Abstract
Learning is shown to play a key role in facilitating performance in a wide range of perceptual skills in both young and older adults. But why do some people learn better or faster than others? Here, we investigated whether individual variability in cognitive functions accounts for differences in learning ability in young and older adults. We tested the ability of young (n=30, 18-30 years) and older (n=30, 65-90 years) adults to discriminate radial and concentric global forms (Glass patterns) embedded in background noise before and after training. We manipulated a) the amount of background noise, and b) the similarity between global forms, using linear morphing between concentric and radial patterns. We trained participants at either fixed or variable amounts of background noise across trials for four to five daily sessions. In addition to this perceptual learning task we tested a range of cognitive abilities, including cognitive inhibition, visual short term memory, selective and divided attention. Our results showed that learning improved performance (i.e. decreased discrimination thresholds) in both young and older adults. Correlating performance in the cognitive and perceptual learning tasks showed that high performance in cognitive inhibition correlated significantly with threshold reduction after training. Interestingly, learning performance correlated significantly with performance in attentional tasks when observers were trained with variable amounts of background noise across trials. In contrast, for fixed amounts of background noise, learning performance correlated significantly with visual short term memory. Finally, threshold reduction correlated more strongly with performance in attentional tasks for good learners (i.e. individuals with high learning performance), while performance in memory tasks for weak learners. Our findings suggest that a) performance in attentional and memory tasks may account for individual variability in learning, and b) task context determines which cognitive abilities are critical for visual learning across the lifespan.
Meeting abstract presented at VSS 2012