Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 16.
Published in final edited form as: J Pers. 2019 Dec 12;88(4):822–832. doi: 10.1111/jopy.12529

The Association between Actor/Partner Optimism and Cognitive Functioning among Older Couples

Jeewon Oh 1, William J Chopik 1, Eric S Kim 2
PMCID: PMC7564516  NIHMSID: NIHMS1633295  PMID: 31784979

Abstract

Objective

Higher optimism has been linked with health, well-being, and cognitive functioning. Spouses also play an important role on people’s health, especially in older adulthood. Yet, whether a spouse’s level of optimism is associated with an individual’s cognitive functioning is understudied. Thus, we examined this question.

Method

Participants were 4,457 heterosexual couples (N=8,914; Mage=66.73, SD=9.67) from the Health and Retirement Study—a large, diverse, prospective, and nationally representative sample of U.S. adults aged >50. Optimism was assessed at baseline (t1) and cognition was measured every two years with up to five repeated assessments of cognition data over the eight-year follow-up period (t1; t2; t3; t4; t5).

Results

Results from multi-level dyadic data analyses showed small but positive associations between actor optimism and actor cognitive functioning (memory: r = .16, mental status =.10), as well as partner optimism and actor cognitive functioning (memory: r = .04, mental status = .03). These associations mostly persisted over time.

Conclusions

Possessing higher optimism, and also having a partner with higher optimism, were both associated with higher cognitive functioning. Thus, with further research, optimism (at both the individual and couple level) might emerge as an innovative intervention target that helps older adults enjoy greater cognitive functioning as they age.

Keywords: optimism, cognitive functioning, actor-partner interdependence model, Health and Retirement Study

People vary in what they expect their futures to hold. Some expect the best while others expect the worst. Optimism characterizes the former and is defined as the generalized expectation that good things will happen in the future (Kim, Chopik, & Smith, 2014; Scheier & Carver, 1993). Although optimism is associated with a number of positive benefits, including cognitive functioning, it is unclear whether the optimism of a person’s spouse might also be associated with better cognitive functioning. The current study tests this possibility by investigating the effects of partner optimism on individual cognitive impairment in a large sample of older couples followed over time.

Benefits of optimism

Emerging research has observed that higher optimism is associated with reduced incidence of chronic illnesses (e.g., reduced risk of incident cardiovascular disease (Boehm & Kubzansky, 2012; Chopik, Kim, & Smith, 2015; Kim et al., 2014). However, if illness does occur, optimism is associated with a slower progression of disease (Ironson et al., 2005) and a reduced risk of mortality from several causes of death (i.e., heart disease, stroke, respiratory disease, infection). Although the mechanisms through which optimism influences physical health has not been fully characterized, there are at least three plausible underlying biobehavioral mechanisms, including: 1) promoting the acquisition and maintenance of healthy lifestyles (e.g., eating a balanced diet, exercising regularly), 2) directly affecting relevant neurobiological processes (e.g., lipid levels, autonomic functioning), and 3) promoting the acquisition and maintenance of a broader set of social and psychological assets that enhance a person’s ability to maintain health (Kubzansky et al., 2018).

For example, people with higher optimism may might perceive stressors as less stressful and/or emotionally recover from stressful events more rapidly. In one illustrative study that examined optimism in a group of people who were affected by a national disaster and another group of people who were non-affected, optimistic people in both groups reported fewer depressive- and obsessive-compulsive symptoms compared to their pessimistic counterparts (Van der Velden et al., 2007). Such findings might also be explained by how optimists tend to use healthier coping strategies when in stressful situations that require the recruitment of problem-solving strategies (e.g., optimists tend to use active coping strategies such as accepting reality, taking direct and planned action; Scheier & Carver, 1993). The well documented effects of social support on health is another potential pathway because, interpersonally, optimists enjoy typically enjoy greater relationship satisfaction and higher relationship quality over time (Assad, Donnellan, & Conger, 2007; Srivastava, McGonigal, Richards, Butler, & Gross, 2006).

Several of these hypothesized biobehavioral mechanisms can help explain the results from prior research that higher optimism is associated with better cognitive functioning, including measures derived from intelligence tests (e.g., verbal reasoning, arithmetic) and performance-based indicators of cognitive functioning (e.g., memory; e.g. Vance, Wadley, Ball, Roenker, & Rizzo, 2005). For example, higher optimism has been associated with reduced risk of cognitive impairment in older adulthood (Gawronski et al., 2016). Optimism also appears to buffer against the negative impact that traumatic brain injuries have on cognitive functioning (TBIs; Ramanathan, Wardecker, Slocomb, & Hillary, 2011). In combination, these studies suggest that being more optimistic is likely beneficial for individuals’ cognitive functioning, but an understudied related question is whether the optimism of one person is beneficially for the cognitive functioning of their spouse?

Potential benefits of partner optimism

Other people in a person’s social network—especially romantic partners—can have a large impact on a person’s health and cognition in late life (Antonucci & Akiyama, 1987; Hoppmann, Gerstorf, & Luszcz, 2008). For example, spouses often share cognitive ability trajectories over time (Hoppmann & Gerstorf, 2009), and husbands’ perceptual speed predicts wives’ perceptual speed over time (but interestingly, not the other way around; Gerstorf, Hoppmann, Anstey, & Luszcz, 2009).1 However, it remains unknown whether one’s psychological characteristics can “cross-over” and affect the cognitive functioning of their partner (Lee, Paddock, & Feeney, 2012).

At least three hypothesized mechanisms illustrate how one person’s optimism could influence their partner’s cognitive functioning. First, based on theories of Social Control (e.g., Social Control Theory), optimistic people might promote cognitive functioning in their partner by maintaining an atmosphere of healthy norms via informal social control. For example, higher optimism is associated with healthier behaviors, which predicts higher cognitive functioning, and optimistic people might reinforce norms in their partner that frequent physical activity is desirable (Lewis & Butterfield, 2007; Boehm et al., 2018). Second, evidence indicates that some couples coordinate their memories to make recall and mental structuring easier (i.e. shared remembering; Harris, Keil, Sutton, Barnier, & McIlwain, 2011). For example, couples that divide up responsibility and/or cue each other remember more information during a recall task. Further, when partners recall shared experiences/memories together (vs. individually), richer details about the memory tend to emerge (Harris, Barnier, Sutton, & Keil, 2014).

Since optimists are generally more likely to seek (and receive) social support when facing difficult situations, these types of memory-enhancing behaviors might be more likely (e.g. Andersson, 2012). Third, a person might enhance their cognitive functioning by learning and imitating their optimistic partners’ (cognitive) health-enhancing behaviors and thinking patterns. Optimists tend to use active coping strategies (vs. avoiding the problem) more frequently (Assad et al., 2007; Scheier & Carver, 1993) which may have implications for cognitive functioning. For instance, one lab experiment compared the effects of different emotion regulation strategies on cognitive functioning (e.g., recall ability for a conflict discussions). Couples that thought about positive aspects of their partner/relationship and the good times they shared together before entering a conflict discussion later remembered more details of the discussion than couples that suppressed emotions during a conflict discussion. On the other hand, those who suppressed emotions recalled less (Richards, Butler, & Gross, 2003). Likewise, if people learn how to use these types of strategies from their optimistic partner, they might reap downstream cognitive benefits (e.g., greater memory). In light of prior research showing how a person’s own level of higher optimism is associated with reduced risk of cognitive impairment, as well as potential mechanisms that can explain how one person’s optimism could influence the cognitive functioning of their partner, there reason to suggest that a dyadic association is possible—however, it has yet to be formally tested.

Using data from nearly 4,500 couples who were followed up to eight years, we hypothesized that people with higher levels of optimism would have higher levels of cognitive functioning. We also hypothesized that people who had partners with higher optimism would themselves experience higher levels of cognitive functioning over time.

Method

The Health and Retirement Study (HRS) is a nationally representative and prospective panel study that has surveyed more than 22,000 Americans aged 50+ and their partners (who may be younger) every two years (Sonnega et al., 2014). Starting in 2006, the HRS began collecting psychosocial constructs, including optimism, in a random 50% of the sample.

The other half of the sample filled out the same psychosocial assessments in 2008. The two HRS sub-cohorts were combined into one sample for the present analyses to increase statistical power and precision (Chopik et al., 2015). Thus, optimism was assessed at 2006/2008 (baseline).

Cognition assessments were administered to all study participants at baseline and every two years after, resulting in up to five repeated assessments of cognition data over the eight-year follow-up period (t1;2006/2008; t2;2008/2010; t3;2010/2012; t4;2012/2014; t5;2014).

In our analytic sample, we included all HRS couples who had data on optimism, the covariates, and at least one wave of cognition data (see Limitations section for further discussion).2 When considering power, our sample exceeded the suggested minimum of 782 couples that are required to estimate an effect of r = .10 at p = .05 and 80% power (Kenny, Kashy, & Cook, 2006).

The University of Michigan’s Institute for Social Research is responsible for the study and provides extensive documentation about the protocol, instrumentation, sampling strategy, and statistical weighting procedures.

Measures

Optimism.

Optimism was measured at baseline (t1; 2006/2008) using the Life Orientation Test-Revised (LOT-R). Studies have shown that the LOT-R has good reliability and validity (Scheier et al., 1994). A sample item is, “In uncertain times, I usually expect the best.” Participants are asked to rate the extent to which they agree with each item on a scale ranging from 1(strongly disagree) to 6(strongly agree). In total, six items were used to assess optimism (α=.76). The mean of all items was taken to create a scale, with higher scores reflected higher optimism.

Cognitive functioning.

Cognitive functioning was measured with six different indices (see Choi, Schoeni, Martin, & Langa, 2018; Crimmins, Kim, Langa, & Weir, 2011 for full details). 1) Ten-word immediate recall (range: 0–10 recalled words), 2) ten-word delayed recall (range: 0–10 recalled words), 3) a serial 7s subtraction test of working memory (range: 0–5 successful trials), 4) counting backwards to assess attention and processing speed (range: 0–2; 0=unsuccessful, 1=successful on second attempt, 2=successful on first attempt), 5) a summed recall measure for correctly identifying day of the week, date of the month, month of the year, and year were used as a battery of cognition measures (range: 0–4; 1 point for accurately answering each component), and 6) quality of their memory (“How would you rate your memory at the present time?”) on a 5-point scale (range: 1(poor) to 5(excellent)).

We conducted exploratory factor analysis on these six cognitive indices, and the results suggested that a two-factor solution best fit the data. The first factor contained both the immediate and delayed word recall measures, and the second factor contained the serial 7’s subtraction task, counting backwards, and day/date/month/year identification measures. The self-reported quality of memory was excluded from the analyses because it 1) loaded weakly on both factors and 2) might be an indicator of cognitive functioning that is biased by the optimism of the respondent (i.e., more optimistic individuals might artificially rate their memory as better than it objectively is according to the performance-based measures). The scores on each measure were standardized (mean = 0, SD = 1) and then averaged to create two factors of cognitive functioning: a memory factor (αs > .86) and a mental status/information processing factor (αs > .42; we termed this factor to be mental status because measures in this factor are often used to assess mental functioning in the case of possible head injury). Details of the factor analysis can be found in the Supplementary Tables 1 and 2.

Covariates.

Sex, age, and education were measured at baseline and used as covariates in our models. Sex was treated as a dichotomous variable. Participants reported their age in whole numbers. Participants reported the years of education they attained (range: 0 (none) to 17 (17+ years of education)).

We added neuroticism (assessed at baseline) as a covariate in a supplementary analysis because neuroticism is a potential confounder. Neuroticism was assessed on a scale of that arranged from 1(a lot) to 4(not at all) using the MIDI personality scales (Lachman & Weaver, 1997). Participants indicated how well each of following four words described them: moody, worrying, nervous and calm. The first three items were reverse-scored and then the mean of all items was computed (α =.70).

Statistical Analyses

To accurately account for the interdependence of individuals within dyads, we used multi-level modeling procedures (MLM; Kenny et al., 2006). This approach can estimate both actor effects (associations between a person’s optimism and his/her own cognitive functioning) and partner effects (associations between a person’s optimism and his/her partner’s cognitive functioning) while accounting for the statistical non-independence of members in a couple. We begin with background information about the analyses (e.g., when variables were assessed and how they were coded, and then explain our modeling decisions).

Our three main predictors were: 1) actor optimism, 2) partner optimism, 3) and time. First, we assessed actor optimism and partner optimism at the first wave (t1; 2006/2008) and treated both factors as time-invariant predictors of the two cognitive functioning factors. Second, to examine how our outcomes of main interest (i.e., two factors of cognitive functioning: memory and mental status) changed over time, we modeled the linear effect of time in each analysis. For this analysis, we coded the slope of cognitive functioning across waves in the following way: cohort 1 (−2), cohort 2 (−1), cohort 3 (0), cohort 4 (1), cohort 5 (2). Third, actor optimism and partner optimism were evaluated as potential moderators of change in cognitive functioning. Specifically, both actor and partner effects of optimism were tested as moderators of the slope (i.e., the models included actor optimism × time and partner optimism × time interaction terms). Additionally, we grand-mean centered all predictor variables (partner/actor optimism, age, education) and also contrast-coded sex (−1=men, 1=women).

In our main analyses, we conducted two multi-level models, one for each predictor. First, we examined memory as a function of actor optimism, partner optimism, time, and the interactions between actor/partner optimism and time. Second, we examined mental status as a function of actor optimism, partner optimism, time, and the interactions between actor/partner optimism and time. All models adjusted for sex, age, and education.

Additional Analyses

In a series of supplementary analyses, we evaluated the possibility of a multiplicative effect of actor optimism and partner optimism on cognitive functioning by creating and testing interaction terms between actor and partner optimism. For example, if an actor × partner interaction was significant, it could indicate that less optimistic individuals might benefit the most from having an optimistic partner. Alternatively, it could indicate that the effects of optimism on cognitive functioning were only present (or particularly evident) if both members were optimistic. A non-significant interaction would indicate that an optimism-cognitive functioning association does not vary depending on a partner’s level of optimism.3

Results

Participants were 4,457 heterosexual couples (N=8,914 individuals), ranging in age from 25 to 97 years old (Mage=66.73, SD=9.67; 50% were women; less than 1% of our sample were aged <50). 81% of participants identified as being: White, 8.6% Black, 8.1% Hispanic, or 2.3% other. Participants averaged 12.79 (SD = 3.06) years of education and were married for an average of 37.19 years (SD = 15.89). Means, standard deviations, and correlations for all variables are presented in Table 1.

Table 1.

Means, Standard Deviations (SD), and Correlations Between Study Variablesa,b,c

2006/2008 2008/2010 2010/2012 2012/2014 2014
1 2 3 4 5 6 7 8 9 10 11 12
2006/2008 1. Actor optimism
2. Partner optimism .282**
3. Memory .205** .108**
4. Mental status .135** .103** .351**
2008/2010 5. Memory .178** .084** .546** .241**
6. Mental status .123** .099** .319** .508** .326**
2010/2012 7. Memory .190** .102** .603** .258** .552** .282**
8. Mental status .142** .091** .333** .469** .302** .507** .383**
2012/2014 9. Memory .163** .068** .517** .212** .553** .253** .572** .293**
10. Mental status .133** .087** .314** .440** .293** .474** .356** .514** .374**
2014 11. Memory .153** .076** .576** .198** .497** .234** .618** .294** .569** .327**
12. Mental status .116** .088** .326** .419** .293** .464** .368** .488** .358** .548** .407**
M 4.49 4.49 .00 .01 .00 .01 .00 .01 .00 .01 .00 .01
SD 0.96 0.96 .94 .71 .94 .71 .94 .70 .95 .70 .95 .70
Open in a new tab
**

p < .001.

a

N = 3361–8821.

b

Memory = mean of standardized immediate recall and delayed recall scores.

c

Mental status = mean of standardized serial 7s, backward counting, and day/date/month/year naming tasks.

How are actor and partner optimism associated with cognitive functioning over time?

The results from the multi-level models are presented in Tables 2 (memory factor) and 3 (mental status factor). Cognitive functioning declined over the 8-year study period (rs > |.28|) in a model adjusting for age, education and sex. Higher levels of actor optimism were associated with better memory (r = .16) and mental status (r = .10). Higher levels of partner optimism were also associated with better memory (r = .04) and mental status (r = .03).

Table 2.

Actor and Partner Optimism on Memory (i.e., Immediate Recall and Delayed Recall)

95% CI
b SE df t p LBa UB rb
Intercept −.08 .01 4007.86 −10.47 <.001 −.09 −.06 −0.16
Agec −.03 .00 5182.48 −44.15 <.001 −.03 −.03 −0.52
Sexd .15 .01 4280.54 23.13 <.001 .14 .17 0.33
Educatione .08 .00 6541.37 34.43 <.001 .08 .09 0.39
Time −.07 .00 3403.58 −25.88 <.001 −.08 −.07 −0.41
Actor Optimism .11 .01 7476.10 14.36 <.001 .10 .13 0.16
Partner Optimism .02 .01 7381.67 3.17 <.001 .01 .04 0.04
Actor Optimism × Time <.001 .003 6006.01 .16 .87 −.01 .01 0.002
Partner Optimism × Time −.003 .003 5926.18 −.94 .35 −.01 .00 −0.01
Actor Optimism × Partner Optimism −.01 .01 4110.06 −1.96 .050 −.03 .00 −0.03
Open in a new tab
a

LB = lower bound, UB = upper bound for the 95% Confidence Intervals (CI)

b

r = partial correlations

c

Age, education, actor optimism, and partner optimism were grand mean centered.

d

Sex was coded as: men (−1), women (1).

e

Education was coded as: years of education.

Table 3.

Actor and Partner Optimism on Mental Status (i.e., Serial 7s, Backward Counting, Day/Date/Month/Year Naming Task).

95% CI
b SE df t p LBa UB rb
Intercept −.03 .01 3774.40 −5.34 <.001 −.05 −.02 −0.09
Agec −.01 .001 4941.60 −12.66 <.001 −.01 −.01 −0.18
Sexd −.01 .01 4065.10 −2.17 .03 −.02 .00 −0.03
Educatione .07 .002 6223.58 33.89 <.001 .07 .07 0.39
Time −.04 .002 3335.75 −17.03 <.001 −.05 −.04 −0.28
Actor Optimism .05 .01 6674.17 8.02 <.001 .04 .07 0.10
Partner Optimism .01 .01 6637.21 2.18 .03 .00 .03 0.03
Actor Optimism × Time .01 .003 5833.95 2.51 .01 .00 .01 0.03
Partner Optimism × Time −.002 .003 5746.16 −.96 .33 −.01 .00 −0.01
Actor Optimism × Partner Optimism −.01 .01 3995.89 −1.60 .11 −.02 .00 −0.03
Open in a new tab
a

LB = lower bound, UB = upper bound for the 95% Confidence Intervals (CI)

b

r = partial correlations

c

Age, education, actor optimism, and partner optimism were grand mean centered.

d

Sex was coded as: men (−1), women (1).

e

Education was coded as: years of education.

We tested actor optimism and partner optimism as potential moderators of changes in cognitive functioning over time. We observed that slopes were largely not moderated by actor and partner optimism. These results can also be interpreted as the effects of actor/partner optimism on cognitive functioning mostly persisting over time and not diminishing as time progressed.4 However, the slope of time was moderated by actor optimism in one instance—for mental status. Upon decomposing this interaction, we observed that the slope of time was largely similar between individuals high in optimism (b = −.04, SE = .003, p < .001, r = −.14) and low in optimism (b = −.05, SE = .004, p < .001, r = −.18), but people low in optimism declined at a slightly faster rate over time.

Do actor and partner optimism interact to predict cognitive functioning?

Our next question assessed whether there were any multiplicative effects between partner optimism and actor optimism when predicting cognitive functioning. We tested this possibility by including an actor optimism × partner optimism interaction term in both models. For memory, although the interaction term was significant (p = .050) in the main analysis, it was no longer significant in our supplementary analyses (e.g., controlling for neuroticism). Hence, we do not devote more space on this effect. For mental status, we did not observe a significant interaction. These results suggest that the effects of actor optimism on cognitive functioning are mostly independent of partner optimism (and vice versa).5

Are actor and partner optimism associated with cognitive functioning over and above actor neuroticism?

As neuroticism is a potential confounder, we evaluated the possibility that neuroticism might account for the effects that we examined (see Supplementary Tables 6 and 7 for the full results). In models that adjusted for neuroticism, actor optimism (rs > .07) and partner optimism (rs > .03) were still significant predictors of cognitive functioning. Neuroticism also predicted lower cognitive functioning for memory (r = −.05) and mental status (r = −.04) in these models. These results suggest that, despite optimism’s correlation with other personality factors like neuroticism, optimism likely has predictive validity above and beyond this other dimension of personality.

Discussion

We observed that higher actor optimism and higher partner optimism were both positively associated with better functioning on two measures of cognitive functioning: memory and mental status. This small but positive effect of optimism mostly persisted over time. Partner optimism did not moderate the effect of actor optimism. Taken together, these results suggest that having a partner high in optimism can be beneficial for one’s cognitive functioning in older adulthood, above and beyond the influence of other potential confounders such as own’s own level of optimism, age, education, and neuroticism. These results also highlight the importance of considering the context of spousal relationships when understanding late-life cognitive functioning.

Why is partner optimism associated with cognitive functioning in late-life?

Why might partner optimism be related to cognitive functioning after accounting for one’s own optimism? As Gerstorf & Hoppmann (2009) state, the study of spousal influence on cognition in late-life is an emerging field that requires further research and theory development, so little theorizing and evidence exists for cross-over patterns like the one that we evaluated in this study. However, in the Introduction, we described three potential pathways, including: 1) the promotion of healthy norms that promote cognitive functioning via informal social control, 2) a coordination of memory recall efforts in couples that facilitates recall, and 3) and learning healthy emotional regulation and coping strategies from the partner. Social relationships are yet another potential pathway. For example, optimists 1) are more likely to seek (and receive) social support when facing difficult situations, 2) are more liked, 3) have a larger network of friends, and 4) these friends provide more social support during stressful times (Andersson, 2012; Brissette, Scheier, & Carver, 2002; Carver, Kus, & Scheier, 1994; Karademas, 2006; Nes & Segerstrom, 2006; Segerstrom, 2001; Trunzo & Pinto, 2003). This additional social support may help spouses and relationships by building stronger relationships and social capital—antecedents and factors that help promote cognitive functioning later life (Chopik, 2017; Welker et al., 2014). Empirically testing these potential mechanisms between partner optimism and cognitive functioning is an exciting avenue of future research.

Limitations, Strengths, and Future Directions

This study had several limitations. First, many optimism effects in the current study were relatively small in magnitude in standardized units (i.e., partial correlations). Large sample sizes can capture smaller effects and estimate more precise effect sizes. However, they can often yield findings that are statistically significant but of little practical significance (Cohen, 1990). Nevertheless, it is important to quantify the size of the contribution that the psychological characteristics of partners have on cognitive functioning. For instance, it is interesting to observe that partner optimism (rs >.03) is associated with cognitive functioning to a similar degree as actor neuroticism (rs >. 04). When compared to other standardized effect sizes, these protective associations of actor and partner optimism are small and may seem less meaningful compared to the effects of time (r = .28–41) or age (r = 18-.52). However, considering unstandardized effect sizes helps place these effects in another perspective: when predicting mental status, a unit increase in partner optimism (b = .01) was equivalent to being a year younger (b = −.01). There might be additional factors that moderate the influence of optimism (or a partner’s optimism) on a person’s cognitive functioning over time (e.g., relationship quality, relationship duration, presence of other social ties). Future research should formally test these potential moderators.

Relatedly, there is likely variability in effect sizes of partner optimism when predicting the more proximal mechanisms that are closely related to cognitive functioning. For example, partner optimism might have a small direct effect on cognitive functioning but a larger indirect effect through physical health, adaptive coping styles, or social capital—all of which we hypothesized to be mediating predictors of the link between optimism and cognitive functioning. Future research can examine this possibility by including a broader array of cognitive and health-related mechanisms.

An important direction for future research is to better pinpoint the direction of effects between optimism and cognitive functioning, and also how these directional effects are modified in light of the influence that partners can have on a person’s health. For example, based on assortative mating, partners might initially select each other based on how similar their levels of optimism and cognitive functioning are. However, having repeated measures of both optimism and cognitive functioning over large stretches of time, and using innovative statistical techniques, can more accurately disentangle directionality issues in the context of close relationships (e.g., whether optimism predicts better cognitive functioning, or higher cognitive functioning predicts higher levels of optimism, or both; Schimmack & Lucas, 2010; Wortman & Lucas, 2016). There is likely a bi-directional effect, but further research is needed to examine the nature of the effects. Whether or not cognitive functioning of one person predicts variation in their own or their partner’s optimism is a fascinating topic for future research.

Participants with only one wave of cognitive functioning data had lower scores on nearly all study variables, including: individual optimism, partner optimism, cognitive functioning, and demographic characteristics (e.g., education).6 This may have restricted the range of many of these variables—attenuating the size of associations that we could observe. Thus, future research that examines more representative samples of older couples may be able to make stronger conclusions about the magnitude of the effects of spousal optimism on cognitive functioning.

This study also has several strengths, including the use of a large and diverse sample of older couples that were followed prospectively for eight years. Further, HRS is one of the few cohort studies with detailed information about optimism, potential confounders, and repeated measures of cognitive functioning. Such features enabled us to measure the primary independent variable of interest, optimism, with a psychometrically validated and widely used measure. Additionally, the prospective nature of the data helped mitigate concerns that the associations we observed were attributable to an association in the reverse direction or retrospective reporting bias.

Conclusion

Higher optimism, and having a partner with higher optimism, were both associated with higher cognitive functioning in older adulthood. These findings are consistent with accumuating evidence on the benefits of optimism and close relationships for health and well-being across the lifespan. In light of our rapidly aging population, identifying factors that enable us to age well, and with our cognitive health intact, is becoming an increasingly more urgent issue. Our study identified partner optimism as an additional psychosocial factor that might contribute to maximizing people’s cognitive health across the lifespan.

Supplementary Material

supplementary tables

Acknowledgments

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Preparation of this manuscript was supported by Award 2R03AG054705-01A1 from the National Institute of Health to William Chopik, and Grant K99AG055696 from the National Institute on Aging to Eric Kim. The Health and Retirement Study is sponsored by the National Institute on Aging (NIA U01AG09740) and is conducted by the University of Michigan. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders. The first author was not supported by any federal or institutional grant/award.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

1

Another factor to consider (beyond shared trajectories of cognitive functioning over time) is that spouses might be similar in cognitive functioning because of assortative mating (i.e., people pair with partners who have similar cognitive functioning). To evaluate this idea, we examined covariation between actor’s intercepts and partner’s slopes in cognitive functioning through the random effects. For both factors, the variability in the actors’ intercepts was not associated with the variability in their partners’ slopes (ps > .14). Yet, actors’ intercepts and partners’ intercepts were significantly associated for both factors (estimates > .03, p < .001), while actors’ slopes and partner’s slopes were significantly related only for mental status (estimate = .002, p = .003). Finally, actors’ intercepts and slopes (as well as partners’ intercepts and slopes) were significantly associated for both factors (estimates > .01, p < .004). However, such an examination in relation to cognitive functioning and optimism developing within and across partners is beyond the scope of the current paper and impossible given the current data.

2

Although our study was on older couples (and the mean age was 66.73, SD = 9.67), we did not have a specific age exclusion criterion and thus analyzed all available data for our main analysis. Given that some older adults marry individuals younger than age 50, there are a few younger participants in our study. However, we tried analyzing the data after including only couples in which both members were 50+. We did not end up using results from these analyses for three reasons. First, we ran into convergence issues for estimating changes in the memory factor using SEM. Second, results for both factors were largely identical to results from analyses that included all participants. Third, people younger than age 50 only comprised .07% of the sample, so we elected to report the results from the larger sample.

3

One alternative way of testing our question is to estimate a second-order dyadic growth curve model predicting intercepts and slopes in cognitive functioning from actor/partner optimism (see Peugh, DiLillo, & Panuzio, 2013). This approach has the added benefit of modeling and thus controlling for measurement error by estimating latent intercepts and slopes from the multiple assessments of the two cognitive functioning factors. We attempted to run these models at the recommendation of the Editor and Reviewers, but ran into some convergence issues. The models converged only after constraining the actor and partner slope covariance to 0, but model fit was poor; further, oversimplifying the covariance structure has been found to provide biased estimates of the standard errors of fixed effect estimates (Wang, Yang, & Liu, 2018). Therefore, we elected to focus on the results from the multi-level modeling procedure. Nevertheless, the results of the structural equation model approach are reported in the Supplementary Tables 3 and 4. Worth noting, the results presented in the supplement are similar as those presented in the main text here using multi-level modeling.

4

A common test in dyadic data analysis is to examine whether the models are “distinguishable” (i.e., moderated) by some individual characteristic that the partners within a couple vary on meaningfully. Oftentimes, the most common distinguishing variable is sex. We examined whether sex moderated any of the effects of interest in the current study. Most actor and partner optimism effects were not distinguishable by sex. The one exception was a three-way partner optimism × time × sex interaction. Decomposing this interaction revealed that the partner optimism × time effect was significant among women (b = −.01, SE = .004, t = −1.96, p = .05) but not men (b = .003, SE = .004, t = .84, p = .40). Because the effect for women was very close to p = .05, we hesitate to devote substantial space to discussing it. Decomposing this interaction revealed that the effect of partner optimism declined over time, such that having a husband high in optimism was associated with better mental status for their wife at time 1 but not time 5.

5

The supplementary materials present results of a model that operationalized cognitive functioning differently. This alternate measure of cognitive functioning was created by summing participants’ scores on immediate recall, delayed recall, serial 7s, and backward counting (range: 0–27). This alternative composite measure is a common way to assess cognitive functioning in the fields of epidemiology and medicine, and is based on prior research that has validated the score and showed that it predicts dementia diagnoses in 74% of participants (Crimmins et al., 2011). Like in our main results, results that used this alternate operationalization of cognitive functioning revealed that both actor optimism and partner optimism predicted better cognitive functioning.

6

We compared participants who contributed one wave of data to those who had at least two waves of data on cognitive measures. Those who contributed only one wave of data (8.5% of the sample) had lower optimism (d = .32), had partners with lower optimism (d = .15), were in longer marriages (d = .20), had fewer years of education (d = .34), and were older (d = .51). Participants with one wave of data also performed worse on every index of cognitive functioning in the current study: immediate word recall (d = .71), delayed word recall (d = .69), serial 7’s (d = .45), backwards counting (d = .37), and day/date/month/year naming (d = .75).

References

  1. Andersson M (2012). Dispositional Optimism and the Emergence of Social Network Diversity. Sociol Q, 53(1), 92–115. 10.1111/j.1533-8525.2011.01227.x [DOI] [Google Scholar]
  2. Antonucci TC, & Akiyama H (1987). Social networks in adult life and a preliminary examination of the convoy model. Journal of Gerontology, 42, 519–527. 10.1093/geronj/42.5.519. Document Type: journal. [DOI] [PubMed] [Google Scholar]
  3. Assad KK, Donnellan MB, & Conger RD (2007). Optimism: An enduring resource for romantic relationships. Journal of Personality and Social Psychology, 93(2), 285. [DOI] [PubMed] [Google Scholar]
  4. Boehm JK, Chen Y, Koga H, Mathur MB, Vie LL, & Kubzansky LD (2018). Is Optimism Associated With Healthier Cardiovascular-Related Behavior?: Meta-Analyses of 3 Health Behaviors. Circulation research, 122(8), 1119–1134. [DOI] [PubMed] [Google Scholar]
  5. Boehm JK, & Kubzansky LD (2012). The heart’s content: The association between positive psychological well-being and cardiovascular health. Psychological bulletin. 10.1037/a0027448 [DOI] [PubMed]
  6. Brissette I, Scheier MF, & Carver CS (2002). The role of optimism in social network development, coping, and psychological adjustment during a life transition. Journal of Personality and Social Psychology, 82(1), 102–111. 10.1037/0022-3514.82.1.102 [DOI] [PubMed] [Google Scholar]
  7. Carver CS, Kus LA, & Scheier MF (1994). Effects of good versus bad mood and optimistic versus pessimistic outlook on social acceptance versus rejection. Journal of Social and Clinical Psychology, 13(2), 138–151. 10.1521/jscp.1994.13.2.138 [DOI] [Google Scholar]
  8. Chang EC, & Sanna LJ (2001). Optimism, pessimism, and positive and negative affectivity in middle-aged adults: A test of a cognitive-affective model of psychological adjustment. Psychology and aging, 16(3), 524–531. 10.1037/0882-7974.16.3.524 [DOI] [PubMed] [Google Scholar]
  9. Choi H, Schoeni RF, Martin LG, & Langa KM (2018). Trends in the Prevalence and Disparity in Cognitive Limitations of Americans 55–69 Years Old. J Gerontol B Psychol Sci Soc Sci, 73(suppl_1), S29–s37. 10.1093/geronb/gbx155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chopik WJ (2017). Associations among relational values, support, health, and well-being across the adult lifespan. Personal Relationships, 24(2), 408–422. 10.1111/pere.12187 [DOI] [Google Scholar]
  11. Chopik WJ, Kim ES, & Smith J (2015). Changes in optimism are associated with changes in health among older adults. Social Psychological and Personality Science, 6, 814–822. 10.1177/1948550615590199 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chopik WJ, Kim ES, & Smith J (2018). An examination of dyadic changes in optimism and physical health over time. Health Psychology, 37, 42–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cohen J (1990). Things I have learned (so far). American Psychologist, 45(12), 1304–1312. 10.1037/0003-066X.45.12.1304 [DOI] [Google Scholar]
  14. Conversano C, Rotondo A, Lensi E, Della Vista O, Arpone F, & Reda MA (2010). Optimism and its impact on mental and physical well-being. Clinical Practice and Epidemiology in Mental Health, 6, 25–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Crimmins EM, Kim JK, Langa KM, & Weir DR (2011). Assessment of cognition using surveys and neuropsychological assessment: the Health and Retirement Study and the Aging, Demographics, and Memory Study. J Gerontol B Psychol Sci Soc Sci, 66 Suppl 1, i162–171. 10.1093/geronb/gbr048 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ganguli M, Fu B, Snitz BE, Hughes TF, & Chang C-CH (2013). Mild cognitive impairment Incidence and vascular risk factors in a population-based cohort. Neurology, 10.1212/WNL.1210b1013e318295d318776. [DOI] [PMC free article] [PubMed]
  17. Gawronski KAB, Kim ES, Langa KM, & Kubzansky LD (2016). Dispositional Optimism and Incidence of Cognitive Impairment in Older Adults. Psychosom Med, 78(7), 819–828. 10.1097/PSY.0000000000000345 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gerstorf D, Hoppmann CA, Anstey KJ, & Luszcz MA (2009). Dynamic links of cognitive functioning among married couples: Longitudinal evidence from the Australian Longitudinal Study of Ageing. Psychology and aging, 24(2), 296–309. 10.1037/a0015069 [DOI] [PubMed] [Google Scholar]
  19. Harris CB, Barnier AJ, Sutton J, & Keil PG (2014). Couples as socially distributed cognitive systems: Remembering in everyday social and material contexts. Memory Studies, 7(3), 285–297. [Google Scholar]
  20. Harris CB, Keil PG, Sutton J, Barnier AJ, McIlwain DJF (2011). We remember, we forget: Collaborative remembering in older couples. Discourse Processes, 48(4), 267–303. [Google Scholar]
  21. Hart SL, Vella L, & Mohr DC (2008). Relationships among depressive symptoms, benefit-finding, optimism, and positive affect in multiple sclerosis patients after psychotherapy for depression. Health Psychology, 27(2), 230–238. 10.1037/0278-6133.27.2.230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hoppmann CA, & Gerstorf D (2009). Spousal interrelations in old age—A mini-review. Gerontology, 55(4), 449–459. 10.1159/000211948 [DOI] [PubMed] [Google Scholar]
  23. Hoppmann CA, Gerstorf D, & Luszcz M (2008). Spousal social activity trajectories in the Australian Longitudinal Study of Ageing in the context of cognitive, physical, and affective resources. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 63B(1), P41–P50. 10.1093/geronb/63.1.P41 [DOI] [PubMed] [Google Scholar]
  24. Ikeda A, Schwartz J, Peters JL, Fang S, Spiro A 3rd, Sparrow D, … Kubzansky LD (2011). Optimism in relation to inflammation and endothelial dysfunction in older men: the VA Normative Aging Study. Psychosom Med, 73(8), 664–671. 10.1097/PSY.0b013e3182312497 [DOI] [PubMed] [Google Scholar]
  25. Ironson G, Balbin E, Stuetzle R, Fletcher MA, O’Cleirigh C, Laurenceau J, … Solomon G (2005). Dispositional optimism and the mechanisms by which it predicts slower disease progression in HIV: proactive behavior, avoidant coping, and depression. International journal of behavioral medicine, 12(2), 86–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Karademas EC (2006). Self-efficacy, social support and well-being. The mediating role of optimism. Personality and Individual Differences, 40(6), 1281–1290. 10.1016/j.paid.2005.10.019 [DOI] [Google Scholar]
  27. Kenny DA, Kashy DA, & Cook WL (2006). Dyadic data analysis. New York: Guilford Press. [Google Scholar]
  28. Kim ES, Chopik WJ, & Smith J (2014). Are people healthier if their partners are more optimistic? The dyadic effect of optimism on health among older adults. Journal of Psychosomatic Research, 76(6), 447–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kim ES, Hagan KA, Grodstein F, DeMeo DL, De Vivo I, & Kubzansky LD (2017). Optimism and cause-specific mortality: a prospective cohort study. American journal of epidemiology, 185(1), 21–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kubzansky LD, Huffman JC, Boehm JK, Hernandez R, Kim ES, Koga HK, … Labarthe DR (2018). Positive psychological well-being and cardiovascular disease: JACC health promotion series. Journal of the American College of Cardiology, 72(12), 1382–1396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lee J, Paddock SM, & Feeney K (2012). Emotional distress and cognitive functioning of older couples: A dyadic analysis. Journal of aging and health, 24(1), 113–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lewis MA, & Butterfield RM (2007). Social Control in Marital Relationships: Effect of One’s Partner on Health Behaviors 1. Journal of Applied Social Psychology, 37(2), 298–319. [Google Scholar]
  33. Nes LS, & Segerstrom SC (2006). Dispositional optimism and coping: A meta-analytic review. Personality and social psychology review, 10(3), 235–251. 10.1207/s15327957pspr1003_3 [DOI] [PubMed] [Google Scholar]
  34. Peterson C, Park N, & Kim ES (2012). Can optimism decrease risk of illness and disease among the elderly? Aging Health, 8, 5–8. [Google Scholar]
  35. Plassman BL, Langa KM, Fisher GG, Heeringa SG, Weir DR, Ofstedal MB, … Wallace RB (2007). Prevalence of dementia in the United States: the aging, demographics, and memory study. Neuroepidemiology, 29(1–2), 125–132. 10.1159/000109998 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Ramanathan DM, Wardecker BM, Slocomb JE, & Hillary FG (2011). Dispositional optimism and outcome following traumatic brain injury. Brain Injury, 25(4), 328–337. [DOI] [PubMed] [Google Scholar]
  37. Richards JM, Butler EA, & Gross JJ (2003). Emotion regulation in romantic relationships: The cognitive consequences of concealing feelings. Journal of Social and Personal Relationships, 20(5), 599–620. [Google Scholar]
  38. Roberts BW, Smith J, Jackson JJ, & Edmonds G (2009). Compensatory conscientiousness and health in older couples. Psychological Science, 20(5), 553–559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Scheier MF, & Carver CS (1993). On the Power of Positive Thinking: The Benefits of Being Optimistic. Current Directions in Psychological Science, 2(1), 26–30. 10.1111/1467-8721.ep10770572 [DOI] [Google Scholar]
  40. Scheier MF, Carver CS, & Bridges MW (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67(6), 1063–1078. 10.1037/0022-3514.67.6.1063 [DOI] [PubMed] [Google Scholar]
  41. Schimmack U, & Lucas RE (2010). Environmental influences on well-being: A dyadic latent panel analysis of spousal similarity. Social Indicators Research, 98(1), 1–21. [Google Scholar]
  42. Segerstrom SC (2001). Optimism, goal conflict, and stressor-related immune change. Journal of Behavioral Medicine, 24(5), 441–467. 10.1023/a:1012271410485 [DOI] [PubMed] [Google Scholar]
  43. Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JWR, & Weir DR (2014). Cohort profile: The Health and Retirement Study (HRS). International Journal of Epidemiology, 43(2), 576–585. 10.1093/ije/dyu067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Srivastava S, McGonigal KM, Richards JM, Butler EA, & Gross JJ (2006). Optimism in close relationships: How seeing things in a positive light makes them so. Journal of Personality and Social Psychology, 91(1), 143–153. 10.1037/0022-3514.91.1.143 [DOI] [PubMed] [Google Scholar]
  45. Taylor AM, Ritchie SJ, & Deary IJ (2017). Associations of intelligence across the life course with optimism and pessimism in older age. Intelligence, 62, 79–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Trunzo JJ, & Pinto BM (2003). Social support as a mediator of optimism and distress in breast cancer survivors. Journal of Consulting and Clinical Psychology, 71(4), 805–811. 10.1037/0022-006x.71.4.805 [DOI] [PubMed] [Google Scholar]
  47. Vance DE, Wadley VG, Ball KK, Roenker DL, & Rizzo M (2005). The effects of physical activity and sedentary behavior on cognitive health in older adults. Journal of aging and physical activity, 13(3), 294–313. [DOI] [PubMed] [Google Scholar]
  48. Van der Velden PG, Kleber RJ, Fournier M, Grievink L, Drogendijk A, & Gersons BP (2007). The association between dispositional optimism and mental health problems among disaster victims and a comparison group: A prospective study. Journal of affective disorders, 102, 35–45. 10.1016/j.jad.2006.12.004 [DOI] [PubMed] [Google Scholar]
  49. Welker KM, Baker L, Padilla A, Holmes H, Aron A, & Slatcher RB (2014). Effects of self-disclosure and responsiveness between couples on passionate love within couples. Personal Relationships, 21(4), 692–708. 10.1111/pere.12058 [DOI] [Google Scholar]
  50. Wortman J, & Lucas RE (2016). Spousal similarity in life satisfaction before and after divorce. Journal of Personality and Social Psychology, 110(4), 625–633. 10.1037/pspp0000065 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplementary tables
NIHMS1633295-supplement-supplementary_tables.docx (33.8KB, docx)

RESOURCES