Fragmented pictures revisited: Long-term changes in repetition priming, relation to skill learning, and the role of cognitive resources

Background: Whereas age-related declines in declarative memory have been demonstrated in multiple cross-sectional and longitudinal studies, the effect of age on non-declarative manifestations of memory, such as repetition priming and perceptual skill learning, are less clear. The common assumption, based on cross-sectional studies, is that these processes are only mildly (if at all) affected by age. Objective: To investigate long-term changes in repetition priming and age-related differences in identification of fragmented pictures in a 5-year longitudinal design. Method: Healthy adults (age 28-82 years) viewed drawings of objects presented in descending order of fragmentation. The identification threshold (IT) was the highest fragmentation level at which the object was correctly named. After a short interval, old pictures were presented again along with a set of similar but novel pictures. Five years later the participants repeated the experiment. Results: At baseline and 5-year follow-up alike, one repeated exposure improved IT for old (priming) and new (skill acquisition) pictures. However, long-term retention of priming gains was observed only in young adults. Working memory explained a significant proportion of variance in within-occasion priming, long-term priming, and skill learning. Conclusion: Contrary to cross-sectional results, this longitudinal study suggests perceptual repetition priming is not an age-invariant phenomenon and advanced age and reduced availability of cognitive resources may contribute to its decline.