Task-specific memory self-efficacy (TSMSE) was experimentally manipulated through provision of information about task difficulty, to determine its effect on free recall for 56 older (age 63-86) and 56 younger (age 16-25) adults. The implications of using prediction-based measures of TSMSE were addressed. After completing one recall trial of a list of 20 words, half the participants were told a second list comprised more difficult words; the others were told the second list would be similar to the first they had received. Free recall and TSMSE were measured before and after this manipulation. The manipulation reduced TSMSE for participants expecting a harder list of words, but not differently for younger compared with older adults. Younger and older adults' recall declined at the second recall trial, but there was no difference between those expecting a harder list and those expecting a similar list. Recall was predicted by domain-specific memory self-efficacy as well as a traditional measure of TSMSE. The study demonstrated the malleability of memory self-efficacy, but called into question assertions about its salience as a mediator of older adults' poorer memory performance.
Copyright 2018 Elsevier B.V., All rights reserved.