This proposal constitutes an application for an ADAMHA Research Scientist Award (RSA). The research activity consists of new experimental investigations focusing on working memory constraints and individual differences in visual reasoning, as well as new theory development. The PI also intends to extend her familiarity with the experimental methodologies and theories in the neuropsychology of language, as well as connectionist simulation techniques. The proposed research examines how the limited capacity of working memory constrains problem solving and visualization in visual reasoning tasks. One series of experiments investigates a problem-solving task whose solution requires that subjects mentally generate a series of goals and subgoals and then retrieve them appropriately. The hypothesis is that experimental manipulations that increase the goal management load should cause selective decrements in performance, partly at times when the load is high and particularly for individuals who have a low capacity for mental goal management. A second series examines the working memory demands of spatial and non- spatial strategies that can be used in a deductive reasoning task (linear syllogisms). The experiments examine what aspect of working memory resources determines the shift from one strategy to the other, and what the processing consequences are of each strategy. Other studies investigate complex visual tasks that require the visualization of mechanical motion. The process of understanding a mechanical device involves both the goal management that is found in ordinary reasoning, as well as the visualization of constrained motion. These experiments examine the contribution of goal management ability and spatial ability to performance in complex visualization tasks. The dependent measures to be used include both response latencies (the time to choose a next move in a problem or to mentally generate a constrained spatial transformation) and error patterns. In addition, pupil dilation measures will be used to index the consumption of cognitive resources. The obtained results will be integrated within a capacity-constrained computational model of working memory. Within this model, a production- system architecture with graded activations propagated by productions, both the processing and storage functions of working memory are fueled by activation. The total amount of activation available to the system has an upperbound that corresponds to the maximum capacity of an individual. The goal of the theory is to explain how performance and strategy selection in visual reasoning are influenced by working memory capacity. This research can indicate how the availability of cognitive resources affects visual thinking in different individuals, from the most gifted to those with neurological impairments.
| Newman, Sharlene D; Just, Marcel Adam; Keller, Timothy A et al. (2003) Differential effects of syntactic and semantic processing on the subregions of Broca's area. Brain Res Cogn Brain Res 16:297-307 |
