Aging and the time course of semantic activation.

Fifty-four young (M age = 20.2 years) and 54 elderly participants (M age = 68.4 years) completed a task in which a prime was followed by a target. People were to respond "yes" if the target was a word and "no" otherwise. On word-target trials, three kinds of prime were presented: associated (e.g., the prime-target pair DOG-CAT), unassociated (e.g., SEW-CAT), and neutral (e.g., BLANK-CAT). The stimulus onset asynchrony (SOA) between onset of the prime and target was varied, with SOAs of 150, 450, and 1,000 ms. The most important finding was an age difference in the minimum SOA at which priming (i.e., faster response time on associated than on unassociated trials) was observed. It is argued that the age difference in onset of priming reflects a slowing with age in one or more aspects of semantic activation. Contrary to the Hasher and Zacks (1979) hypothesis, there is evidence of an age difference in the speed of automatic activation.