do: why they exercise or don’t, whether they are racist or ageist, or whether they are more easily persuaded by certain kinds of messages than others. Thus, whereas self-reported information is well suited to answer certain questions, it is seriously limited in answering others.

This paper briefly reviews some methodological and analytical approaches that hold significant promise for the field of aging research, including the measurement of implicit constructs and experience sampling. The measurement of change over time, which is essential to a deep understanding of aging, is also an area in which rapid progress is being made. This overview is followed by three short papers: two address measurement concerns and ways to minimize bias in self-reports; the third looks at the emerging use of neuroscience in social cognition research on aging. Breakthroughs in thought and theory often occur after improvements in measurement techniques and methodology are made; some of the latest developments discussed briefly here illustrate the potential for psychological research on aging.


Explicit constructs and processes are those that are subject to conscious awareness and direct self-report; implicit constructs and processes are not. There has been a recent explosion of interest in social psychology in the area of implicit social cognition, with the term being used in several different ways (see Petty, Wheeler, and Tormala, 2003). Some researchers are interested in implicit constructs such as attitudes, goals, and motives. For example, do people have evaluative predispositions of which they are unaware (e.g., an implicit attitude such as an unrecognized dislike of old people)? Other researchers are interested in implicit biases, in which people are perfectly aware of their attitudes or motives but do not know where they come from, or in implicit effects, in which people are aware of their attitudes or motives but do not know what effect those attitudes or motives have on their thoughts and actions.

To examine implicit constructs and processes, social psychologists have developed a battery of implicit measures that do not call for conscious self-reports of the construct or process. The earliest such measures were in essence disguised self-reports (e.g., thematic apperception tests) or behavioral observations (e.g., how close one sat next to a stranger) from which researchers inferred an underlying attitude or motive. Recently, implicit measures based on reaction times have demonstrated considerable utility in predicting behaviors that could not be predicted by direct self-reports (e.g., Dovidio, Kawakami, and Gaertner, 2002). Furthermore, even when direct self-reports were useful in predicting behavior, implicit measures have been

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement