Email updates

Keep up to date with the latest news and content from BMC Geriatrics and BioMed Central.

Open Access Research article

Cognitive function is associated with risk aversion in community-based older persons

Patricia A Boyle12*, Lei Yu13, Aron S Buchman13, David I Laibson4 and David A Bennett13

Author Affiliations

1 Rush University Medical Center, Rush Alzheimer's Disease Center, 600 S. Paulina, Suite 1020, Chicago, IL 60612 USA

2 Department of Behavioral Sciences, Rush Alzheimer's Disease Center, 600 S. Paulina, Suite 1020, Chicago, IL 60612 USA

3 Department of Neurological Sciences, Rush Alzheimer's Disease Center, 600 S. Paulina, Suite 1020, Chicago, IL 60612 USA

4 Harvard University, Department of Economics Littauer M-12, Harvard University, Cambridge, MA 02138 USA

For all author emails, please log on.

BMC Geriatrics 2011, 11:53  doi:10.1186/1471-2318-11-53

Published: 11 September 2011



Emerging data from younger and middle-aged persons suggest that cognitive ability is negatively associated with risk aversion, but this association has not been studied among older persons who are at high risk of experiencing loss of cognitive function.


Using data from 369 community-dwelling older persons without dementia from the Rush Memory and Aging Project, an ongoing longitudinal epidemiologic study of aging, we examined the correlates of risk aversion and tested the hypothesis that cognition is negatively associated with risk aversion. Global cognition and five specific cognitive abilities were measured via detailed cognitive testing, and risk aversion was measured using standard behavioral economics questions in which participants were asked to choose between a certain monetary payment ($15) versus a gamble in which they could gain more than $15 or gain nothing; potential gamble gains ranged from $21.79 to $151.19 with the gain amounts varied randomly over questions. We first examined the bivariate associations of age, education, sex, income and cognition with risk aversion. Next, we examined the associations between cognition and risk aversion via mixed models adjusted for age, sex, education, and income. Finally, we conducted sensitivity analyses to ensure that our results were not driven by persons with preclinical cognitive impairment.


In bivariate analyses, sex, education, income and global cognition were associated with risk aversion. However, in a mixed effect model, only sex (estimate = -1.49, standard error (SE) = 0.39, p < 0.001) and global cognitive function (estimate = -1.05, standard error (SE) = 0.34, p < 0.003) were significantly inversely associated with risk aversion. Thus, a lower level of global cognitive function and female sex were associated with greater risk aversion. Moreover, performance on four out of the five cognitive domains was negatively related to risk aversion (i.e., semantic memory, episodic memory, working memory, and perceptual speed); performance on visuospatial abilities was not.


A lower level of cognitive ability and female sex are associated with greater risk aversion in advanced age.