Background
Most decisions made in everyday life involve choices between immediate and delayed rewards. From seemingly minor choices, such as deciding to pass up mouth-watering French fries in favor of a healthy salad, to larger ones, such as deciding to forego spending money from today’s paycheck to invest it for retirement, opportunities for intertemporal tradeoffs abound. Temporal discounting, the tendency to prefer smaller, immediate rewards over larger, delayed ones, is a concept that has been widely studied in the economics, behavioral economics, neuroscience and psychology literatures and is related to several important real world financial and health behaviors [
1‐
3]. Individuals who discount future rewards tend to spend more, save less, and make poorer investment decisions [
4,
5]. Further, those who discount tend to under-utilize health insurance yet exercise less, use alcohol and drugs more, engage in unsafe sex, and are more likely to be obese [
6‐
8]. Thus, temporal discounting is associated with a vast array of behaviors that have significant individual and societal-level consequences.
Recognition of the important role of temporal discounting in human behavior has generated interest in identifying the factors associated with it, and data from studies of younger and middle-aged adults have demonstrated that something akin to cognitive ability (e.g., performance on aptitude or achievement tests) is negatively associated with temporal discounting [
9‐
13]. To date, however, little is known about temporal discounting among older persons. Some studies suggest that older persons may discount less steeply than younger persons, although findings are mixed [
14‐
19], but we are not aware of studies that have examined whether cognitive ability is related to temporal discounting among older adults. Considering that aging is a time when some of life’s most significant financial and health decisions are made (
e.g., asset decumulation, asset allocation, annuitization, intergenerational transfers, Medicare/health insurance choices, end of life decisions), an understanding of the factors associated with discounting may have major financial and public health implications.
In this study, we examined the relation of cognition, related demographic (i.e., age, sex, education) and contextual (i.e., income, chronic medical conditions) factors with temporal discounting among more than 380 older persons without dementia from the Rush Memory and Aging Project [
20]. All participants underwent detailed cognitive evaluations and assessments of temporal discounting using standard behavioral economics questions in which participants were asked to choose between an immediate, smaller payment of $10 versus a delayed, larger payment that ranged from $10.75 to $30. The associations of global cognitive function and five specific cognitive abilities with temporal discounting were examined via mixed effect models adjusted for age, sex, education, income, and chronic medical conditions.
Discussion
In this study, we examined the association of cognitive function, demographic and contextual factors with temporal discounting in a cohort of more than 380 community-based older persons free of dementia. In a mixed effect model adjusted for age, sex, education, income, and chronic diseases, global cognitive function was significantly associated with temporal discounting. Subsequent analyses indicated that this association was driven by perceptual speed and visuospatial abilities. This is the first study that we are aware of to examine whether demographic, contextual and cognitive factors are associated with temporal discounting among community-based older persons. Findings indicate that a lower level of cognitive function is associated with greater temporal discounting among very old community-based persons without dementia.
In recent years, temporal discounting has been the focus of many psychology, economics, behavioral economics, and neuroeconomics studies of younger persons. Findings from these studies have yielded at least two important insights: first, people do not always behave in a dynamically consistent manner (
i.e., in ways consistent with maximizing utility), as suggested in traditional economic models; and second, temporal discounting is associated with numerous real world economic and health outcomes linked to success and well-being [
3,
4,
32]. Data from highly educated younger persons and adults suggest a negative association between indicators of cognitive ability and temporal discounting, but this association has not been examined in large-scale studies of aging [
9‐
13,
33]. This represents a significant gap in knowledge, given that aging is a time when numerous complex and influential life decisions are made just as cognitive function may begin to decline. Consistent with findings reported in younger persons, we found that cognitive ability is negatively related to discounting among older persons. This association was robust to adjustment for other factors such as income and the number of chronic medical conditions.
Some prior studies have investigated whether discounting rates differ in younger versus older persons, and the prevailing suggestion from these studies is that older persons tend to discount less steeply than younger persons; however, the evidence is mixed [
15,
31,
34,
35]. We are not aware of studies that have directly examined the association of cognition, demographics and related contextual factors with discounting in community-based persons as old as those studied here. Our findings suggest that the level of cognitive function is the major determinant of an older person’s ability to wait patiently for larger but delayed rewards. Given that temporal discounting has been shown to be associated with many important real world decisions involving the ability to delay gratification (e.g., drug and alcohol use, savings behavior) among younger persons, this finding may have important public health implications for older persons. For example, overwhelming evidence from the real world suggests that older persons are vulnerable to poor and impulsive decision-making. Notably, older persons are selectively vulnerable to fraud and lose more than 2.9 billion dollars annually to financial fraud alone (not counting all other types of fraud). Similarly, older persons frequently make poor financial and health choices with respect to personal matters (e.g., many persons take social security distributions early, often to their own detriment; many also take health supplements not supported by evidence-based medicine). It is conceivable that temporal discounting may be a factor in situations in which older persons make poor decisions and fall prey to scam artists. Future studies are needed to clarify the public health implications of discounting in old age and may suggest policy implications for vulnerable older persons.
To date, most of the existing studies on cognition and discounting have examined indices of general cognitive ability; however, a few focused specifically on working memory because of its relatively strong association with overall intellectual function [
12,
13]. In one such study, increased working memory load during a discounting task was associated with greater overall discounting [
13]. Others have reported similar associations with working memory, and this relation has been interpreted as meaning that active attention to goal specific information and the integration of complex information are important determinants of discounting [
12]. In the present study, we found that perceptual speed and visuospatial abilities were related to discounting but working memory was not. The difference between study findings may in part reflect divergent approaches to measuring discounting across studies. Our questions were in some ways simpler than those used previously (
e.g., we included fewer and more simply worded questions and held the delay time constant for the larger payments); thus, the tasks used in earlier studies were more difficult and may have demanded more working memory than that employed in this study. Further, perceptual speed involves the ability to efficiently process information and make mental comparisons, and it is intuitive that these abilities are associated with the ability to advantageously decide between an immediate, smaller payment and a delayed, larger one. Moreover, perceptual speed, like working memory, is considered a component of the broader domain of executive function, which includes higher-level cognitive processing and abstract reasoning abilities that support the ability to wait for a delayed reward.
The finding that visuospatial abilities were associated with temporal discounting was unexpected; however, some data suggest that visuospatial deficits may precede the onset of dementia in old age and are associated with a deterioration in independent activities such as driving) [
36]. Somewhat akin to driving (albeit less complex), temporal discounting tasks require processing relational associations and estimating reward size and magnitude, cost benefit ratios, and timing. Thus, it is possible that persons with poorer visuospatial ability were less able to efficiently identify relevant patterns (
i.e., delayed payments are larger) and integrate multiple sources of information. However, additional research is needed to clarify the potential association of visuospatial abilities with discounting.
Strengths of this study include the comprehensive evaluation of cognition and temporal discounting in a fairly large cohort of community-based older persons. This enabled us to examine the association of global cognition and five specific cognitive abilities with temporal discounting among persons carefully evaluated and determined to be free of dementia. Limitations included the volunteer cohort, which may limit the generalizability of findings, and the single assessment of discounting using a series of hypothetical questions. Although a task involving actual monetary payment may be preferable, prior studies using hypothetical tasks like that used here have shown that the tendency to discount future rewards (or the inability to delay gratification) is associated with numerous psychological problems including gambling, substance abuse, depression and anxiety, as well as poorer academic, occupational and financial outcomes [
3,
4,
6,
8,
10,
22,
27,
37]. These data provide strong support for the ecological validity of measures such as that used here. Longitudinal study would enable us to examine the dynamic changes that may occur in both cognition and temporal discounting in aging. Future studies are needed to investigate whether and how the associations of cognition and age with temporal discounting change over time with advancing age and to clarify the association of discounting with decision-making and health outcomes in old age.
Author contributions
PAB, DAB, and DIL were involved in the study concept and design, acquisition of subjects and/or data, analysis and interpretation of data, and preparation of manuscript. LY, ES, RSW and ASB were involved in the analysis and interpretation and preparation of the manuscript. All authors read and approved the final manuscript.
Competing interests
The authors have no disclosures to report.