Journal of Experimental Psychology: General

A Route to Well-Being: Intelligence Versus Wise

Reasoning
Igor Grossmann, Jinkyung Na, Michael E. W. Varnum, Shinobu Kitayama, and Richard E.
Nisbett
Online First Publication, August 6, 2012. doi: 10.1037/a0029560

CITATION
Grossmann, I., Na, J., Varnum, M. E. W., Kitayama, S., & Nisbett, R. E. (2012, August 6). A
Route to Well-Being: Intelligence Versus Wise Reasoning. Journal of Experimental
Psychology: General. Advance online publication. doi: 10.1037/a0029560
Journal of Experimental Psychology: General © 2012 American Psychological Association
2012, Vol. 141, No. 4, 000 0096-3445/12/$12.00 DOI: 10.1037/a0029560

A Route to Well-Being: Intelligence Versus Wise Reasoning

Igor Grossmann Jinkyung Na

University of Waterloo University of Texas at Dallas

Michael E. W. Varnum Shinobu Kitayama and Richard E. Nisbett

Peking University University of Michigan

Laypeople and many social scientists assume that superior reasoning abilities lead to greater well-being.
However, previous research has been inconclusive. This may be because prior investigators used
operationalizations of reasoning that favored analytic as opposed to wise thinking. We assessed wisdom
in terms of the degree to which people use various pragmatic schemas to deal with social conflicts. With
a random sample of Americans, we found that wise reasoning is associated with greater life satisfaction,
less negative affect, better social relationships, less depressive rumination, more positive versus negative
words used in speech, and greater longevity. The relationship between wise reasoning and well-being
held even when controlling for socioeconomic factors, verbal abilities, and several personality traits. As
in prior work, there was no association between intelligence and well-being. Further, wise reasoning
mediated age-related differences in well-being, particularly among middle-aged and older adults. Impli-
cations for research on reasoning, well-being, and aging are discussed.

Keywords: aging, intelligence, reasoning, well-being, wisdom

The happiness of your life depends upon the quality of your thoughts. greater well-being than their younger counterparts (e.g.,
—Marcus Antonius Carstensen, Pasupathi, Mayr, & Nesselroade, 2000; Mroczek &
Kolarz, 1998). Moreover, standard intelligence tests do not do a
Scholars since at least the time of Aristotle have speculated that
good job of capturing people’s ability to think about social rela-
superior reasoning leads to greater well-being. This is consistent
tions (e.g., Sternberg, 1999) or real-world decision making (e.g.,
with laypeople’s intuitions and beliefs about themselves—those
Stanovich, 2009).
people who report greater well-being believe that they have supe-
rior reasoning abilities (Campbell, Converse, & Rogers, 1976, We propose that superior reasoning may, in fact, be related to
Study 3; Diener & Fujita, 1995). However, various large-scale well-being but that this is true for pragmatic (as opposed to
studies have shown no relationship between standard measures of abstract) reasoning. By pragmatic reasoning, we mean reasoning
intelligence and well-being (e.g., Sigelman, 1981; Watten, Sy- that is influenced by life experiences and situated in a social
versen, & Myhrer, 1995; Wirthwein & Rost, 2011). Furthermore, context. Such reasoning strategies have been described as wise by
abstract reasoning abilities and other types of fluid intelligence both philosophers and psychologists. Although wisdom has been
decline over adulthood (Salthouse, 2004), yet older adults report defined in many ways (Sternberg & Jordan, 2005), there is some
consensus that wisdom involves the use of certain types of prag-
matic reasoning that are prosocial and that help to navigate im-
portant challenges in social life. For instance, Baltes and col-
leagues (e.g., Baltes & Smith, 2008; Baltes & Staudinger, 2000)—
Igor Grossmann, Department of Psychology, University of Waterloo, who developed the Berlin wisdom paradigm— have defined
Waterloo, Ontario, Canada; Jinkyung Na, Center for Vital Longevity, wisdom as knowledge useful for dealing with life problems, in-
University of Texas at Dallas; Michael E. W. Varnum, Department of
cluding an awareness of the varied contexts of life and how they
Psychology, Peking University, Beijing, People’s Republic of China; Shi-
nobu Kitayama and Richard E. Nisbett, Department of Psychology, Uni- change over time, recognition that values and goals differ among
versity of Michigan. people, and acknowledgment of the uncertainties of life (together
The work presented in this article was completed as part of Igor Gross- with ways to manage these uncertainties). Similarly, Basseches
mann’s dissertation research. It supported by the International Max Plank (1980) and Kramer (1983)—representing the neo-Piagetian view
Research LIFE Fellowship and the Rackham Predoctoral Dissertation of reasoning—formulated a set of cognitive schemas they believed
Fellowship awarded to Igor Grossmann and National Institute on Aging to be involved in wise thinking, including acknowledgment of
Grant 5RO129509-02 awarded to Richard E. Nisbett and Shinobu Ki-
others’ points of view, appreciation of contexts broader than the
tayama. We thank Robert J. Sternberg and Jacqui Smith for thoughtful
comments on a previous version of the article.
issue at hand, sensitivity to the possibility of change in social
Correspondence concerning this article should be addressed to Igor relations, acknowledgment of the likelihood of multiple outcomes
Grossmann, Department of Psychology, University of Waterloo, Waterloo, of a conflict, concern with conflict resolution, and preference for
Ontario N2L 3G1, Canada. E-mail: [email protected] compromise in resolving opposing viewpoints.

1
2 GROSSMANN, NA, VARNUM, KITAYAMA, AND NISBETT

Little work has directly tested the relationship between wise Overview of the Present Research
reasoning and well-being. The only two studies we are aware of
that have examined this question used the Berlin wisdom para- To address the relationship between wise reasoning and well-
digm. These studies found inconclusive results. In one study, wise being, we measured wise reasoning about real-world intergroup
reasoning was unrelated to negative affect,1 but it was weakly and interpersonal dilemmas and various indicators of well-being.
positively related to some aspects of positive affect (e.g., feeling To maximize the possibility of detecting conditions under which
interested or inspired; Kunzmann & Baltes, 2003). In another wise reasoning may be related to well-being, we tested a diverse
study, wise reasoning was unrelated to people’s positive or nega- set of socioemotional tasks. In addition to exploring the relation-
tive emotional responses but negatively related to global judgment ship between wise reasoning and well-being, we also addressed the
of life satisfaction (particularly among the top 15% on wise rea- question of how aging impacts this relationship. Specifically, we
soning; Mickler & Staudinger, 2008). It is important to consider predicted that older adults would show greater well-being, partly
that the materials in these studies included rather abstract de- because they are wiser than younger adults when reasoning about
scriptions of personal problems. For instance, participants were social conflicts (Grossmann et al., 2010; Studies 1–2; Worthy,
asked to read and respond to such scenarios as “A 14-year-old girl Gorlick, Pacheco, Schnyer, & Maddox, 2011).
wants to move away from home right away” or “Somebody gets a
phone call from a good friend. The friend says that she or he Method
cannot go on anymore and that she or he has decided to commit
suicide” (Baltes & Staudinger, 2000). Such briefly described sce- Participants
narios provided little information about social context, which may
be a critical factor in the assessment of wise reasoning (Sternberg, We recruited a stratified random sample of 241 Americans in
2004). Thus, it remained unclear if wisdom-related forms of rea- Washtenaw County, Michigan, with an approximately equal num-
soning are linked with well-being. With the present work, we ber of participants of both genders and of each of three age groups
aimed to fill this gap in the literature by using a novel measure of (25– 40 years, 41–59 years, 60 –90 years) and an adequate number
wise reasoning to systematically investigate its relationship to a of adults from lower educated strata (see Table 1). Participants
large number of well-being indicators. were informed that we were interested in human reasoning and
We built on the idea that people acquire wisdom through expe- they were compensated with $70 for taking part in each of the two
rience and through successful mastery of various challenging life 2-hr individual experimental sessions.
experiences (Pascual-Leone, 1990; Rowley & Slack, 2009; Stern-
berg, 1998). Such experiences are heterogeneous in nature and Procedure and Materials
may result in idiosyncratic ways of thinking about conflict. There-
fore, and consistent with the process-oriented view of wisdom Wise reasoning. Part of the data on wise reasoning came from
(Kramer, 2000; Sternberg, 1998), we conceptualized wisdom as a the Michigan Wisdom Study (Grossmann et al., 2010). The qual-
set of reasoning strategies that may be applicable and beneficial itative part of the study was conducted in a face-to-face interview
across a large number of social conflicts. In other words, we setting of the Robert B. Zajonc Experimental Laboratories at the
defined wisdom not through the availability of static knowledge Institute for Social Research, Ann Arbor, Michigan. Participants
about a particular conflict and its solution but rather through the were informed that we were interested in people’s reasoning about
use of dynamic reasoning strategies that can be applied in various various future events. Care was taken to ensure that the setting did
domains. Building on earlier theoretical work conceptualizing not induce fear of cognitive ability testing and possible ageism-
wisdom as an inherently social construct (Rowley, 2006; Stern- related stereotype threat (Hess & Blanchard-Fields, 1999). Specif-
berg, 2007), we measured wisdom using naturalistic, context-rich ically, the test setting was designed to provide a social atmosphere
materials concerning social conflicts and by examining reasoning (e.g., use of a meeting room instead of a laboratory cubicle, neutral
in a structured interview with a researcher. art pictures on the walls of the room), and the wisdom-related
We measured six broad strategies of wise reasoning in our stimulus materials were formatted as regular newspaper articles.
content analyses of participants’ responses to social dilemmas Further, the interview was conducted using a semistructured pro-
(Baltes & Staudinger, 2000; Basseches, 1980; Kramer, 1983; cedure and involved additional probes to each question. In Session
Staudinger & Glück, 2011). These components were (a) consider- 1, participants read three newspaper articles describing intergroup
ing the perspectives of people involved in the conflict, (b) recog- tensions (over ethnic differences, politics, and natural resources) in
nizing the likelihood of change, (c) recognizing multiple ways in unfamiliar countries (Grossmann et al., in press; Grossmann et al.,
which the conflict might unfold, (d) recognizing uncertainty and 2010). After each story, the interviewer instructed participants to
the limits of knowledge, (e) recognizing the importance of or talk about the future development of the conflict, guided by three
searching for a compromise between opposing viewpoints, and (f) questions in the following order: “What do you think will happen
recognizing the importance of or predicting conflict resolution. after that?” “Anything else?” and “Why do you think it will
The validity of these dimensions as measures of wise reasoning happen this way?” Participants’ responses were audiorecorded. In
has been demonstrated in a recent study, which surveyed a large Session 2, participants read three letters describing interpersonal
pool of wisdom researchers and counseling practitioners. The dilemmas (between friends, spouses, and neighbors) selected from
results from this study indicated that wisdom researchers and
practitioners rated responses that are high on these dimensions as 1
This was particularly true in analyses with age as a covariate. Zero-
wiser than those that were low on these dimensions (Grossmann et order correlations in this study showed a negative relationship between
al., 2010, Study 3). wise reasoning and both forms of affect (positive or negative).
 REASONING AND WELL-BEING 3

Table 1 coders were trained on sample materials until they reached high
Sample Characteristics interrater reliability. These coders then scored transcripts of each
response on the six aspects of wise reasoning: (a) considering the
Demographic perspective of the parties involved, (b) recognizing the likelihood
category % or M (SD)
of change, (c) recognizing multiple possibilities regarding how a
Gender 50.9% women conflict might unfold, (d) recognizing limits of one’s own knowl-
Age in years 49.48 (16.65) edge and acknowledging uncertainty, (e) searching for compro-
Age group
mise, and (f) predicting conflict resolution. Raters coded partici-
25–40 years 38.2%
41–59 years 28.9% pants’ transcripts on a scale from 1 ⫽ not at all to 3 ⫽ a great deal;
60–90 years 32.9% see Table 2 for example responses. To increase the external va-
African American 12.3% lidity of these ratings, we used separate groups of raters for
Asian American 3.5% intergroup and interpersonal scenarios. Agreement between two
European American 80.26%
Latino 3.9% respective raters was good (interrater rs ⱖ .85; coder discrepancies
Education were resolved in group discussion between the coders and Igor
High school or less 11.30% Grossmann). Considering the diverse nature of the scenarios,
Some college 29.40% scores across various scenarios showed acceptable reliability as
College 33.90%
indicated by internal consistency and intraclass correlation mea-
Postsecondary 25.30%
Family income/year sures (for Session 1, Cronbach’s ␣ ⫽ .61; for Session 2, Cron-
Under $40,000 29.6% bach’s ␣ ⫽ .77; across both sessions, Cronbach’s ␣ ⫽ .78).
$40,000–$60,000 19.7%
$60,000–$80,000 20.2%
Over $80,000 30.5%
Well-Being and Longevity
Participants completed a set of well-being indicators and asso-
ciated measures of emotion regulation tendencies.
the Dear Abby advice column. Participants then answered ques- Positive versus negative affect. Participants were asked to
tions similar to those used in Session 1. These tasks lasted 30 min recall 10 situations (Kitayama, Park, Sevincer, Karasawa, &
per session on average. Uskul, 2009). Some of these episodes involved social relations
Coding procedure. Participants’ transcripts were masked, (e.g., having a positive interaction with friends), some were
and age-related information was removed. Two hypothesis-blind related to study and work (e.g., being overloaded with work),

Table 2
Example Responses for the Immigration Story, Indicating Low and High Wise Reasoning

Low High

Search for a compromise

I’m sure that each, each culture will keep their original customs. They might want to let them continue with their ways and maybe at the same
It’s not likely that someone that’s lived a certain way is going time maybe try to do some kind of promotion to encourage them to better
to change just because they moved to a new area. . . . People assimilate into the culture though, not throw away their own culture, but to
are pretty true to their nature and they’re not really big on try to make the country more unified, maybe bring customs together that
change so I’m sure that it won’t be an easy thing for them to might be similar for both cultures, in order to unify the country.
change their culture.

Recognizing multiple ways the conflict might unfold

Most likely there is going to be very similar things as going on It’s hard to predict in a short period of time if the Kyrgyz will be assimilated
in the United States . . . the economic drivers are going to with the Tajiks or whether there will begin to be unrest and civil disunity
want to keep the immigration going and more traditionalists because of this immigration. . . . I don’t think it will necessary happen
are going to want to stem it and make laws like only either way. You have several possibilities. Either taking historically into
speaking Tajik, instead of both. . . . It seems like that’s account what has happened: One you will have assimilation and happiness
happened throughout the world history when you get a large or two you will have constant bickering and at least social disunity and
number of immigration coming in. social warfare.

Considering perspectives of people involved in the conflict

I think eventually it’ll be like what’s happening in the U.S. is I think there’ll be friction between those two ideas. People do assimilate
that all the different cultures will merge. If it’s really eventually but it often takes a couple generations to do that. . . . There’ll be
productive for the country I think the cultures will merge and influences both ways but people who are in particular countries that
there’ll be more peace within them. . . . I think that they’ll let receive immigrants, they always see it from their point of view, namely
them come in, because they need the labor force. . . . There’s that these immigrants are changing the country. They don’t necessarily see
going to be some conflict of social interests and their it from the other point of view. Also, immigrants might be upset because
cultures. . . . There’s going to be a little bit of culture their children are not the way they would be if they were back in their
difference. It sounds like that’s the main issue here, it’s just homeland.
the culture. I think usually that ends up working out.
Note. The examples were adapted from Grossmann et al., 2010.
4 GROSSMANN, NA, VARNUM, KITAYAMA, AND NISBETT

and some others concerned daily hassles and bodily conditions wise reasoning relates to longevity 5 years after completion of the
of the self (e.g., being caught in a traffic jam). Participants were first session of our study by examining publicly available death
asked to remember the latest occasion when each of the 10 records.
situations happened to them. They were asked to report the Emotional discourse. In an attempt to supplement the main
extent to which they experienced a series of emotions in these well-being measures reported in this study with implicit measures
situations. The list of emotion terms contained seven positive that are less susceptible to demand effects, we explored emotional
(e.g., feeling of closeness, pride, elated, happy) and five nega- discourse patterns. We reanalyzed participants’ narratives about
tive (e.g., ashamed, frustrated, angry, unhappy) emotions. Six- the social conflicts using the Linguistic Inquiry and Word Count
point scales that ranged from 1 ⫽ not at all to 6 ⫽ very strongly program (Pennebaker, Francis, & Booth, 2001). This program
were used in rating emotional experience. To minimize possible analyzes texts on a probabilistic basis by comparing files on a
age-related positivity effect in memory (Carstensen & Mikels, word-by-word basis with a dictionary of 2,290 words and word
2005), we followed the recommendations by Kahneman and stems that are organized into several different language categories.
colleagues (Kahneman, Krueger, Schkade, Schwarz, & Stone, The analysis computes the percentage of total words found in the
2004) and performed our analyses on the episodes from the text that belong to these language categories. We took the relative
preceding 2 days.2 We averaged the scores for positive (Cron- percentage of positive to negative affect words contained in the
bach’s ␣ ⫽ .88) versus negative emotions (Cronbach’s ␣ ⫽ narratives as an index of positive versus negative thought acces-
.68). sibility. We focused on this index, because previous research
Relationship satisfaction. We asked participants to consider indicated that counting positive versus negative words in verbal
their personal network including people “who are important in and written discourse about oneself and others is positively asso-
your life right now.” Participants reported the initials of such ciated with greater self-report positive affect (Pennebaker, Mehl,
people in three concentric circles, ranging from “people to whom & Niederhoffer, 2003) and lower neuroticism and greater agree-
you feel so close that it is difficult to imagine life without them” ableness (Pennebaker & King, 1999) and predicts longevity (Dan-
(innermost circle) to “people who are close enough and important ner, Snowdon, & Friesen, 2001).
enough in your life that they should be placed in your personal
network” (outer circle). Network members included relatives
Covariate Measures: Speed of Processing, Cognitive
(44%, M ⫽ 12.74, SD ⫽ 8.71), nonkin friends (42%, M ⫽ 14.86,
SD ⫽ 14.07), and acquaintances (14%, M ⫽ 5.11, SD ⫽ 6.20).
Abilities, and Personality
Next, participants indicated initials of those network members who Given the novelty of our measure of wise reasoning, we also
“have given you advice and social support during the last month.” tested its relationships to cognitive abilities and personality traits.
We counted the number of such individuals as an index of positive Participants completed two tests of processing speed (Hedden et
relations (M ⫽ 5.39, SD ⫽ 4.50). Finally, we asked them to al., 2002) and the digit span subtest of the Wechsler Adult Intel-
indicate initials of those member who “have caused annoyances ligence Scale (WAIS; rforward-backward ⫽ .49, p ⬍ .001). Speed-
and troubles during the last month,” which we counted as an index of-processing scores were significantly correlated (r ⫽ .69, p ⬍
of negative relations (M ⫽ 1.85, SD ⫽ 2.11). Following previous .001) and thus were standardized and collapsed into a single index.
research (Fiori, Antonucci, & Akiyama, 2008), we obtained the Participants also completed comprehension and vocabulary sub-
index of relationship satisfaction by examining the relative pro- tests of WAIS (r ⫽ .52, p ⬍ .001).
portion of distinctly positive versus negative relationships. To The Big Five personality dimensions were measured using the
normalize the skewed distribution, we log-transformed the result- 10-item personality measure, which has shown high convergent
ing scores. validity with other Big Five personality measures in past research
Rumination. Participants next completed the five-item (Gosling, Rentfrow, & Swann, 2003) and the highest convergence
Brooding subscale of the Ruminative Response Scale, which ex- with the 60-item NEO Five-Factor Inventory in comparison to
amines an emotion regulation tendency to respond to distress by other short Big Five measures (Furnham, 2008). The two items per
repeatedly reflecting on past negative experiences (Treynor, Gon- each dimension were significantly correlated (rExtraversion ⫽ .53,
zalez, & Nolen-Hoeksema, 2003). For instance, one of the ques- p ⬍ .001; rAgreeableness ⫽ .19, p ⬍ .05; rConscientiousness ⫽ .37, p ⬍
tions asked participants to indicate how often they think “What am .001; rNeuroticism ⫽ .48, p ⬍ .001; rOpenness ⫽ .40, p ⬍ .001).
I doing to deserve this?” from 1 ⫽ almost never to 4 ⫽ almost Following an established procedure (Gosling et al., 2003), we
always. In the present study, this scale has showed acceptable collapsed the pairs of scores into a single index for each of the five
reliability, which was comparable to previous research (Cron- personality dimensions.
bach’s ␣ ⫽ .69).
Life satisfaction. Participants further answered a life satisfac-
tion question (a common technique in well-being research; Kah- Control Variables: Perceived Health and Social Class
neman, Diener, & Schwarz, 1999)—“All things considered, how Perceived health was measured with a three-item health ques-
satisfied are you with your life as a whole these days?”— on a tionnaire (e.g., “Compared to other people your own age, how
10-point scale ranging from 1 ⫽ not at all to 10 ⫽ very much
satisfied. This measure was administered via a survey distributed
2
a year after completion of the experimental sessions, resulting in a Participants reported at least one positive and one negative episode
from the past 2 days. The main results looked very similar when examining
smaller sample size (N ⫽ 141). all episodes; for example, wise reasoning was not related to positive affect
Longevity. Finally, because subjective well-being is a strong (r ⫽ ⫺.02), but wise reasoning was significantly negatively related to
predictor of longevity (Chida & Steptoe, 2008), we explored how negative affect (r ⫽ ⫺.16, p ⫽ .02).
 REASONING AND WELL-BEING 5

would you rate your physical health?”; Hedden et al., 2002; Table 4
Cronbach’s ␣ ⫽ .78). Participants also provided demographic Relationship Between Wise Reasoning, Well-Being, Analytic
information including their age, education, family income, and Abilities, and Personality
occupation. We coded participants’ occupations using the Interna-
tional Socioeconomic Index of Occupational Status (Ganzeboom Wise reasoning
& Treiman, 1996). Intergroup Interpersonal
Variable (Session 1) (Session 2) Total

Results Relation to well-being

Life satisfaction (n ⫽ 144) .16ⴱ .07 .17ⴱ
Positive affect (n ⫽ 222) .03 ⫺.04 .01
Preliminary Analyses: Interitem Correlations and Negative affect
Factor Extraction (reverse scored; n ⫽ 222) .18ⴱⴱ .23ⴱⴱ .27ⴱⴱⴱ
Relationship quality
Results from exploratory analyses are summarized in Table 3. (log; n ⫽ 166) .23ⴱⴱ .15ⴱ .25ⴱⴱⴱ
All pairwise correlations of wise reasoning dimensions were sig- Depressive brooding
nificant except for limits of knowledge being unrelated to change (reverse scored; n ⫽ 185) .33ⴱⴱⴱ .18ⴱⴱ .33ⴱⴱⴱ
Positive vs. negative words in
and conflict resolution (mean r ⫽ .25). Principal component anal- speech (n ⫽ 221) .16ⴱ .16ⴱ .19ⴱⴱ
yses and a scree plot provided evidence for a single factor, which
accounted for over 38% of the variance. An alternative two-factor Relation to cognitive abilities and personality
solution yielded highly correlated factors (r ⫽ .67), with the Analytic abilities
second factor explaining a comparable amount of variance as each Speed of processing ⫺.25ⴱⴱⴱ ⫺.05 ⫺.25ⴱⴱⴱ
of the six items by itself (17%). In the interest of parsimony and to WAIS Digit Span ⫺.01 .02 .01
WAIS–Comprehension .03 .22ⴱⴱ .08
enhance measurement reliability, we collapsed the scores into a WAIS–Vocabulary .15ⴱ .31ⴱⴱⴱ .25ⴱⴱⴱ
single index.3 Big Five personality dimensions
Neuroticism ⫺.13 ⫺.11 ⫺.15
Extraversion ⫺.06 ⫺.07 ⫺.06
Relation to Cognitive Abilities and Personality Openness .05 .01 .06
Agreeableness .24ⴱ .20† .27ⴱⴱ
As shown in Table 4, speed of processing was negatively Conscientiousness .09 .16 .12
correlated with wise reasoning about intergroup conflicts (an
artifact of differential age effects on processing speed and wise Note. The values reported are zero-order correlation coefficients. Person-
ality measures were assessed via a mailed questionnaire 2 years after
reasoning, rpartial ⫽ .05, ns), whereas verbal cognitive abilities completion of the laboratory sessions, leading to an attrition (n ⫽ 104).
were positively associated with wise reasoning. Specifically, WAIS ⫽ Wechsler Adult Intelligence Scale.
WAIS–Comprehension was positively associated with wise rea- †
p ⱕ .10. ⴱ p ⱕ .05. ⴱⴱ p ⱕ .01. ⴱⴱⴱ p ⱕ .001.
soning about interpersonal scenarios, and WAIS–Vocabulary
was positively associated with wise reasoning about both inter-
group and interpersonal conflicts. Among the Big Five factors, dency to brood, and a more positive way of talking about social
only Agreeableness was positively related to wise reasoning. conflicts but not more positive affect in daily life. This pattern of
results was highly consistent across both intergroup and interper-
Relation to Well-Being sonal scenarios.
In contrast to wise reasoning, cognitive abilities showed no
Table 4 indicates a significant association between wise reason- systematic relationship to the well-being indicators. Faster speed
ing and a wide range of well-being indicators. Participants who of processing was negatively related to positive affect (r ⫽ ⫺.13,
scored high on wise reasoning reported less negative affect in daily p ⫽ .06) and relationship quality (r ⫽ ⫺.15, p ⫽ .05) and positively
life, better relationship quality, greater life satisfaction, less ten- related to brooding (r ⫽ .12, p ⫽ .10). Greater crystallized abilities were
correlated with less positive affect (rWAIS–Comprehension ⫽ ⫺.19, p ⫽
.006; r WAIS–Vocabulary ⫽ ⫺.26, p ⫽ .001) and less negative affect
(r WAIS–Vocabulary ⫽ ⫺.14, p ⫽ .07). However, higher scores on the
Table 3
WAIS–Comprehension task were positively related to frequency
Factor Loadings for the Six Dimensions and
of positive versus negative words in the discourse (r ⫽ .13, p ⫽
Dimension Intercorrelations

Factor 3
In Session 1 (intergroup conflicts), we explored wise reasoning
Dimension loading 1 2 3 4 5 6
about fictional scenarios, which had similar narrative structure. We
1. Change .56 — .23ⴱⴱⴱ .23ⴱⴱⴱ .26ⴱⴱⴱ .38ⴱⴱⴱ .03 attempted to increase the external validity in Session 2 (interpersonal
2. Compromise .70 — .44ⴱⴱⴱ .35ⴱⴱⴱ .27ⴱⴱⴱ .14ⴱ conflicts) by selecting scenarios from the real newspaper column. The
3. Flexibility .76 — .35ⴱⴱⴱ .25ⴱⴱⴱ .43ⴱⴱⴱ greater diversity likely resulted in somewhat lower internal reliability of
4. Perspective .64 — .17ⴱⴱ .17ⴱⴱ wisdom scores in Session 2 (Cronbach’s ␣ ⫽ .50) than in Session 1
5. Resolution .55 — .03 (Cronbach’s ␣ ⫽ .71). We attempted to increase measurement reliabil-
6. Limits of ity by performing the main analyses on the composite index across
knowledge .44 — intergroup and interpersonal scenarios (Cronbach’s ␣ on six wisdom
dimensions across both sessions ⫽ .71). Results were comparable
ⴱ ⴱⴱ ⴱⴱⴱ
p ⱕ .05. p ⱕ .01. p ⱕ .001. across both types of scenarios (see Table 4).
6 GROSSMANN, NA, VARNUM, KITAYAMA, AND NISBETT

.06). The remaining correlations between cognitive abilities and reasoning on longevity remained significant in analyses with
well-being indicators were negligible (|rs| ⬍ .09). such covariates as sociodemographic factors (gender, educa-
Replicating and extending previous work, personality factors tion, occupational status, income), perceived health (B ⫽
showed a number of significant correlations with well-being ⫺1.389, SE ⫽ 0.806, |t| ⫽ 1.72, p ⫽ .09), or verbal cognitive
(Diener, Oishi, & Lucas, 2003). Neuroticism was related to less abilities (B ⫽ ⫺1.325, SE ⫽ 0.782, |t| ⫽ 1.70, p ⫽ .09).
life satisfaction (r ⫽ ⫺.52, p ⬍ .001), less positive affect (r ⫽
⫺.25, p ⫽ .01), more negative affect (r ⫽ .22, p ⫽ .03), and Age, Wisdom, and Well-Being
more brooding (r ⫽ .51, p ⬍ .001). Agreeableness was related
to more life satisfaction (r ⫽ .23, p ⫽ .03), more positive affect We subsequently examined whether wise reasoning mediates
(r ⫽ .23, p ⫽ .02), less negative affect (r ⫽ ⫺.19, p ⫽ .05), the relationship between age and well-being by performing a
greater relationship quality (r ⫽ .36, p ⫽ .001), less brooding structural equation analysis with well-being indices as indicators
(r ⫽ .25, p ⫽ .02), and more positive versus negative words in of a latent well-being construct. Each well-being indicator showed
discourse (r ⫽ .22, p ⫽ .02). Extraversion was related to lower a significant contribution to the latent well-being construct, and the
life satisfaction (r ⫽ ⫺.22, p ⫽ .04) but also less negative affect model showed an acceptable fit, ␹2(19) ⫽ 27.13, root-mean-square
(r ⫽ ⫺.18, p ⫽ .07). Finally, conscientiousness was linked to error of approximation ⫽ .04, comparative fit index ⫽ .96).
more positive and less negative affect (r ⫽ .24, p ⫽ .02, and r ⫽ Indicators such as life satisfaction (␤ ⫽ .549, t ⫽ 6.41, p ⬍ .001),
⫺.21, p ⫽ .03, respectively). negative affect (reverse coded; ␤ ⫽ .516, t ⫽ 7.16, p ⬍ .001),
relationship quality (␤ ⫽ .345, t ⫽ 4.00, p ⬍ .001), and brooding
(reverse coded; ␤ ⫽ .718, t ⫽ 10.37, p ⬍ .001) showed substantial
Wisdom and Well-Being: Control Analyses With contributions, whereas the contributions of positive affect (␤ ⫽
Cognitive Abilities, Personality, and Sociodemographic .184, t ⫽ 2.24, p ⫽ .025) and positive speech (␤ ⫽ .238, t ⫽ 2.83,
Factors p ⫽ .005) were comparably smaller.
As Figure 1 illustrates, age was positively related to wise
We next ran a series of multivariate regression models, in
reasoning and well-being, and the effect of wise reasoning on
which we simultaneously included wise reasoning and a set of
well-being was significant when controlling for age. Moreover,
covariates as predictors of well-being. Specifically, we tested
the indirect effect of age on well-being via wise reasoning was
how wise reasoning predicts well-being when controlling for a
significant (see Figure 1 for 95% confidence intervals [CIs]
set of crystallized cognitive abilities (WAIS comprehension and
from a nonparametric bootstrap test with 2,000 random replace-
vocabulary tests), Agreeableness, or length of elaboration
ments, the technique of choice for assessing indirect effect in
(quantified as the number of words in the narrative). As Table
smaller samples; Preacher, Rucker, & Hayes, 2007). Follow-up
5 indicates, wise reasoning remained a significant predictor of
analyses indicated that the indirect effect was significant for
the majority of well-being indicators when simultaneously con-
middle-aged (at Mage; Sobel Z ⫽ 1.90, p ⫽ .057; 90% CI [.001,
trolling for verbal abilities, Agreeableness, and response length.
.004]) and older adults (1 standard deviation above the mean
There were two exceptions. First and similar to analyses with-
age, Sobel Z ⫽ 2.43, p ⫽ .015, 90% CI [.001, .004]) but not for
out covariates, wise reasoning did not predict more positive
younger adults (1 standard deviation below the mean age, Sobel
affect. Second, wise reasoning did not significantly predict
Z ⫽ 1.24, ns, 90% CI [0, .003]), suggesting that the indirect
more life satisfaction when controlling for Agreeableness (␤ ⫽
effect of age on well-being via wise reasoning was moderated
.14, t ⫽ 1.51, p ⫽ .13). Moreover, structural equation analyses
by age (Preacher et al., 2007).
in which all well-being indicators were modeled as part of a
Further analyses with wise reasoning, age, and their interac-
latent well-being construct regressed on wise reasoning and
tions predicting well-being showed a significant effect of wise
covariates indicated a significant overall effect of wise reason-
reasoning (␤ ⫽ .220, t ⫽ 2.19, p ⫽ .028) and age (␤ ⫽ .367, t ⫽
ing in each of the three models. In addition, Table 6 shows that
4.15, p ⬍ .001) and a marginal interaction (␤ ⫽ .158, t ⫽ 1.77,
including wise reasoning in the model with socioeconomic
p ⫽ .077). Comparable analyses with gender and wise reason-
factors and perceived health predicting well-being improves the
ing predicting well-being showed a significant effect of wise
fit of regression models for each well-being indicator except for
reasoning on well-being (␤ ⫽ .483, t ⫽ 6.28, p ⬍ .001) but no
positive affect, as indicated by the Model II R2 significance
significant effects of gender (␤ ⫽ .055, t ⫽ 0.69, ns), nor was
level.
there a Gender ⫻ Wise Reasoning interaction (␤ ⫽ .007, t ⫽

Longevity 4
The association between wise reasoning and longevity/mortality was
Because mortality is quite rare among younger adults, lon- further moderated by participants’ age (B ⫽ 0.186, SE ⫽ 0.065, |t| ⫽ 2.87,
gevity analyses focused on participants above 45 years of age. p ⫽ .004). Simple slopes analyses (Aiken et al., 1996) indicated that
The results were comparable when exploring the impact of wise middle-aged participants who scored low (10th quantile) on wise reasoning
showed a mortality probability as high as that of their older counterparts
reasoning on longevity on the full sample. The results of a (B ⫽ 0.022, SE ⫽ 0.037, |t| ⬍ 1, ns.). However, middle-aged participants
probit model regression with wise reasoning and age as predic- who scored above average (50th quantile) on wise reasoning showed
tors and longevity (1 ⫽ dead, 0 ⫽ alive; 13 deaths out of 127 significantly lower mortality probability than their older counterparts (B ⫽
participants) as the dependent variable showed a significant 1.208, SE ⫽ 0.036, |t| ⫽ 3.33, p ⬍ .001). In addition, wise reasoning did
not significantly contribute to greater longevity among participants above
effect of age (B ⫽ 0.048, SE ⫽ 0.019, |t| ⫽ 2.58, p ⫽ .01), and 67 years of age (|B| ⬍ 2.299, SE ⫽ 1.644, |t| ⫽ 1.40, p ⫽ .16). Note,
a marginally significant effect of wisdom on longevity (B ⫽ however, that the death numbers in the present study were fairly small, thus
⫺1.325, SE ⫽ 0.782, |t| ⫽ 1.70, p ⫽ .09).4 The effect of wise interaction results have to be interpreted with caution.
 REASONING AND WELL-BEING 7

Table 5
Wisdom and Well-Being: Control Analyses With Cognitive Abilities and Personality

Variable LS POS NEG REL BROOD SPEECH WB

Model I
Wise reasoning .165/1.896† .064/0.961 .275/4.316ⴱⴱⴱ .251/3.138ⴱⴱ .343/4.891ⴱⴱⴱ .201/2.970ⴱⴱ .497/6.243ⴱⴱⴱ
WAIS–Comprehension ⫺.057/0.599 ⫺.083/1.103 .060/0.797 .012/0.134 .079/0.992 .173/2.201ⴱ .093/0.983
WAIS–Vocabulary ⫺.012/0.121 ⫺.218/2.748ⴱⴱ .023/0.279 ⫺.087/0.959 ⫺.077/0.957 ⫺.123/1.279 ⫺.099/0.974
R2 .031/1.014 .076/2.027ⴱ .081/2.331ⴱ .070/1.576 .124/2.476ⴱ .065/1.749† .240/3.195ⴱⴱⴱ

Model II
Wise reasoning .138/1.509 ⫺.007/0.103 .268/4.216ⴱⴱⴱ .183/2.229ⴱ .309/4.264ⴱⴱⴱ .174/2.542ⴱ .359/3.981ⴱⴱⴱ
Agreeableness .160/1.322 .214/2.116ⴱ .158/1.600 .320/3.493ⴱⴱⴱ .183/1.749† .150/1.420 .433/3.683ⴱⴱⴱ
R2 .044/1.138 .046/1.056 .097/2.346ⴱ .136/2.352ⴱ .129/2.585ⴱⴱ .052/1.482 .406/3.648ⴱⴱⴱ

Model III
Wise reasoning .166/1.743† ⫺.001/0.015 .271/4.044ⴱⴱⴱ .214/2.493ⴱ .355/4.993ⴱⴱⴱ .212/3.044ⴱⴱ .491/5.998ⴱⴱⴱ
Response length ⫺.011/0.124 .048/0.661 .040/0.569 .051/0.637 ⫺.065/0.928 ⫺.064/0.925 ⫺.030/⫺0.355
R2 .028/0.853 .002/0.328† .075/2.161ⴱ .048/1.396 .130/2.421ⴱⴱ .049/1.471 .232/3.125ⴱⴱ
Note. The values to the left of the slash are standardized beta coefficients; values to the right of the slash are t values. LS ⫽ life satisfaction; POS ⫽
positive affect; NEG ⫽ less negative affect; REL ⫽ relationship quality; BROOD ⫽ less rumination; SPEECH ⫽ positive versus negative words when
talking about social conflicts; WB ⫽ latent well-being construct with contribution from each well-being indicator; WAIS ⫽ Wechsler Adult Intelligence
Scale. Missing cases were estimated with the Mplus 6.1 full-maximum likelihood procedure. Model fit indices (root-mean-square errors of approximation)
were, for Model I, .06; for Model II, .04; and for Model III, .04.
†
p ⱕ .10. ⴱ p ⱕ .05. ⴱⴱ p ⱕ .01. ⴱⴱⴱ p ⱕ .001.

0.08, ns). Subsequent simple slope analyses (Aiken, Reno, & the younger adults (1 standard deviation below the mean age,
West, 1996) are illustrated in Figure 2. The relationship be- B ⫽ 0.076, SE ⫽ 0.218, t ⬍ 1, ns).
tween wise reasoning and well-being was marginally significant
among the middle-aged participants (at the mean age, B ⫽ Summary
0.297, SE ⫽ 0.159, t ⫽ 1.86, p ⫽ .06) and significant among the
older adults (1 standard deviation above the mean age; B ⫽ Consistent with prior research, cognitive abilities such as crys-
0.518, SE ⫽ 0.208, t ⫽ 2.49, p ⬍ .01) but not significant among tallized intelligence, processing speed, and working memory

Table 6
Model Comparison: Sociodemographic Factors Versus Sociodemographic Factors and Wise Reasoning

Variable LS POS NEG REL BROOD SPEECH WB

Model IV
Occupation ⫺.065/0.680 ⫺.221/3.018ⴱⴱ .039/0.503 ⫺.103/1.194 ⫺.009/0.113 ⫺.070/0.893 ⫺.061/0.623
Education (years) .062/0.665 ⫺.006/0.076 .045/0.595 .103/1.163 .034/0.421 .041/0.533 .076/0.807
Income (log) ⫺.002/0.023 ⫺.089/1.318 ⫺.050/0.735 ⫺.090/1.179 ⫺.141/1.990ⴱ .122/1.787† ⫺.133/1.541
Perceived health .198/2.543ⴱⴱ .282/4.520ⴱⴱⴱ .191/2.887ⴱⴱ .160/2.145ⴱ .285/4.259ⴱⴱⴱ .070/1.033 .417/5.264ⴱⴱⴱ
么 ⫽ 1⁄2/乆 ⫽ ⫺1⁄2 .039/0.491 .043/0.682 .044/0.671 ⫺.170/2.309ⴱ .053/0.766 ⫺.072/1.080 .034/0.412
R2 .043/1.384 .116/2.822ⴱⴱ .052/1.783† .072/1.846† .108/2.527ⴱⴱ .027/1.227 .198/2.934ⴱⴱ

Model V
Occupation ⫺.084/0.887 ⫺.225/3.052ⴱⴱ .007/0.089 ⫺.124/1.462 ⫺.047/0.606 ⫺.093/1.189 ⫺.111/1.216
Education (years) .052/0.564 ⫺.007/0.098 .030/0.407 .083/0.945 .006/0.081 .028/0.377 .047/0.540
Income (log) ⫺.002/0.029 ⫺.089/1.324 ⫺.048/0.729 ⫺.086/1.156 ⫺.142/2.110ⴱ .120/1.801† ⫺.131/1.637
Perceived health .192/2.460ⴱ .280/4.487ⴱⴱⴱ .178/2.755ⴱⴱ .144/1.944ⴱ .267/4.111ⴱⴱⴱ .058/0.869 .387/5.143ⴱⴱⴱ
么 ⫽ 1⁄2/乆 ⫽ ⫺1⁄2 .051/0.641 .046/0.721 .065/1.021 ⫺.152/2.067ⴱ .078/1.189 ⫺.057/0.874 .064/0.847
Wise reasoning .152/1.716† .029/0.452 .269/4.272ⴱⴱⴱ .208/2.573ⴱⴱ .323/4.695ⴱⴱⴱ .191/2.834ⴱⴱ .454/5.984ⴱⴱⴱ
R2 .063/1.687† .118/2.835ⴱ .112/2.915ⴱ .108/2.358ⴱ .197/3.782ⴱⴱⴱ .063/1.913† .396/4.799ⴱⴱⴱ
Note. The values to the left of the slash are standardized beta coefficients; values to the right of the slash are t values. LS ⫽ life satisfaction; POS ⫽
positive affect; NEG ⫽ less negative affect; REL ⫽ relationship quality; BROOD ⫽ less rumination; SPEECH ⫽ positive versus negative words
when talking about social conflicts; WB ⫽ latent well-being construct with contribution from each well-being indicator. Missing cases were
estimated with the Mplus 6.1 full-maximum likelihood procedure. Model fit indices (root-mean-square errors of approximation) were, for Model IV,
.05, and, for Model V, .05.
†
p ⱕ .10. ⴱ p ⱕ .05. ⴱⴱ p ⱕ .01. ⴱⴱⴱ p ⱕ .001.
8 GROSSMANN, NA, VARNUM, KITAYAMA, AND NISBETT

showed no systematic relationship to well-being. In contrast, the

ability to reason wisely about social conflicts was associated with
greater global life satisfaction, greater satisfaction with social
relationships, less negative affect in daily life, a lower propensity
to brood, relatively positive discussion social conflicts, and greater
longevity. This positive association between wise reasoning and
well-being was consistent when looking separately at wise reason-
ing about intergroup and interpersonal conflicts and when control-
ling for crystallized cognitive abilities or individual differences in
personality. It is important to note that wise reasoning was not
related to greater positive affect in daily life. The latter finding
suggests that people who reason wisely are more content without
being chronically more positive.5 Figure 2. The association between wise reasoning and well-being as
Moreover, wise reasoning explained a greater amount of vari- moderated by participants’ age in a multivariate regression analysis. Sim-
ance in individual well-being than did gender or various socioeco- ple slopes for the minimum (min), mean, and maximum (max) ages in the
nomic indicators. We also observed two age-related patterns. First, sample are shown.
the effect of age on well-being was partially mediated by wise
reasoning. Second, the link between wise reasoning and well-being
was stronger among older adults and absent among younger adults. wise reasoning and well-being, as well as between wise reasoning
and longevity. In the present article, we suggested that wiser
Discussion reasoning about social conflicts leads to greater well-being and not
vice versa. A reversal would suggest that positive affect facilitates
Psychologists have long sought to identify strategies that are a desire to explore the world (e.g., Fredrickson, 1998), which, in
reliably associated with greater well-being (Seligman, 2002). The turn, might lead to more experience and stimulate the development
observation that wise reasoning improves into old age (Grossmann of wise reasoning. But positive affect was not related to well-
et al., 2010; Worthy et al., 2011) in conjunction with experimental being; thus, it seems unlikely that wisdom is the product of chronic
work on the malleability of wise reasoning (Kross & Grossmann, positive affect. At the same time, it is plausible that some core
2012) suggests that it may be possible to train people to reason emotional and social intelligence may contribute toward both
wisely. Our findings further suggest that wise reasoning is a greater wise reasoning and good mental health. Further, well-being
potential psychological mechanism that may explain age-related has been linked to greater longevity on its own right (Carstensen et
differences in well-being. This finding dovetails with work by al., 2011), suggesting that some common underlying mechanism
Carstensen and colleagues demonstrating that as people age, they for wise reasoning and well-being has an effect on longevity.
shift their priorities toward interpersonal issues and develop Therefore, additional longitudinal research with the simultaneous
greater emotional competence (Charles & Carstensen, 2010). Ex- assessment of wise reasoning, emotion regulation strategies (Gross
tending this work, the present research demonstrated that older & John, 2003), and social intelligence (Cantor & Kilhlstrom, 1987)
adults show greater ability to reason wisely about social conflicts would be useful to gain deeper insight into the directionality of the
than younger adults and that among middle-aged and older adults, relationship between wise reasoning and well-being and the pro-
such reasoning is positively linked to socioemotional benefits. cesses that may underlie this relationship.
In interpreting the results of our research, one important con- Another important question deals with the divergence of the
sideration deals with the directionality of the relationship between present findings and those of previous research that has explored
the relationship between wisdom-related knowledge and well-
being (Kunzmann & Baltes, 2003; Mickler & Staudinger, 2008).
First of all, in contrast to the previous studies, we provided
meaningful contexts, which would help older participants reason in
a wise way. Also, because of the lack of contextual information,
participants in their studies should have relied on their own knowl-
edge about specific life events (e.g., suicide attempt). We believe
that such focus on specific factual knowledge may be problematic.
A large body of research in cognitive psychology has demon-
strated that knowledge acquired through experiences in one situ-
ation often fails to transfer to another (Singley & Anderson, 1989).
Therefore, when measuring knowledge about specific life events,
Figure 1. Wise reasoning mediating the effect of age on well-being in a
structural equation model. Standardized coefficients are shown. The value
it may be important to sample from a heterogeneous set of life
under the age ¡ well-being path reveals the relationship between age and
well-being, after controlling for wise reasoning. The values in the square 5
The null relationship between wise reasoning and positive affect dove-
brackets correspond to the 95% confidence interval from a bootstrap test tails with some developmental work, suggesting that the greater well-being
performed to assess the significance of the indirect effect. The mediation is in older age is mainly accounted by steep declines in negative affect and
significant if the confidence interval does not include zero. ⴱⴱ p ⱕ .01. only a modest increase in positive affect (Stone, Schwartz, Broderick, &
ⴱⴱⴱ
p ⱕ .005. Deaton, 2010).
 REASONING AND WELL-BEING 9

domains to make accurate estimations about one’s declarative Charles, S. T., & Carstensen, L. L. (2010). Social and emotional aging.
knowledge about such events. Conflating wise reasoning with Annual Review of Psychology, 61, 383– 409. doi:10.1146/annurev.psych
specific knowledge about life dilemmas may also be problematic .093008.100448
for some other reasons. It is plausible that wiser people may show Chida, Y., & Steptoe, A. (2008). Positive psychological well-being and
mortality: A quantitative review of prospective observational studies.
a lower likelihood of encountering challenging life dilemmas and
Psychosomatic Medicine, 70, 741–756. doi:10.1097/PSY
therefore have less specific knowledge about them. This may be
.0b013e31818105ba
because wiser individuals are more likely to recognize and poten- Danner, D. D., Snowdon, D. A., & Friesen, W. V. (2001). Positive
tially preempt some of these dramatic life events. Further, scoring emotions in early life and longevity: Findings from the nun study.
individuals with great specific knowledge about life tragedies as Journal of Personality and Social Psychology, 80, 804 – 813. doi:
wiser could introduce confounds with regard to the link between 10.1037/0022-3514.80.5.804
wisdom and individual well-being: People who have greater Diener, E., & Fujita, F. (1995). Resources, personal strivings, and subjec-
knowledge about very difficult situations presumably learned tive well-being: A nomothetic and idiographic approach. Journal of
about them from personal experiences; thus, their psychological Personality and Social Psychology, 68, 926 –935. doi:10.1037/0022-
well-being might be lower. An important avenue for future re- 3514.68.5.926
search would be to test the extent to which the development of Diener, E., Oishi, S., & Lucas, R. E. (2003). Personality, culture, and
wisdom-related reasoning depends on personal or vicarious life subjective well-being: Emotional and cognitive evaluations of life. An-
nual Review of Psychology, 54, 403– 425. doi:10.1146/annurev.psych.54
experiences and if such experiences strengthen or weaken the link
.101601.145056
between wise reasoning and well-being.
Fiori, K. L., Antonucci, T. C., & Akiyama, H. (2008). Profiles of social
relations among older adults: A cross-cultural approach. Ageing and
Society, 28, 203–231. doi:10.1017/S0144686X07006472
Conclusion
Fredrickson, B. L. (1998). What good are positive emotions? Review of
Our results suggest that lay beliefs about the relationship be- General Psychology, 2, 300 –319. doi:10.1037/1089-2680.2.3.300
tween reasoning abilities and well-being are correct, with one Furnham, A. (2008). Relationship among four Big Five measures of
caveat. Whereas wise reasoning about social conflicts contributes different length. Psychological Reports, 102, 312–316. doi:10.2466/pr0
.102.1.312-316
to well-being, abstract cognitive abilities (as measured by intelli-
Ganzeboom, H. B. G., & Treiman, D. J. (1996). Internationally comparable
gence tests) do not. On the practical side, the present work sug-
measures of occupational status for the 1988 International Standard
gests that despite the cognitive declines often associated with older Classification of Occupations. Social Science Research, 25, 201–239.
age, the increasing number of older adults may be of great value doi:10.1006/ssre.1996.0010
for the social and emotional well-being of future communities. Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A very brief
measure of the Big-Five personality domains. Journal of Research in
Personality, 37, 504 –528. doi:10.1016/S0092-6566(03)00046-1
References Gross, J. J., & John, O. P. (2003). Individual differences in two emotion
Aiken, L. S., Reno, R. R., & West, S. G. (1996). Multiple regression: regulation processes: Implications for affect, relationships, and well-
Testing and interpreting interactions. Newbury Park, CA: Sage. being. Journal of Personality and Social Psychology, 85, 348 –362.
Baltes, P. B., & Smith, J. (2008). The fascination of wisdom: Its nature, doi:10.1037/0022-3514.85.2.348
ontogeny, and function. Perspectives on Psychological Science, 3, 56 – Grossmann, I., Karasawa, M., Izumi, S., Na, J., Varnum, M. E. W.,
64. doi:10.1111/j.1745-6916.2008.00062.x Kitayama, S., & Nisbett, R. E. (in press). Aging and wisdom: Culture
Baltes, P. B., & Staudinger, U. M. (2000). Wisdom: A metaheuristic matters. Psychological Science.
(pragmatic) to orchestrate mind and virtue toward excellence. American Grossmann, I., Na, J., Varnum, M. E., Park, D. C., Kitayama, S., & Nisbett,
Psychologist, 55, 122–136. doi:10.1037/0003-066X.55.1.122 R. E. (2010). Reasoning about social conflicts improves into old age.
Basseches, M. (1980). Dialectical schemata: A framework for the empirical PNAS: Proceedings of the National Academy of Sciences of the United
study of the development of dialectical thinking. Human Development, States of America, 107, 7246 –7250. doi:10.1073/pnas.1001715107
23, 400 – 421. doi:10.1159/000272600 Hedden, T., Park, D. C., Nisbett, R. E., Ji, L., Jing, Q., & Jiao, S. (2002).
Campbell, A., Converse, P. E., & Rogers, W. L. (1976). The quality of Cultural variation in verbal versus spatial neuropsychological function
American life: Perceptions, evaluations, and satisfactions. New York, across the lifespan. Neuropsychology, 16, 65–73. doi:10.1037/0894-
NY: Russell Sage Foundation. 4105.16.1.65
Cantor, N., & Kilhlstrom, J. (1987). Social intelligence. New York, NY: Hess, T. M., & Blanchard-Fields, F. (1999). Social cognition and aging.
Prentice Hall. San Diego, CA: Academic Press.
Carstensen, L. L., & Mikels, J. A. (2005). At the intersection of emotion Kahneman, D., Diener, E., & Schwarz, N. (1999). Well-being: The foun-
and cognition: Aging and the positivity effect. Current Directions in dations of hedonic psychology. New York, NY: Russell Sage Founda-
Psychological Science, 14, 117–121. doi:10.1111/j.0963-7214.2005 tion.
.00348.x Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone,
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. R. (2000). A. A. (2004, December 3). A survey method for characterizing daily life
Emotional experience in everyday life across the adult life span. Journal experience: The day reconstruction method. Science, 306, 1776 –1780.
of Personality and Social Psychology, 79, 644 – 655. doi:10.1037/0022- doi:10.1126/science.1103572
3514.79.4.644 Kitayama, S., Park, H., Sevincer, A. T., Karasawa, M., & Uskul, A. K.
Carstensen, L. L., Turan, B., Scheibe, S., Ram, N., Ersner-Hershfield, H., (2009). A cultural task analysis of implicit independence: Comparing
Samanez-Larkin, G. R., . . . Nesselroade, J. R. (2011). Emotional North America, Western Europe, and East Asia. Journal of Personality
experience improves with age: Evidence based on over 10 years of and Social Psychology, 97, 236 –255. doi:10.1037/a0015999
experience sampling. Psychology and Aging, 26, 21–33. doi:10.1037/ Kramer, D. (1983). Post-formal operations? A need for further conceptu-
a0021285 alization. Human Development, 26, 91–105. doi:10.1159/000272873
10 GROSSMANN, NA, VARNUM, KITAYAMA, AND NISBETT

Kramer, D. (2000). Wisdom as a classical source of human strength: Sigelman, L. (1981). Is ignorance bliss? A reconsideration of the folk
Conceptualization and empirical inquiry. Journal of Social and Clinical wisdom. Human Relations, 34, 965–974. doi:10.1177/
Psychology, 19, 83–101. doi:10.1521/jscp.2000.19.1.83 001872678103401104
Kross, E., & Grossmann, I. (2012). Boosting wisdom: Distance from the Singley, M. K., & Anderson, J. R. (1989). The transfer of cognitive skill.
self enhances wise reasoning, attitudes, and behavior. Journal of Exper- Cambridge, MA: Harvard University Press.
imental Psychology: General, 141, 43– 48. doi:10.1037/a0024158 Stanovich, K. E. (2009). What intelligence tests miss: The psychology of
Kunzmann, U., & Baltes, P. B. (2003). Wisdom-related knowledge: Af- rational thought. New Haven, CT: Yale University Press.
fective, motivational, and interpersonal correlates. Personality and So- Staudinger, U. M., & Glück, J. (2011). Psychological wisdom research.
cial Psychology Bulletin, 29, 1104 –1119. doi:10.1177/ Annual Review of Psychology, 62, 215–241. doi:10.1146/annurev.psych
0146167203254506 .121208.131659
Mickler, C., & Staudinger, U. M. (2008). Personal wisdom: Validation and Sternberg, R. J. (1998). A balance theory of wisdom. Review of General
age-related differences of a performance measure. Psychology and Ag- Psychology, 2, 347–365. doi:10.1037/1089-2680.2.4.347
ing, 23, 787–799. doi:10.1037/a0013928 Sternberg, R. J. (1999). The theory of successful intelligence. Review of
Mroczek, D. K., & Kolarz, C. M. (1998). The effect of age on positive and General Psychology, 3, 292–316. doi:10.1037/1089-2680.3.4.292
negative affect: A developmental perspective on happiness. Journal of Sternberg, R. J. (2004). Words to the wise about wisdom? A commentary
Personality and Social Psychology, 75, 1333–1349. doi:10.1037/0022- on Ardelt’s critique of Baltes. Human Development, 47, 286 –289.
3514.75.5.1333 doi:10.1159/000079155
Pascual-Leone, J. (1990). An essay on wisdom: Toward organismic pro- Sternberg, R. J. (2007). Wisdom, intelligence, and creativity synthesized.
cesses that make it possible. In R. J. Sternberg (Ed.), Wisdom: Its nature, New York, NY: Cambridge University Press.
origins, and development (pp. 244 –278). New York, NY: Cambridge Sternberg, R. J., & Jordan, J. (Eds.). (2005). A handbook of wisdom:
University Press. doi:10.1017/CBO9781139173704.013 Psychological perspectives. New York, NY: Cambridge University
Pennebaker, J. W., Francis, M. E., & Booth, R. J. (2001). Linguistic Inquiry Press. doi:10.1017/CBO9780511610486
and Word Count (LIWC): LIWC2001 [Computer software]. Mahwah, Stone, A. A., Schwartz, J. E., Broderick, J. E., & Deaton, A. (2010). A
NJ: Erlbaum. snapshot of the age distribution of psychological well-being in the
Pennebaker, J. W., & King, L. A. (1999). Linguistic styles: Language use United States. PNAS: Proceedings of the National Academy of Sciences
as an individual difference. Journal of Personality and Social Psychol- of the United States of America, 107, 9985–9990. doi:10.1073/pnas
ogy, 77, 1296 –1312. doi:10.1037/0022-3514.77.6.1296 .1003744107
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psycho- Treynor, W., Gonzalez, R., & Nolen-Hoeksema, S. (2003). Rumination
logical aspects of natural language use: Our words, our selves. Annual reconsidered: A psychometric analysis. Cognitive Therapy and Re-
Review of Psychology, 54, 547–577. doi:10.1146/annurev.psych.54 search, 27, 247–259.
.101601.145041 Watten, R. G., Syversen, J. L., & Myhrer, T. (1995). Quality of life,
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing mod- intelligence and mood. Social Indicators Research, 36, 287–299. doi:
erated mediation hypotheses: Theory, methods, and prescriptions. Mul- 10.1007/BF01078818
tivariate Behavioral Research, 42, 185–227. doi:10.1080/ Wirthwein, L., & Rost, D. H. (2011). Giftedness and subjective well-being:
00273170701341316 A study with adults. Learning and Individual Differences, 21, 182–186.
Rowley, J. (2006). What do we need to know about wisdom? Management doi:10.1016/j.lindif.2011.01.001
Decision, 44, 1246 –1257. doi:10.1108/00251740610707712 Worthy, D. A., Gorlick, M. A., Pacheco, J. L., Schnyer, D. M., & Maddox,
Rowley, J., & Slack, F. (2009). Conceptions of wisdom. Journal of W. T. (2011). With age comes wisdom: Decision making in younger and
Information Science, 35, 110 –119. doi:10.1177/0165551508092269 older adults. Psychological Science, 22, 1375–1380. doi:10.1177/
Salthouse, T. A. (2004). What and when of cognitive aging. Current 0956797611420301
Directions in Psychological Science, 13, 140 –144. doi:10.1111/j.0963-
7214.2004.00293.x
Seligman, M. E. P. (2002). Authentic happiness: Using the new positive Received April 6, 2012
psychology to realize your potential for lasting fulfillment. New York, Revision received July 4, 2012
NY: Free Press. Accepted July 4, 2012 䡲