Age, memory changes, and the varying utility of recognition as a media effects pathway

Abstract

Numerous media campaign evaluations use self-reported exposure measures in assessing media effects, and many of these measures rely on participant recognition of campaign material. Rather than accepting the utility of such measures at face value, however, we should probe their limits. We can predict at least one theoretically important limit on the basis of what we know about age-related memory changes. Specifically, the utility of a recognition item as an indicator of past exposure should decline as audiences age, especially for the most elderly. Analysis of data from a science communication project evaluation offers support for these ideas, demonstrating that age predicts recognition error and that the relationship between experimentally assigned physical exposure and subsequent self-reported recognition wanes among adults 70 and older. We also examine the possibility that media content might nonetheless affect older adults in ways often overlooked by reliance on recognition items.