game-like training task, designed to be both engaging and adaptive to the abilities of individual participants. The first study identified a measure of verbal episodic memory as well as the volume of a brain region involved in language, verbal memory and cognitive control (the left inferior frontal gyrus) as predictors of individual learning rates on the training task. These two neuro-cognitive measures were more predictive of task learning when considered in conjunction than when considered separately, indicating a complimentary effect. The second study compared daily performance on the training task with several daily factors known to influence cognition, including perceived wellbeing, stress, business, and sleep. Auto-regressive analyses conducted in Study 2 were able to identify meaningful predictors of performance-over-time on the training task in fifty percent of cases. This pattern of influences varied greatly between participants, indicating a highly individualized influence of these variables. The third study observed that individual differences in learning of the training task were related to training-related gains in a measure of nonverbal reasoning, with participants who learned the training task faster showing relatively greater transfer (i.e. gains) to that measure of reasoning. Collectively, the three studies presented in this dissertation offer a novel insight into training-related cognitive benefits via the identification of a discreet “path of transfer” resultant from this training. Specifically, these studies identify a pattern of influence by which verbal episodic memory (and its related brain region) is determinant of learning of a working-memory task (the training task), and learning of that task itself is determinant of training-related gains to nonverbal reasoning. This pattern of findings serves as a testable hypothesis for future studies of working-memory based cognitive training in older adults, and this “path of transfer” approach may serve as a useful tool in examining the concept of transfer more generally.