The attentional demands of mnemonic control processes.

In order to assess the processing (i.e., attentional) demands of different control processes, subjects were required to perform a secondary choice reaction time task in addition to primary verbal tasks. The performance of the secondary task yielded a measure termedexpended processing capacity (EPC), which was used to infer the attentional demands of the primary tasks. Two factors, mediator type (experimenter-supplied vs. subject-generated) and intentionality (incidental vs. intentional), were varied in a paired-associate situation in an effort to affect the degree of elaboration processing. Only mediator type had an effect on recall accuracy and on EPC during both initial processing and recall. Subject-generated mediators resulted in higher recall and in higher EPC during initial processing, but in lower EPC during recall, than did experimenter-supplied mediators. Implications of these findings are discussed.