Journal of Affective Disorders 128 (2011) 24–32
Contents lists available at ScienceDirect
Journal of Affective Disorders
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j a d
Research report
Personality and the perception of health and happiness
C. Robert Cloninger a,⁎, Ada H. Zohar b
a
b
Dept. of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA
Dept. of Behavioral Sciences, Ruppin Academic Center, Israel
a r t i c l e
i n f o
Article history:
Received 25 February 2010
Accepted 3 June 2010
Available online 26 June 2010
Keywords:
Personality
Character
Health
Wellness
Happiness
Well-being
a b s t r a c t
Background: Health is a state of physical, mental, and social well-being. Personality traits
measure individual differences in adaptive functioning and mental health, but little is known
about how well personality accounts for health's affective aspects (i.e., “happiness”) and its
non-affective aspects (i.e., “wellness”) in the general population.
Methods: 1102 volunteer representatives of the Sharon area of Israel completed the
Temperament and Character Inventory (TCI), the Positive and Negative Affect Scale (PANAS),
the Satisfaction with Life Scale (SWLS), the Multidimensional Scale of Perceived Social Support
(PSS), and the subjective health assessment of the General Health Questionnaire (GHQ).
Multidimensional personality profiles were used to evaluate the linear and non-linear effects of
interactions among dimensions on different aspects of well-being.
Results: Self-directedness was strongly associated with all aspects of well-being regardless of
interactions with other dimensions. Cooperativeness was strongly associated with perceived
social support, and weakly with other aspects of well-being, particularly when Selfdirectedness was low. Self-transcendence was strongly associated with positive emotions
when the influence of the other character dimensions was taken into account. Personality
explained nearly half the variance in happiness and more than one-third of the variance in
wellness.
Limitations: Our data are cross-sectional and self-reported, so they are subject to personal
perceptual bias.
Conclusions: The emotional, social, and physical aspects of well-being are interdependent, but
specific configurations of TCI Self-directedness, Cooperativeness, and Self-transcendence
influence them differentially. Interactions among different combinations of character traits
have strong effects on the perception of both wellness and happiness.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Much is known about the relationship of personality to
psychopathology (Cloninger, 1999), but much less is known
about the relationship of personality to health as a state of
physical, mental, and social well-being (Cloninger, 2004). Yet
a mental health professional needs to understand the
relationship of personality to well-being in order to help
motivate both the promotion of health and the reduction of
⁎ Corresponding author. Tel.: + 1 314 362 7005; fax: + 1 314 362 5594.
E-mail address: clon@wustl.edu (C.R. Cloninger).
0165-0327/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jad.2010.06.012
distress and disability (Amering and Schmolke, 2009;
Cloninger, 2006). Health is much more than the absence of
disease or infirmity, so clinicians need to understand how to
promote health in all its aspects (WHO, 1946, 2001).
To proceed scientifically, we must be able to measure the
distinct aspects of well-being in a reliable way. Several
reliable ways of measuring well-being are available, including
measures that focus on the presence of positive emotions and
the absence of negative emotions, life satisfaction (Pavot and
Diener, 1993), social engagement (Zimet et al., 1990), and
physical wellness (Knauper and Turner, 2003). The term
“subjective well-being” or “happiness” is often used to refer
to a combination of the absence of negative emotions and the
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
presence of positive emotions, life satisfaction, and social
engagement (Ryan and Deci, 2001). Such measures of
subjective well-being emphasize the importance of the
hedonic aspects of experience, such as pleasure, satisfaction,
and happiness (Kahneman et al., 2003). However, the
identification of life satisfaction and subjective well-being
with pleasure and positive emotion has been seriously
challenged because people can regard themselves as well
even when they are suffering, if they regard their unpleasant
experience as meaningful and purposeful (Cassell, 1999;
Keyes et al., 2002; Ryan and Deci, 2000; Ryff and Keyes,
1995).
The term “psychological well-being” or “eudaimonia” has
been used to refer to well-being that arises from a combination
of character strengths involving facets of Self-directedness (e.g.
autonomy, life purpose, environmental mastery, and selfacceptance), Cooperativeness (e.g., positive relations with
others), and Self-transcendence (e.g., personal growth and
self-actualization) (Ryan and Deci, 2001; Ryff and Keyes, 1995;
Schmutte and Ryff, 1997). Recently even those who use the term
“happiness” from hedonic and positive psychology are actually
describing the concept of eudaimonic well-being or “psychological wealth” in which lasting satisfaction emerges from
meaningful values and goals as a process of character development (Diener and Biswas-Diener, 2008). Mature character traits
are associated with eudaimonic well-being and promote both
emotional and physical health (Cloninger, 2004; Ryff et al., 2004,
2006). Temperament traits, like Harm Avoidance (e.g., anxiety
proneness) or Novelty Seeking (e.g., anger proneness), are
associated with hedonic well-being and also have some
associations with both physical and emotional health (Cloninger et al., 1998; Moldin et al., 1993). However, the association of
health and “wellness” with character strengths appears to be
stronger and more consistent than with temperament (Manderscheid et al., 2010; Ruini et al., 2003; Ryff et al., 2004).
Likewise, recovery of quality of life often emphasizes the
importance of the same character traits, even when slightly
different terminology is used to emphasize the role of hope,
empathy, and respect in person-centered care (Amering and
Schmolke, 2009). It is remarkable that theoretically diverse
approaches have converged in their substantive conclusions
about how to describe a health-promoting character profile.
Despite this remarkable convergence in broad constructs,
there is little data about the precise relationships between
comprehensive models of personality and the different
components of health and well-being in the general population. The data available is limited to factor analytically derived
tests that confound temperament and character (Schmutte
and Ryff, 1997) as a result of making the simple but invalid
assumption of linearity of effects in the complex adaptive
systems underlying the within-person development of personality and well-being (Cervone, 2004, 2005; Cloninger,
2008). Unfortunately, such simplifying linearity assumptions
are inappropriate for phenomena in which every antecedent
variable can have different outcomes (“multi-finality”) and
every outcome can have different antecedents (“equifinality”), which is always the case for personality and health
(Cicchetti and Rogosch, 1996; Cloninger et al., 1997; Schmutte
and Ryff, 1997).
Accordingly, we set out to identify a large sample from the
general population of the Sharon region in Israel to measure
25
personality and different aspects of health and well-being in a
prospective longitudinal study in which we are assessing
subjects thoroughly with well-standardized self-reports,
psychiatric interviews, and also conducting physical examinations and laboratory testing. Israel provides a particularly
informative population because of its long cultural tradition
of learning-based encouragement of personal responsibility
and community progress (Pease, 2009). This tradition of
encouraging character development has resulted in Israel's
“Economic Miracle”, an outstanding record of capacity for
coping and productivity under stressful conditions (Brooks,
2010; Senor and Singer, 2009). Our study participants were
highly diverse adults and included a representative number
of active workers in highly successful businesses, farms, and
professions.
We studied multidimensional profiles of personality so
that our person-centered approach would allow an understanding of what happens to an individual person with
specific combinations of personality traits adapting within his
or her biopsychosocial context. We focused here on character
traits as measured by the Temperament and Character
Inventory (TCI) because character has been found to be
strongly related to well-being whereas temperament traits is
only weakly associated (Cloninger, 2004; Ruini et al., 2003).
Our aims were to evaluate the interactions among specific
combinations of character traits in relation to both the
affective (“happiness”) and non-affective (“wellness”)
aspects of health in a large and diverse general population.
We analyzed the influence of character on well-being using
both linear and non-linear methods to evaluate the practical
importance of recognizing the complexity of developmental
processes.
2. Methods
Participants were 1102 consecutive community volunteers
from the Sharon area of Israel. Participation was solicited via
mailbox advertisement, and via a series of thirteen public
lectures given in community centers, protected living projects,
and a college campus. The conditions for participation included
being 40 years of age or older, and having adequate Hebrew to
complete the study questionnaire. Participants were not paid,
but they are enrolled in a four-year longitudinal study in which
they will be eligible for a free medical checkup at a large
medical facility. The participants had a mean age of 57.8 years.
36.8% were men and 63.2% women. Among the participants,
69.1% were married, 17.3% were divorced, 9.2% were widowed
and the rest were single. There was a range of education: 2.0%
had only primary school education, and 2.1% had Ph.D.s. The
rest were evenly distributed among high school education,
further vocational education, BA and MA degrees. Altogether
the participants in this study reflected the demographic
characteristics of age-matched residents of the Sharon (Central
Bureau of Statistics, 2007).
2.1. Data analysis
All questionnaires were optically scanned using OMR
software (Remark Office version 7.0) and imported into SPSS
version 17 for Windows, in which analyses were conducted. All
26
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
measures reported here were analyzed in standard form with
mean zero and standard deviation of one to facilitate comparison.
3. Results
3.1. Character profiles
2.2. Measures
2.2.1. Temperament and character
The TCI-140 includes 140 items that are answered on a 5
point Likert-like scale from absolutely false to absolutely true. It
measures four dimensions of temperament: Harm Avoidance
(HA), Novelty Seeking (NS), Reward Dependence (RD), and
Persistence, (PS) (Zohar and Cloninger, 2010). There are three
dimensions of character: Self-directedness (SD), Cooperativeness (CO), and Self-Transcendence, (ST). There are 20 items for
each of the 7 dimensions except for ST, which has 16 allowing
for inclusion of 4 validity items to assess inattention or
carelessness. The TCI-140 was translated into Hebrew for the
current study by a process of translation, independent back
translation. The reliabilities (alpha) of the scales analyzed here
were 0.88 for SD, 0.79 for CO, and 0.89 for ST. In the current
study, three participants got all 4 validity items wrong, and four
got 3 validity items wrong. These 7 participants constitute
0.63% of the total sample. Their TCI-140's were removed from
subsequent analyses.
2.2.2. Affect, Life satisfaction, and Social support
The Positive and Negative Affect Scale (PANAS, Watson et
al., 1988) includes 20 items half of which describe negative
affective states and half positive, which are endorsed on a
Likert-like scale. It yields two subscales, one of negative affect
and one of positive affect. Scale reliability estimates were
α = 0.79 and α = 0.83 respectively.
Life satisfaction, sometimes called subjective well-being, was
assessed with a five-item report of well-being answered on a
seven-point Likert-like scale (Pavot and Diener, 1993). It was
translated for the current study by a process of translation,
independent back translation, comparison and revision. Its
potential scores ranged from 5 (lowest) to 5 (highest), and
these were transformed to standard form with mean 0 and
standard deviation 1. Scale reliability was α = 0.87.
Social support was assessed with the Multidimensional
Perceived Social Support Scale (Zimet et al., 1990). It includes
four items to assess perceived support from friends, four items
about family support, and four items about support by an
intimate partner. The 12 items are rated on a 7-point scale each,
giving a potential range of scores from 12 to 84, which were
transformed to standard form. It has been used extensively in
Hebrew translation. Scale reliability in the current study was
α=0.93.
2.2.3. Health behavior and Health status
Participants were asked about the frequency of their weekly
physical activity, sexual activity, smoking behavior, and their
subjective evaluation of their health over the last 30 days on a 5category scale. The subjective health assessment analyzed here
was the first item of the General Health Questionnaire (Knauper
and Turner, 2003), asking “how would you describe your health
over the past 30 days?” It was rated very bad (1), bad (2),
mediocre (3), good (4), or excellent (5). Scores are tabulated here
in standard form.
To form the character profiles, the sample was divided
into subjects above and below the median for each of the
three character traits, after excluding the 48 participants who
were in the middle third of the distribution for all three traits.
Then the participants were grouped according to all the
possible combinations of high and low character scores to
define the 8 possible character configurations shown in
Table 1. The character profiles are listed in the order that were
previously observed in the USA to be associated with greater
happiness and character integration (Cloninger, 2004).
3.2. The relationship of character profile with positive affect
The mean PANAS positive affect scores were compared
among people in the 8 character profiles, as shown in Fig. 1. We
evaluated the linear effects of character on positive affect by
analysis of variance, which revealed highly significant differences among the groups (F= 35.03, p = 0.000). The comparison between groups with the Bonferroni range correction
showed that the creative (SCT) profile was significantly higher
than all others with the exception of organized (SCt) profile.
The depressive (sct) profile was significantly lower than all 7
others.
We evaluated the non-linear influence of each of the
character dimensions on positive affect by paired comparisons of the effect of extremes of each character dimension
when the other two were controlled. As a detailed example,
the analysis of the influence of Self-transcendence is shown in
Table 2. Higher Self-transcendence was consistently associated with higher positive affect for each of the four possible
configurations of Self-directedness and Cooperativeness.
Likewise higher Self-directedness was consistently associated
with higher positive affect for each of the possible configurations of Cooperativeness and Self-transcendence (paired
t = 3.44 to 6.51, p = 0.000). The association of higher
Cooperativeness with positive affect was also highly significant when Self-directedness was low: higher Cooperativeness was associated with higher positive affect than lower
Cooperativeness in the contrast of moody vs disorganized
profiles (sCT vs scT, t = 2.8909, p = 0.004) and for dependent
vs depressive profiles (sCt vs sct, t = 2.996, p = 0.003). When
Self-directedness was high, the association of higher Cooperativeness with higher positive affect was weakly significant
for the contrast of organized vs autocratic profiles (SCt vs Sct,
Table 1
Frequency distribution of TCI character profiles.
Character Profile
N
Valid %
SCT—creative
SCt—organized
ScT—fanatical
Sct—autocratic
sCT—moody
sCt—dependent
scT—disorganized
Sct—depressive
Total
168
176
116
71
57
112
183
164
1047
16.0
16.8
11.1
6.8
5.4
10.7
17.5
15.7
100
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
27
Fig. 1. Positive affect as a function of character profile.
Fig. 2. Negative affect as a function of character profile.
t = 2.03, p = 0.043) and a trend for the contrast of creative vs
fanatical profiles (SCT vs ScT, t = 1.61, p = 0.109). Hence the
association of positive affect with character profiles was
highly non-linear.
between moody and disorganized profiles (sCT vs scT, t =
−2.14, p = 0.033) and there was a trend in the contrast
between creative and fanatical profiles (SCT vs ScT, t = −1.77,
p = 0.078). Cooperativeness has no significant association with
negative affect in other profiles. Self-transcendence was not
associated with lower negative affect in any contrast, and was
weakly associated with higher negative affect in the contrast of
disorganized vs depressive profiles (scT vs sct, t = 2.35,
p = 0.020).
3.3. The relationship of character profile with negative affect
The mean PANAS scores for negative affect were also
compared among the people with the 8 TCI character profiles,
as shown in Fig. 2. Analysis of variance showed that the groups
were significantly different from one another (F = 39.27,
p = 0.000). The comparison between groups with the Bonferroni range correction showed that the first four character
profiles with high Self-directedness (that is, SCT, SCt, ScT, and
Sct) were significantly lower than the last four character
profiles with low Self-directedness (that is, sCT, sCt, scT, and
sct). Thus the single strongest linear discriminator for negative
affect is Self-directedness: all those above the median for Selfdirectedness, whatever their other character scores, were
significantly less prone to negative affect.
The non-linear interactions of character dimensions on
negative affect were different than those observed for positive
affect. Self-directedness had a highly significant inverse
association with negative affect for each of the four possible
configurations of the other two character traits (t= −5.47 to
−8.07, p = 0.000). Cooperativeness had a weak association
when Self-transcendence was high: higher Cooperativeness
was associated with lower negative affect in the contrast
Table 2
The association of high Self-transcendence with positive affect.
High ST
profile
Low ST
profile
Difference in
positive affect
Paired-t
(probability)
Creative—SCT
Fanatical—ScT
Moody—sCT
Disorganized—scT
Organized—SCt
Autocratic—Sct
Dependent—sCt
Depressive—sct
41
40
38
36
6.07
4.65
3.06
4.10
vs
vs
vs
vs
38
37
36
34
(p = 0.000)
(p = 0.000)
(p = 0.003)
(p = 0.000)
3.4. Correlations among indicators of health and happiness
The relationships among our five indicators of health and
happiness were examined. As shown in Table 3, these
included positive and negative affects (PANAS), life satisfaction (SWLS), perceived social support (PSS), and perceived or
subjective health (SH). Table 3 summarizes the correlations
among these measures. Positive and Negative Affectivity
were largely uncorrelated (r = −0.09). Scores for the three
non-affective measures of life satisfaction, perceived social
support, and perceived health were weakly but positively
correlated with one another (r = + 0.23 to + 0.31), so we
formed a Composite Health Index (CHI) as the mean of these
three non-affective measures in standard form. Each individual measure of health was strongly correlated with the CHI
(r = +0.7 to +0.8).
Analysis of variance was used to compare the 8 character
profile groups for the additional three measures of well-being
besides affect. The profile groups differed significantly for all
three measures of well-being, including life satisfaction
(F = 41.08, p = 0.000), perceived social support (F = 14.61,
p = 0.000), and subjective health (F = 24.81, p = 0.000). The
means of the profile groups are depicted in Fig. 3 for all three
indicators of health and happiness. Post-hoc group comparisons using the Bonferroni correction showed that the means
of the creative (SCT) and of the organized (SCt) profile were
significantly higher than those of all profiles that were not
28
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
Table 3
Correlations × 100 among measures of aspects of well-being: positive emotionality, negative emotionality, life satisfaction, perceived social support, perceived
health, and a Composite Health Index.
Pos affect
Neg affect
Life satisfaction
Social support
Perceived health
Composite health
Composite health
Perceived health
Social support
Life satisfaction
Neg affect
Pos affect
43
−42
81
77
68
(100)
22
−30
31
23
(100)
37
−23
23
(100)
37
−41
(100)
−9
(100)−
(100)
All correlations are significant at p = 0.000 except positive and negative affects (r = −0.09, p = 0.005).
high in both Self-directedness and Cooperativeness (that is,
profiles 3–8).
Profile-based configural analysis of the impact of each
character trait on the non-affective components of well-being
revealed more information than the analysis of variance.
Taking interactions among the character traits into account,
higher Self-directedness was associated with greater life
satisfaction, perceived social support, and perceived health in
all contrasts (Table 4). Higher Cooperativeness was strongly
associated with greater social support in all contrasts, but had
little or no association with life satisfaction or perceived
health (Table 4). Self-transcendence had little or no association with any measure of non-affective health (Table 4).
4. Overall relationship of health to character profile
The CHI appeared to provide a good summary measure of
perceived non-affective health or wellness whereas the
PANAS provided a good summary measure of perceived
mood or happiness. Descriptive statistics for CHI and PANAS
by character profile are summarized in Table 5. Analysis of
variance showed that the groups differed on the CHI
significantly (F = 48.19, p = 0.000). The means of the different groups for the CHI are shown in Fig. 4 for ease of
inspection. Post-hoc group comparisons using the Bonferroni
range correction showed that profiles with high Selfdirectedness (profiles 1–4) are significantly different from
those with low Self-directedness (profiles 5–8).
Compared to the analysis of variance, profile-based
configural analysis revealed much more information about
the impact of character on health. Taking into account
interactions with other traits, the impact of each character
trait on non-affective and affective aspects of health is
summarized in Table 6. Non-affective health (i.e., “wellness”)
is measured by the CHI, and affective health (i.e., “happiness”)
is measured by the presence of positive emotion and the
absence of negative emotion. For both wellness and happiness, higher Self-directedness was strongly associated with
better health regardless of the other two traits. For wellness,
higher Cooperativeness was significantly associated with
wellness (i.e., higher CHI) in the contrast of organized vs
autocratic character configurations (SCt vs Sct, P = 0.000) and
moody vs disorganized character configurations (sCT vs scT,
p = 0.005), and had trends in the same direction for the other
two configurations (Table 6). Cooperativeness increased
wellness largely by enhancing perceived social support
(Table 5). Self-transcendence increased happiness by increasing positive emotions, but did not significantly increase
wellness or reduce negative emotions (Table 6).
Table 4
T-test for each character trait for well-being, social support, and subjective
health.
Fig. 3. Life satisfaction, subjective health, and social support as a function of
character profile.
Life satisfaction
Social support
Subjective health
t
t
t
p
p
p
Self-directedness
SCT vs sCT
7.228
SCt vs sCt
8.407
ScT vs ScT
7.170
Sct vs sct
7.393
0.000
0.000
0.000
0.000
5.208
3.783
4.237
4.232
0.000
0.000
0.000
0.000
5.485
4.135
4.196
2.643
0.000
0.000
0.000
0.009
Cooperation
SCT vs ScT
SCt vs Sct
sCT vs scT
sCt vs sct
−0.874
2.633
1.56
0.593
0.383
0.009
0.120
0.554
2.758
3. 396
3.285
3.077
0.006
0.001
0.001
0.002
1.196
1.383
1.537
0.082
0.233
0.168
0.125
0.934
Self-transcendence
SCT vs SCt
−1.056
ScT vs Sct
2.129
sCT vs sCt
1.151
scT vs sct
0.558
0.288
0.035
0.251
0.577
0.910
0.800
0.526
0.651
−0.204
−0.102
−0.746
−2.432
0.838
0.919
0.456
0.016
0.113
0.244
−0.635
−0.453
SCT = creative; SCt = organized; ScT = fanatical; Sct= a utocratic; sCT =
moody; sCt = dependent; scT = disorganized; sct = depressive.
29
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
Table 5
Descriptive statistics for CHI positive and negative affects by character
profile.
Character
profile
SCT
SCt
ScT
Sct
sCT
sCt
scT
sct
CHI
Positive affect
Negative affect
Mean
SE
Mean
SE
Mean
SE
.424
.461
.288
.184
−.223
−.201
−.471
−.382
.051
.050
.088
.063
.062
.079
.049
.052
41.167
38.511
40.298
37.464
38.267
36.282
36.546
34.250
.345
.337
.592
.422
.415
.531
.330
.349
20.196
19.716
21.439
20.384
24.810
25.042
26.311
24.799
.398
.389
.683
.487
.479
.612
.381
.403
SCT = creative; SCt = organized; ScT = fanatical; Sct = autocratic; sCT =
moody; sCt = dependent; scT = disorganized; sct = depressive.
5. The influence of character profiles on extremes of wellness
and illness
Table 6
T-test for each character trait for chi, negative and positive affect.
CHI
t
Negative affect
Positive affect
p
t
p
t
p
Self-directedness
SCT vs sCT
8.554
SCt vs sCt
7.542
ScT vs scT
7.091
Sct vs sct
6.574
0.000
0.000
0.000
0.000
−7.776
−8.071
−5.466
−7.03
0.000
0.000
0.000
0.000
6.505
3.441
4.769
5.955
0.000
0.000
0.000
0.000
Cooperation
SCT vs ScT
SCt vs Sct
sCT vs scT
sCt vs sct
0.105
3.643
2.822
1.796
0.105
0.000
0.005
0.074
−1.771
−1. 36
−2.141
0.295
0.078
0.175
0.033
0.768
1.608
2.03
2.889
2.996
0.109
0.043
0.004
0.003
Self-transcendence
SCT vs SCt
−0.566
ScT vs Sct
1.086
sCT vs sCt
−0.219
scT vs sct
−1.105
0.572
0.279
0.827
0.270
1.031
1.53
−0.274
2.345
0.303
0.128
0.784
0.020
6.069
4.645
3.064
4.098
0.000
0.000
0.003
0.000
Adaptive functions are often best revealed at the extremes
of a complex dynamic system, so we examined the associations
of character profiles with the extremes of wellness and illness.
The top sixth and the bottom sixth of the distribution of the CHI
were selected and dubbed “best health” and “worst ill-health”
respectively. Then the people in each character profile group
were compared for the proportion that had “best health” and
also “worst health”. The profile groups differed significantly in
the proportion that had extremely good health (Chi Squared
142.22, df = 7, p = 0.0000) and extremely poor health (Chi
Squared 127.08, df = 7, p = 0.0000).
The percentages with best health and worst health are
depicted in Fig. 5. The results illustrate the strong impact of
Self-directedness on overall health with lesser influences
from the other two character traits. Individuals who with
creative (SCT) or organized (SCt) profiles are frequently in
the best of health, whereas those who are depressive (sct) or
disorganized (scT) are frequently in the worst of ill-health.
Among the other configurations, there was a shift from
predominant good health in fanatical (ScT) and autocratic
SCT = creative; SCt = organized; ScT = fanatical; Sct = autocratic; sCT =
moody; sCt = dependent; scT = disorganized; sct = depressive.
Fig. 4. Mean values of Composite Health Index of the 8 character profile
groups.
Fig. 5. Percentage of people in each character profile who have “best health”
or “worst ill-health”.
(Sct) profiles to predominant ill-health in moody (sCT) and
dependent (sCt) profiles.
In order to quantify the overall linear influence of the
three character variables on happiness and wellness, regression analyses were carried out with the HI or CHI as the
dependent variable predicted by the three character traits.
TCI character explained 45% of the variance in wellness (CHI,
R square 0.449, adjusted R square 0.448, F = 169.4, p = 0.000)
and 36% in wellness (CHI, R square 0.363, adjusted R squared
0.362, F = 208.65, p = 0.000). For HI, the linear effect of Selfdirectedness was strongly dominant (beta = 0.935, t = 5.51,
p = 0.000). For CHI, the linear influence of Self-directedness
was strong (beta = 0.57, t = 20.6, p = 0.000), Cooperativeness
was weak but significant (beta 0.07, t = 2.4, p = 0.016), and
Self-transcendence was negligible (beta 0.002, t = 0.96,
p = 0.923). The results were virtually the same for each
gender separately.
30
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
6. Discussion
Character profiles have a strong association with individual differences in health, including both its non-affective
aspect (i.e., “wellness”) and its affective aspect (i.e., happiness). In order to quantify the different components of health
and well-being, we measured wellness as a composite of
perceived health, perceived social support, and life satisfaction. Likewise we measured “happiness” as the difference
between standard scores of positive emotion minus negative
emotion within each individual. We found that character has
a strong impact on the perception of all aspects of health,
including social, emotional, and physical well-being.
All three dimensions of character measured by the TCI
contribute to individual differences in health. TCI Self-directedness clearly has the strongest impact as a foundation for the
regulation of a person's hopes and desires, which influences all
aspects of both wellness and happiness, consistent with theories
of self-efficacy and self-determination (Cervone, 2004; Ryan
and Deci, 2000). Cooperativeness has a strong impact on
perceptions of social support, which also makes a substantial
impact to increase wellness and reduce negative emotions,
consistent with attachment and social engagement theories
(Bowlby, 1983; Ryan and Deci, 2001). Self-transcendence has a
strong impact on awareness of participation in what is beyond
the individual self, which increases the experience of positive
emotions, but has little or no impact on wellness or negative
emotions, consistent with humanistic and existential theories
(Cloninger et al.; Cloninger, 2004; Jaspers, 1968; Rogers, 1995).
Overall, each aspect of character makes a distinct contribution to well-being that depends strongly on its interactions with
the other dimensions of character. The influences of character
on health are highly non-linear, depending on specific configurations whose influences differ for distinct aspects of health.
These distinct patterns of interactions can be understood in
terms of the psychology that defines the role of each dimension
of character in mental self-government and the regulation of a
person's desires, goals, emotions, thoughts, and values.
Self-directedness is measured by a person's being responsible, purposeful, and resourceful (Cloninger et al., 1993). High
Self-directedness is associated with hopeful self-confidence,
which has a crucial role in recovery from a broad range of
physical and mental disorders (Amering and Schmolke, 2009;
WHO, 2001). Low Self-directedness is a strong indicator of
people with personality disorders, who are typically irresponsible, aimless, and helpless (Svrakic et al., 1993). Health has
been defined as a state of physical, mental, and social wellbeing (WHO, 1946, 2001). The great importance of Selfdirectedness is shown by its strong association with all
indicators of health, including those that are predominantly
mental (like affect and life satisfaction), social, or physical (like
perceived health). Self-directedness alone explained nearly
one-third (i.e., square of 0.57= 32%) of the variance in wellness
and nearly half (i.e., 45%) of the variance in happiness.
Cooperativeness also makes a significant but weak contribution to wellness, explaining only 4% of the variance in
wellness, as measured by CHI, in linear regression analyses.
Cooperativeness is measured by a person's social tolerance,
empathy, and helpfulness. The impact of Cooperativeness is
strong on the perception of social support, which in turn has an
indirect influence on both wellness and happiness.
Self-transcendence appears to have a negligible impact on
health in linear regression analyses. Nevertheless, when the
interactions among character traits are taken into account,
Self-transcendence has a strong and consistent impact on the
presence of positive emotions. In other words, people with
higher Self-transcendence are consistently happier than
those with lower Self-transcendence when the configuration
of the other two character traits is taken into account. Selftranscendence is measured by the capacity of a person to
become absorbed in what they enjoy doing and thereby to
identify with what is beyond their own transient existence. As
a result, people who are higher in Self-transcendence are
more likely to experience life with joyful exaltation (Cloninger, 2004, 2007). Self-transcendence has been found to be
particularly important for people to adapt well when facing
suffering or death (Coward and Reed, 1996). In fact, we
observed more positive emotions when people who are
higher in Self-transcendence (T) are compared to those who
are lower in Self-transcendence (t). The shift toward positive
emotions is observed (see Table 2 and Fig. 2) by comparing
people with creative vs organized profiles (SCT vs SCt, with
average positive affect scores of 41.2 vs 38.5), fanatical vs
autocratic profiles (ScT vs Sct, with positive affect scores of
40.3 vs 37.5), moody vs dependent profiles (sCT vs sCt, with
positive affect scores of 38. 3 vs 36.3), and disorganized vs
depressive profiles (scT vs. sct, with positive affect scores of
36.5 vs 34.3). The level of happiness is greater in each paired
comparison between higher vs lower Self-transcendence and
controlling for the other two traits. This consistent non-linear
impact of Self-transcendence on emotional well-being was
missed entirely by linear regression analysis.
The main limitation of these observations is that they are
based on a cross-sectional sample in which association, but not
causation, can be demonstrated. However, the TCI personality
dimensions have been shown to be antecedent causes of
individual differences in psychopathology and personality
disorders in twin and family studies of vulnerability to
neuropsychiatric disorders (Calvo et al., 2009; Ettelt et al.,
2008; Farmer et al., 2003; Gillespie et al., 2003; Smith et al.,
2008; Zohar et al., 2005). They have also been shown to have
predictive validity in prospective studies in the general
population (Grucza and Goldberg, 2007) and with specific
disorders that have extensive effects on all aspects of health,
such as obesity and other eating disorders (Anderson et al.,
2002; Fassino et al., 2004; Leombruni et al., 2007). In addition,
the present findings are baseline observations in a prospective
longitudinal study, so we will be able to test these initial
observations by prediction of outcome. The second limitation is
that all our measures in these analyses were self-reported and
therefore subject to perceptual bias and cognitive distortion.
Fortunately, we are also conducting psychiatric and medical
interviews as well as physical examinations with laboratory
data in our prospective study, so we will be able to compare
perceived health with objective health findings.
Our findings lend empirical support to the WHO definition of
health as a state of physical, mental, and social well-being, and
more than the absence of disease or infirmity (WHO, 1946). Our
findings also support the description of mental health as a state
of well-being in which people can realize and use their own
abilities, can cope with the stresses of life, work and love, and
can contribute to their communities (WHO, 2001). We found
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
that perceptions of health, social support, and life satisfaction
were positively correlated with one another and with positive
emotions (Table 2). In contrast, each of these measures of health
and happiness was inversely correlated with negative emotions.
The correlations among these measures were only weak to
moderate in strength, so different aspects of health have unique
associations with different components of health.
The interdependence of the components of health makes it
clear that much understanding can be lost when people are
reduced to separate organ systems or when physical, mental,
and social well-being are viewed in isolation (Cloninger, 2004).
For example, TCI personality traits predict as much or more
variability in risk for atherosclerosis than traditional measures
like smoking or cholesterol levels (Hintsanen et al., 2009).
Physical, mental, and spiritual well-being simply cannot be
separated without loss of understanding of human nature,
presumably because each of these components of well-being
has evolved as an adaptive response to the challenges of life in
the long history of human evolution (Cloninger, 2009). Multiple
systems of learning and memory are integrated in human
beings—behavioral conditioning, semantic learning, and selfaware consciousness, so each must be considered as an
interdependent aspect of the whole person's adaptive functioning and well-being (Cloninger, 2009).
Configural analysis of personality profiles provides a useful
way to describe the multidimensional nature of human adaptive
functioning. Configurations of temperament and of character
provide a profile of different dimensions of the whole person,
rather than considering variables that distinguish groups without
any account of their interactions within the individual. Such
person-centered analyses are crucial for understanding development because configurations in a non-linear dynamic system are
“meta-stable” that is, they tend to be relatively stable, even
though they are self-organizing and may transform abruptly in
response to even small changes in conditions (Cloninger et al.,
1997). Furthermore, biological and psychosocial processes
function as non-linear dynamical systems in general and only
approximate the assumptions of linearity when there are
negligible changes in background conditions (Waldrop, 1992).
That is, linearity is typically a local approximation, not a generally
valid characteristic of biopsychosocial systems (Cloninger, 2004;
Wright, 1984). Consequently, linear statistical analyses never
justify any conclusions about individual people because even
small changes in other unmeasured or uncontrolled variables can
have a strong impact on results, as we demonstrated here by
doing analyses with and without assuming linearity. We clearly
found that well-being depends on specific interactions that are
strong and predictable from an understanding of the non-linear
dynamics of personality development.
Role of funding source
The US–Israel Binational Science Foundation provided funds only and
monitors their administration, but otherwise has no role in the analysis or
report.
Conflict of interest
The authors declare no conflict of interests.
Acknowledgement
The authors acknowledge the support of the United States
of America–Israel Binational Science Foundation.
31
References
Amering, M., Schmolke, M., 2009. Recovery in Mental Health: World
Psychiatric Association Evidence and Experience in Psychiatry. John
Wiley & Sons, New York.
Anderson, C.B., Joyce, P.R., Carter, F.A., et al., 2002. The effect of cognitive–
behavioral therapy for bulimia nervosa on temperament and character
as measured by the temperament and character inventory. Compr.
Psychiatry 43, 182–188.
Bowlby, J., 1983. Attachment Second Edition: Attachment and Loss. Basic
Books, New York.
Brooks, D., The Tel Aviv Cluster, New York Times, January 11, 2010.
Calvo, R., Lazaro, L., Castro-Fornieles, J., et al., 2009. Obsessive-compulsive
personality disorder traits and personality dimensions in parents of
children with obsessive-compulsive disorder. Eur. Psychiatry 24, 201–206.
Cassell, E.J., 1999. Diagnosing suffering: a perspective. Ann. Intern. Med. 131,
531–534.
Central Bureau of Statistics, 2007. Tables 2.6 and 8.3. http://www.cbs.gov.il/
Accessed November 14 2007.
Cervone, D., 2004. The architecture of personality. Psychol. Rev. 111, 183–204.
Cervone, D., 2005. Personality architecture: within-person structures and
processes. Annu. Rev. Psychol. 56, 423–452.
Cicchetti, D., Rogosch, F.A., 1996. Equifinality and multifinality in developmental psychopatology. Dev. Psychopathol. 8, 597–600.
Cloninger, C.R. (Ed.), 1999. Personality and Psychopathology (American Psychopathological Association Series). American Psychiatric Press, Washington, D.C.
Cloninger, C.R., 2004. Feeling Good: the Science of Well-Being. Oxford
University Press, New York.
Cloninger, C.R., 2006. The science of well-being: an integrated approach to
mental health and its disorders. World Psychiatry 5, 71–76.
Cloninger, C.R., 2007. Spirituality and the science of feeling good. South. Med.
J. 100, 740–743.
Cloninger, C.R., 2008. The psychobiological theory of temperament and
character: comment on Farmer and Goldberg (2008). Psychol. Assess. 20,
292–299 (discussion 300-294).
Cloninger, C.R., 2009. The evolution of human brain functions: the functional
structure of human consciousness. Aust. NZ J. Psychiatry 43, 994–1006.
Cloninger, C.R., Abou-Saleh, M.T., Mrazek, D.A., et al., Biological perspective
on psychiatry for the person. Psychopathology.
Cloninger, C.R., Svrakic, D.M., Przybeck, T.R., 1993. A psychobiological model
of temperament and character. Arch. Gen. Psychiatry 50, 975–990.
Cloninger, C.R., Svrakic, N.M., Svrakic, D.M., 1997. Role of personality selforganization in development of mental order and disorder. Dev.
Psychopathol. 9, 881–906.
Cloninger, C.R., Bayon, C., Svrakic, D.M., 1998. Measurement of temperament
and character in mood disorders: a model of fundamental states as
personality types. J. Affect. Disord. 51, 21–32.
Coward, D.D., Reed, P.G., 1996. Self-transcendence: a resource for healing at
the end of life. Issues Ment. Health Nurs. 17, 275–288.
Diener, E., Biswas-Diener, R., 2008. Happiness: Unlocking the Secrets of
Psychological Wealth. Blackwell Publishing, Malden, MA.
Ettelt, S., Grabe, H.J., Ruhrmann, S., et al., 2008. Harm avoidance in subjects
with obsessive-compulsive disorder and their families. J. Affect. Disord.
107, 265–269.
Farmer, A., Mahmood, A., Redman, K., et al., 2003. A sib-pair study of the
temperament and character inventory scales in major depression. Arch.
Gen. Psychiatry 60, 490–496.
Fassino, S., Amianto, F., Gramaglia, C., et al., 2004. Temperament and character in
eating disorders: ten years of studies. Eat. Weight Disord. 9, 81–90.
Gillespie, N.A., Cloninger, C.R., Heath, A.C., et al., 2003. The genetic and
environmental relationship between Cloninger's dimensions of temperament and character. Pers. Individ. Differ. 35, 1931–1946.
Grucza, R.A., Goldberg, L.R., 2007. The comparative validity of 11 modern
personality inventories: predictions of behavioral acts, informant
reports, and clinical indicators. J. Pers. Assess. 89, 167–187.
Hintsanen, M., Pulkki-Raback, L., Juonala, M., et al., 2009. Cloninger's
temperament traits and preclinical atherosclerosis: the Cardiovascular
Risk in Young Finns Study. J. Psychosom. Res. 67, 77–84.
Jaspers, K., 1968. The phenomenological approach in psychopathology. Br. J.
Psychiatry 114, 1313–1323.
Kahneman, D., Diener, E., Schwartz, N. (Eds.), 2003. Well-Being: the Foundations
of Hedonic Psychology. Russell Sage Foundation Publications, New York.
Keyes, C.L., Shmotkin, D., Ryff, C.D., 2002. Optimizing well-being: the empirical
encounter of two traditions. J. Pers. Soc. Psychol. 82, 1007–1022.
Knauper, B., Turner, P.A., 2003. Measuring health: improving the validity of
health assessments. Qual. Life Res. 12 (Suppl 1), 81–89.
Leombruni, P., Piero, A., Dosio, D., et al., 2007. Psychological predictors of
outcome in vertical banded gastroplasty: a 6 months prospective pilot
study. Obes. Surg. 17, 941–948.
32
C.R. Cloninger, A.H. Zohar / Journal of Affective Disorders 128 (2011) 24–32
Manderscheid, R.W., Ryff, C.D., Freeman, E.J., et al., 2010. Evolving definitions
of mental illness and wellness. Prev. Chron. Dis. 7, A19.
Moldin, S.O., Scheftner, W.A., Rice, J.P., et al., 1993. Association between major
depressive disorder and physical illness. Psychol. Med. 23, 755–761.
Pavot, W., Diener, E., 1993. Review of the satisfaction with life scale. Psychol.
Assess. 5, 149–161.
Pease, S.L., 2009. The Golden Age of Jewish Achievement: the Compendium of a
culture, a people, and their stunning performance. Deucalion, Sonoma
Valley, USA.
Rogers, C.R., 1995. A Way of Being. Houghton Mifflin, Boston.
Ruini, C., Ottolini, F., Rafanelli, C., et al., 2003. The relationship of psychological
well-being to distress and personality. Psychother. Psychosom. 72, 268–275.
Ryan, R.M., Deci, E.L., 2000. Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. Am. Psychol.
55, 68–78.
Ryan, R.M., Deci, E.L., 2001. On happiness and human potentials: a review of
research on hedonic and eudaimonic well-being. Annu. Rev. Psychol. 52,
141–166.
Ryff, C.D., Keyes, C.L., 1995. The structure of psychological well-being
revisited. J. Pers. Soc. Psychol. 69, 719–727.
Ryff, C.D., Singer, B.H., Dienberg Love, G., 2004. Positive health: connecting wellbeing with biology. Philos. Trans. R. Soc. Lond. B Biol. Sci. 359, 1383–1394.
Ryff, C.D., Dienberg Love, G., Urry, H.L., et al., 2006. Psychological well-being
and ill-being: do they have distinct or mirrored biological correlates?
Psychother. Psychosom. 75, 85–95.
Schmutte, P.S., Ryff, C.D., 1997. Personality and well-being: reexamining
methods and meanings. J. Pers. Soc. Psychol. 73, 549–559.
Senor, D., Singer, S., 2009. Start-up Nation: the Story of Israel's Economic
Miracle, First Edition. Twelve: The Hachette Book Group, New York.
Smith, M.J., Cloninger, C.R., Harms, M.P., et al., 2008. Temperament and
character as schizophrenia-related endophenotypes in non-psychotic
siblings. Schizophr. Res. 104, 198–205.
Svrakic, D.M., Whitehead, C., Przybeck, T.R., et al., 1993. Differential diagnosis
of personality disorders by the seven-factor model of temperament and
character. Arch. Gen. Psychiatry 50, 991–999.
Waldrop, M.M., 1992. Complexity: the Emerging Science at the Edge of Order
and Chaos. Simon & Schuster, New York.
Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief
measures of positive and negative affect: the PANAS scales. J. Pers. Soc.
Psychol. 54, 1063–1070.
WHO, 1946. Definition of Health in Preamble to the Constitution of the World
Health Organization, (2). World Health Organization, Geneva, Switzerland.
WHO, 2001. Mental Health: New Understanding, New Hope. World Health
Organization, Geneva.
Wright, S., 1984. Evolution and the Genetics of Populations: Genetics and
Biometric Foundations, Vol. One. University of Chicago Press, Chicago.
Zimet, G.D., Powell, S.S., Farley, G.K., Werkman, S., Berkoff, B.K., 1990.
Psychometric characteristics of the Multidimensional Scale of Perceived
Social Support. J. Pers. Assess. 55, 610–617.
Zohar, A.H., Cloninger, C.R., 2010. The psychometric properties of the TCI-140
in Hebrew. Eur. J. Psychol. Assess.
Zohar, A.H., Ebstein, R.P., Pauls, D.L., 2005. TPQ profiles of patients with OCD
and GTS and their first degree relatives. World J. Biol. Psychiatry 6,
151–152.