Three hypotheses regarding the locus of age differences in working mem
ory were examined in an experiment in which young, middle-aged, and ol
d adults remembered the names of persons standing in line while also p
erforming mental addition. Instructions as to the relative importance
of these two tasks were manipulated. The results are highly compatible
with the hypothesis of an age-related deficit in a single pool of res
ources that can be allocated to different tasks. They do not support h
ypotheses of deficits in one or more pools used separately for process
ing and storage functions. Middle-aged adults performed as well as you
ng adults and better than old adults.