Brain activation in young and older adults during implicit and explicit retrieval

Positron emission tomography was used to study regional cerebral blood flow (H2(15)O method) in groups of young and older adults during implicit and explicit retrieval, following a procedure devised by Squire et al. (1992). At study, subjects were exposed to four lists of words. Following list presentation, subjects were presented with three-letter word stems under four conditions: (1) silent viewing, (2) completion of word stems that could not form words from the study list, with the instruction to provide the first word that came to mind (baseline), (3) completion of word stems, half of which could form words from the study list, with the instruction to provide the first word that came to mind (priming), and (4) completion of word stems, half of which could form words from the study list, with the instruction to use the stems as cues for recall of list words (memory). The behavioral data indicated an agerelated deficit in cued recall that was reduced in priming. Both age groups showed a similar decrease of blood flow in right posterior cortex during priming relative to baseline. During cued recall, bilateral increases of blood flow were observed in prefrontal cortex and anterior cingulate gyrus for both age groups. The young adults showed selective increases of activity in left cerebellum and Wernicke's area, whereas the older adults showed a selective bilateral activation in the perirhinal region of the medial-temporal cortex during cued recall. The results suggest a simiiar biological basis of priming in both age groups: a decrease in the neural activity required to process a particular stimulus during a subsequent encounter compared with a previous one. In addition, the importance of prefrontal regions for conscious retrieval was substantiated and extended to late adulthood. Finally, the agedifferential activations observed during cued recall were discussed relative to prominent concepts in the current literature on cognitive aging (e.g., speed of processing, self-initiated operations, cross-modal recoding).