William J. Hoyer - US grants

The long-term objective of this research is to identify the knowledge-based and capacity-based factors which are associated with improvement, maintenance, and decline in effective cognitive functioning in healthy older adults. The first proposed experiment seeks to clarify the nature of the effects of prior domain-related knowledge on adult age differences in summarizing the information available in text materials in the target domain. Proposed experiments 2, 3, and 4 examine age-related and skill-related differences in searching for target items in "visually-rich" displays. It is expected that the beneficial effects of skill-related knowledge on the cognitive performance of older adults will be demonstrated by contrasting the visual search performance of older adults under domain-specific and domain-general visual display formats. Adult age differences in the effects of domain-related skill on reducing the capacity demands of visual search are assessed using a dual-task methodology (comparing Experiments 3 and 5). Older adults are expected to show capacity-based limitations in domain-general visual search but not in domain-specific search, and they are expected to show less transfer of the domain-specific skills of visual search to the task of identifying and summarizing the main points of text materials in the target domain.

The overall aim of this project is to specify the mechanisms that are responsible for age-related changes in visual-cognitive performance. Age-related change in several types of visualcognitive performance will be described and explicated in terms of empirically-derived neural network models of aging. Specifically, the factors that contribute to general and task-specific interindividual and intraindividual variability in healthy young, middle-aged, and older adults will be identified and modeled. The proposed neural network models have the unique property that their constituent interconnections can be adjusted so as to represent time-related and practice-related change in the encoding and subsequent transformation of visual information. During the three years of this project, about 400 men and women between the ages of 19 years and 80 years will be tested on measures of 1) perceptual closure or figure completion, and 2) the perception of apparent motion. For each visual-cognitive task, several neurologically plausible network models will be developed, tested, and contrasted with each other in terms of their ability to simulate general and task-specific differences in the distributions of response times and errors for young adults, middle-aged adults, and older adults. The outcome of this line of investigation will have clear implications for the understanding of the mechanisms responsible for effective cognitive functioning during the middle and later years of life.

The proposed experiments build on a program of research aimed at understanding the potentials and limits of effective cognitive functioning during the adult years. Specific aims of the proposed research are 1) to describe cognitive aging in terms of the factors that regulate the rate of skill acquisition, and 2) to describe age differences in the relative efficiency of selected forms of skill learning (item learning, rule learning, feature learning, context learning, and sequence learning). Performance of younger and older adults under conditions that contrast forms of learning (e.g., simple item learning with rule learning) is expected to provide tests of general theories as well as developmental theories of skill acquisition. During the five years of this project, about 1000 women and men between the ages of 20 and 70 years will be tested in eight experiments. Experiments 1-4 examine age-related differences in the effects of practice on item learning, computational speedup, and the shift from computation to item learning under conditions that favor either item learning or computation. Experiments 5-7 are designed to contrast different forms of learning with one other so as to weigh the contribution of each to overall skill. The data from Experiments 5-7 will allow specific age deficits in learning to be identified, and will exploit those deficits to test general theories of learning and automatization. Experiment 8 investigates inter-session disruption and retention effects by age. The outcomes of the proposed research contribute to the understanding of the effective conditions of skill acquisition and retention throughout the adult life span.