Why Does Writing About Important Values Reduce Defensiveness? Self-Affirmation and the Role of Positive Other-Directed Feelings

Abstract

Previous research has repeatedly shown that writing about an important value, compared with writing about an unimportant value, reduces defensiveness in response to self-threatening information, but has not identified why. Study 1 showed that participants who wrote about an important value reported more positive other-directed feelings, such as love and connection, than participants who wrote about an unimportant value. Study 2 replicated this effect, and showed that loving and connected feelings, but not positive or negative self-directed feelings, completely accounted for the effect of a values-affirmation manipulation on smokers' acceptance of information indicating that smoking harms health. These studies, in concert with previous research, suggest that values affirmation reduces defensiveness via self-transcendence, rather than self-integrity (i.e., self-worth or self-images).

Full text

VALUES AND DEFENSIVENESS
Running head: VALUES AND DEFENSIVENESS
Why Does Writing about Important Values Reduce Defensiveness?
Self-Affirmation and the Role of Positive, Other-Directed Feelings
Jennifer Crocker and Yu Niiya
University of Michigan
Dominik Mischkowski
Universitaet Konstanz
January 8, 2008
Word count: 4000 words
Citation: Crocker, J., Niiya, Y., & Mischkowski, D. (2008). Why does writing about important values
reduce defensiveness? Self-affirmation and the role of positive, other-directed feelings.
Psychological Science, 19, 740-747. PMID 18727791
1

VALUES AND DEFENSIVENESS
Abstract
Previous research has repeatedly shown that writing about an important value reduces
defensiveness in response to self-threatening information, compared to writing about an
unimportant value, but has not identified why. Study 1 showed that writing about an important
value induces positive, other-directed feelings such as love and connection compared to a control
condition. Study 2 replicated this effect, and showed that loving and connected feelings, but not
positive or negative self-directed feelings, completely accounted for the effect of the values-
affirmation manipulation on smokers’ acceptance of information that smoking harms health.
These studies, in concert with previous research, suggest that values-affirmation reduces
defensiveness via transcending the self, rather than self-integrity (i.e., self-worth or self-images).
2

VALUES AND DEFENSIVENESS
Few people appreciate learning that their beliefs are erroneous, their behavior contradicts
their values, or they have caused harm to themselves or others. Although such information can
be useful, even life-saving, it can also threaten people’s views of themselves as rational,
intelligent, responsible, healthy, or caring. Consequently, people often respond defensively to
such self-threatening information (Cooper & Fazio, 1984; Pronin, Gilovich, & Ross, 2004).
Defensiveness presents a major obstacle to persuasive communications intended to encourage
people to change harmful or unhealthy behaviors. For example, caffeine drinkers are more likely
than non-caffeine drinkers to find fault with studies indicating a link between caffeine
consumption and health problems (e.g., Harris & Napper, 2005; Sherman, Nelson, & Steele,
2000).
Self-affirmation manipulations reliably reduce defensiveness (see McQueen & Klein,
2006; Sherman & Cohen, 2006; Steele, 1988, for reviews). In particular, people are more
accepting of self-threatening information when they have the opportunity to write a few
sentences about their most important value (see McQueen & Klein, 2006, for a review). In these
studies, people rank-order a list of values (e.g., politics, business, art, religion, social
relationships) from most important to least important, and either write a paragraph about their
most important value and why it is important to them (experimental condition), or their least
important value and why it might be important to others (control condition). Under the guise of
an unrelated study, participants read a bogus article describing negative health consequences of a
behavior such as drinking coffee. Such studies have repeatedly found that people for whom the
information is self-relevant (e.g., coffee drinkers) are less defensive (i.e., less skeptical, less
critical of research methods, more likely to believe conclusions are justified by the data) when
3

VALUES AND DEFENSIVENESS
they have first had the opportunity to write about their most important value, compared to when
they write about their least important value (see Sherman & Cohen, 2006, for a review).
According to self-affirmation theory, writing about values reduces defensiveness because
it affirms the integrity of the self. Steele proposed that people have a self-system that maintains
self-integrity, “a phenomenal experience of the self—self-conceptions and self-images—as
adaptively and morally adequate, that is, as competent, good, coherent, unitary, stable, capable of
free choice, capable of controlling important outcomes, and so on” (Sherman & Cohen, 2006;
Steele, 1988). As Sherman and Cohen (2006, p. 185) note, “When this image of self-integrity is
threatened, people respond in such a way as to restore self-worth.” Thus, according to self-
affirmation theory, reflecting on important values reduces defensiveness by boosting self-images
or self-worth.
Although many studies have demonstrated that writing about important values reduces
defensiveness, research has not identified the mechanism for this effect. Most attempts to
identify a mechanism have focused on how people feel or think about themselves. The most
direct prediction from self-affirmation theory suggests that writing about important values raises
state self-esteem. However, to date research has provided little support for self-esteem as a
mediating mechanism (McQueen & Klein, 2006; Sherman & Cohen, 2006). Studies that include
state self-esteem as a dependent measure have not found that writing about important values
boosts self-esteem relative to the control condition (Schmeichel & Martens, 2005). One study
found that compared to participants in a control condition, people who wrote about an important
value rated the initial letters in their names favorably, indicating higher implicit self-esteem
(Koole, Smeets, van Knippenberg, & Dijksterhuis, 1999). However, preference for initial name
letters did not account for the effect of writing about important values on rumination.
4

VALUES AND DEFENSIVENESS
Self-affirmation theory also predicts that affirming important values boosts self-images or
self-concepts. Steele and Liu (1983) found that participants who completed a scale of economic
and political values rated themselves more positively if this value was important to them than if
it was not important. They did not, however, show that the effects of values scales on self-
concept ratings accounted for the effect of affirming values, nor did they rule out the possibility
that participants with political-economic values had more positive self-concepts before the
values-affirmation.
Other researchers have proposed that self-affirmation manipulations such as writing
about or reflecting on an important value reduce negative affect or increase positive affect
(Tesser, 2000). Several studies have tested this hypothesis; results have been mixed, with some
studies showing no effect on mood, and others showing a positive effect (McQueen & Klein,
2006; Sherman & Cohen, 2006). Evidence that mood or affect mediates the effect of self-
affirmation manipulations is scarce; one study found that a measure of implicit positive affect
explained the effects of values-affirmation on reduced rumination (Koole et al., 1999).
Overall, research clearly indicates that affirming important values reduces defensiveness;
how it does so remains an open question (McQueen & Klein, 2006; Sherman & Cohen, 2006).
Transcending the Self
We propose that reflecting and writing about important values enables people to
transcend concerns about self-image or self-worth. Writing essays about important values
reminds people what they care about beyond themselves, and may induce positive, other-directed
feelings such as love. Consistent with the broaden-and-build theory of positive emotions
(Fredrickson, 1998), love may elevate and inspire people to improve, opening them to potentially
threatening information (Haidt, 2003). Previous research on self-affirmation has not typically
5

VALUES AND DEFENSIVENESS
examined other-directed positive emotions, because self-affirmation theory focuses on self-
image and self-feelings as the central construct.
This line of reasoning suggests two possible patterns of emotion and reduced
defensiveness in values-affirmation studies. First, writing about an important value may reduce
defensiveness only among people who rank relationships as their most important value.
Visualizing a close positive relationship induces warm affect for the relational partner, leading to
increased interest in negative feedback about intellectual ability (Kumashiro & Sedikides, 2005).
Alternatively, writing about any important value may remind people of what they care about
beyond themselves, leading to feelings of love, and enabling them to transcend the self, reducing
defensiveness.
Study 1
Study 1 manipulated values affirmation and examined whether it induces positive other-
directed emotion, especially love.
Method
Participants
College students (N =139; 56% female, 71% White, 14% Asian) received partial credit for
their introductory psychology class for participating in a “study of values.” Their age ranged from
17 to 21 (M = 18.8, SD = .76).
Procedure
In groups of up to 10, participants gave their informed consent. Following the procedures
of Sherman, Nelson, and Steele (2000), participants ranked 6 values (business, art/music/theater,
social life/relationships, science /pursuit of knowledge, religion /morality, government /politics)
in order of their personal importance. Next, participants opened an envelope containing either the
6

VALUES AND DEFENSIVENESS
values-affirmation instruction or the control instruction, ensuring that the experimenter was
unaware of participants’ condition. In the values-affirmation condition (n = 69), participants
wrote about their most important value and why it was important and meaningful to them for 10
min while those in the control condition (n = 70) wrote about their least important value and why
it might be important and meaningful to others.
Participants rated how much they felt 18 feelings when writing their essay on a scale
from 1 (not at all) to 5 (extremely). The items are presented in Figure 1. At the conclusion of the
study, participants were debriefed.
Results and Discussion
Social life was ranked by 41% of males and 54% of females as their most important
value, followed by religion (23% of males, 31% of females) and science (15% of males, 6% of
females). Business, art, or government was chosen by 21% of males and 10% of females. Figure
1 presents the mean difference between the values-affirmation and control conditions for each
feeling, in order of the size of the difference. The two conditions differed for all feelings except
selfish, scared, sad, confused, and angry, ts > 2.5, prep > .95, ds > 0.45. All six positive other-
directed feelings (love, giving, empathic, connected, sympathy, and grateful) showed large
differences between conditions. As expected, ratings of love differed most between the two
conditions, with a mean difference of 2.26, t(135) = 11.47, prep > .99, d = 1.97. Males (2.34,
t(58) = 8.76, prep > .99, d = 2.25) and females (2.15, t(75) = 7.95, prep > .99, d = 1.80) both
showed this effect (F(1, 133) = .22, n.s., for the interaction).
Values-affirmation had greater effect on love than on a composite of general positive
feelings (proud, content, and joyful, α= .84). A 2 (Emotion) X 2 (Values-Affirmation Condition)
repeated-measures ANOVA showed a significant Emotion X Values-Affirmation interaction,
7

VALUES AND DEFENSIVENESS
F(1, 135) = 35.35, prep > .99, η2p = .21, indicating that the values-affirmations had greater effect
on love (Maffirmation = 4.24, SD = 1.04 vs. Mcontrol = 1.97, SD = 1.26) than on general positive
feelings (Maffirmation = 3.74, SD = .82 vs. Mcontrol = 2.48, SD = .99).
Love was greater in the values-affirmation than in the control condition regardless of
which value participants ranked as most important (see Table 1). A 2 (Values-Affirmation
Condition) X 4 (Value Choice) ANOVA showed significant main effects on love for values-
affirmation (F(1, 129) = 87.95, prep > .99, η2p = 0.41) and value choice (F(3, 129) = 7.74, prep > .
99, η2p = 0.15) but no interaction (F(1, 129) = .59, n.s.).
In sum, Study 1 showed that writing about an important value compared to writing about
an unimportant value increases feelings of love, regardless of which value participants rated as
most important, and for both males and females. These results contrast with previous research
showing inconsistent effects of values-affirmation on positive affect (e.g., Cohen, Aronson, &
Steele, 2000), which did not examine positive other-directed feelings. In Study 2, we tested the
hypothesis that love accounts for increased acceptance of potentially self-threatening
information.
Study 2
Method
Participants
8

VALUES AND DEFENSIVENESS
One-hundred two undergraduate students (97% female; 70% White; 12% Asian)
participated in exchange for partial credit toward their introductory psychology class. Their age
ranged from 17 to 22 (M = 18.6, SD = 1.02).
Procedure
At the beginning of the term, participants completed a mass testing survey, which
included a question about smoking (“How many cigarettes do you typically smoke each day?”).
Seventy one nonsmokers and 25 smokers (at least one cigarette per day) were recruited for a
subsequent study (the unequal ns reflect the low number of smokers in the sample). Four
additional participants who did not complete the prescreening survey were included in the non-
smokers group and two in the smoking group based on their responses to the post-test survey,
resulting in an n of 75 non-smokers and 27 smokers. The experimenter was blind to participants’
smoking status.
Upon arrival, participants signed consent forms for two short ostensibly unrelated studies,
one on values and the other on how people evaluate scientific articles. In the first study,
participants completed the values-affirmation manipulation as in Study 1, and rated how much
writing the essay made them feel: loving, connected, empathic, humble, proud, vulnerable,
admirable, superior, powerful, strong, in control, victimized, ashamed, inferior, powerless, weak,
and out of control.
For “the second study,” participants read a 3-page fake research article describing the
increased risk of developing abdominal aortic aneurysm among smokers (Dillard, McCaul, &
Randall, 2006, May). After reading the article, participants answered questions adapted from
Sherman et al. (2000) assessing acceptance of the research findings: 1) How skeptical were you
when reading the article? (1 = Not at all; 9 = To a great extent; reverse scored); 2) To what
9

VALUES AND DEFENSIVENESS
extent do you think that the conclusion of the article was justified on the basis of existing
research findings? (1 = not at all; 9 = very well justified). Participants then rated the author’s
description of the findings of the study (1 = distorted and misleading to 9 = well done and quite
accurate), and rated the methods and results of the study (1= very weak to 9 = very strong). We
averaged these items to create an acceptance measure (α = .89). Finally, participants indicated
how many cigarettes they typically smoke per day, were probed for suspicion, and were fully
debriefed.
Results and Discussion
Most participants (66.7%) chose social life as their most important value, followed by
religion (14.7%) and science (11.8%). Less than 3% chose each of the other values (business, art,
and government).
Emotions
Consistent with Study 1, values-affirmed participants reported significantly greater
positive affect (see Figure 2). Again, the effect was largest for ratings of loving, which differed
so much between the two conditions (2.66, t(100) = 13.85, prep > .99, d = 2.77) that the
distribution violated assumptions of normality (see Figure 3).
To examine whether values-affirmation influenced feeling loving more than self-directed
positive feelings, we conducted a 2 (Emotion) X 2 (Values-Affirmation Condition) repeated-
measures ANOVA, comparing loving with a composite score of proud, strong, and admirable (α
= .85). Consistent with Study 1, participants reported feeling loving (M = 2.99, SD = 1.64) more
than self-directed positive feelings (M = 2.74, SD = 1.20; F(1,100) = 4.20, prep >.88, η2p = .04).
Moreover, the Emotion X Condition interaction was significant, F(1, 100) = 19.40, prep >.99, η2p
= .16, indicating that values affirmation had greater impact on loving (Maffirmation = 4.24, SD = 1.05
10

VALUES AND DEFENSIVENESS
vs. Mcontrol = 1.58, SD = .87) than on self-directed positive feelings (Maffirmation = 3.54, SD = .83 vs.
Mcontrol = 1.84, SD = .88).
Acceptance of the Article
Consistent with past findings, writing about one’s most important value increased
acceptance of the article, but only among smokers, for whom the article was relevant (Values-
Affirmation X Relevance, F(1, 98) = 4.35, prep = .89, η2p = 0.04). Neither main effect was
significant, Fs(1, 98) < 2.2. Values-affirmed smokers were more accepting (M = 6.80, SD =
1.23) than control smokers (M = 5.45, SD = 1.84; t(25) = 2.11, prep = .89, d = 0.86); among non-
smokers, acceptance did not differ by condition (Maffirmation = 6.02, SD = 1.44; Mcontrol = 6.26; SD =
1.94, t(73) = .60, n.s.). In the affirmation condition, smokers (M = 6.80, SD = 1.23) were more
accepting than nonsmokers (M = 6.02, SD = 1.44), although this difference was not robust (t(52)
= -1.63, prep = .81, d = 0.58).
Mediation
Loving explained the relationship between values-affirmation and acceptance of
threatening information for smokers but not for non-smokers. As shown in Figure 4, values-
affirmation significantly increased loving among smokers, which in turn, predicted an increase in
acceptance. When loving was included in the regression analysis predicting acceptance, the
effect of values-affirmation on acceptance dropped from β = .39 to β = .11, n.s., indicating that
loving fully accounts for the effect of values-affirmation on acceptance of threatening
information.
We also tested each other emotion as a mediator among smokers. Values affirmed
smokers felt more connected (β = .74, prep > .99, r2 = .54) and proud (β = .40, prep = .89, r2 = .16)
and marginally more strong (β = .35, prep = .85, r2 = .12) and less ashamed (β = -.35, prep = .84, r2
11

VALUES AND DEFENSIVENESS
= .12). Feeling connected predicted acceptance (β = .41, prep = .89, r2 = .17). Controlling for
feeling connected, values affirmation no longer predicted acceptance (β decreased from .39, prep
= .89 to .14, prep = .40), indicating mediation. The other emotions did not predict acceptance (preps
< .79). Thus, only loving and connected explained the effect of values affirmation on reduced
defensiveness among smokers. When entered into a regression equation predicting acceptance, a
composite of loving and connected predicted acceptance, β = .49, prep >.95, f2 = 0.32, and reduced
the effect of the manipulation on acceptance to β = -.01, prep = .11, f2 = 0, suggesting complete
mediation.
Non-smokers showed no evidence of mediation. Although values-affirmed non-smokers
felt more loving, connected, empathic, humble, vulnerable, powerful, strong, proud, admirable,
and in control (βs > .30, prep > .99, r2 > .12) and less inferior, powerless, and ashamed (βs < -.20,
prep > .88, r2 > .06), values-affirmation did not increase acceptance among non-smokers (β = -.07,
n.s.). None of the emotions predicted nonsmokers’ acceptance of the article (-.22 < βs < .07, n.s.)
except out-of-control, which significantly predicted reduced acceptance (β = -.26, prep > .88, r2
=.07) but was not affected by values affirmation.
In sum, Study 2 again showed that writing about important values leads to feeling loving
more than writing about unimportant values. Study 2 also replicated findings from previous
research that the values-affirmation manipulation increases acceptance of self-threatening
information. Most important, Study 2 showed that feeling loving explained the effect of the
values-affirmation manipulation on defensiveness. Feeling connected, but none of the other
emotions, also mediated the effect.
General Discussion
12

VALUES AND DEFENSIVENESS
Previous research has repeatedly shown that writing about an important value reduces
defensiveness, compared to writing about an unimportant value (Sherman & Cohen, 2006). Self-
affirmation theory claims that writing about important values affirms the integrity of the self,
boosting self-images (Steele & Liu, 1983) or self-worth (Sherman & Cohen, 2006). However,
this research has failed to establish why this manipulation reduces defensiveness.
The results of Study 1 demonstrated that people who write about their most important
value report feeling more love than people who write about their least important value,
regardless of which value they ranked as most important. Although women reported feeling more
love, the values affirmation manipulation increased feelings of love in men as much as women.
Study 2 replicated the effect of the values-affirmation manipulation on feeling loving and
showed that this effect was stronger than for several other positive emotions, including self-
relevant emotions such as pride. Furthermore, Study 2 showed that loving accounted for the
effect of the values-affirmation manipulation on smokers’ acceptance of information that
smoking harms health, although these correlational analyses cannot demonstrate causality. Of
the other emotions included, only feeling connected also explained the effect of values-
affirmation on acceptance.
These findings call for a reconsideration of self-affirmation theory. Like looking for a lost
item under the lamppost because that is where the light is, self-affirmation researchers have
looked for mediators in self-feelings, because that is the focus of self-affirmation theory. Our
results, in concert with previous research, suggest that values-affirmation manipulates feelings of
caring for other people or things, rather than self-worth or self-images.
We suggest that values-affirmation manipulations remind participants of people or things
they care about beyond themselves that are more important than temporary feelings of self-
13

VALUES AND DEFENSIVENESS
threat; rather than affirming the self, values affirmation enables people to transcend the self.
Feeling loving may shift people’s goals from preserving positive self-images to concern for
something that transcends the self. The strong effects on love raise the possibility that values-
affirmation manipulations affect care-giving hormones, such as oxytocin (Gonzaga, Turner,
Keltner, Campos, & Altemus, 2006). Intranasal administration of oxytocin increases trust (Zak,
Kurzban, & Matzner, 2004) and down-regulates stress-related cortisol responses (Henry &
Wang, 1998). The present results raise the intriguing possibility that oxytocin may increase
learning of potentially self-threatening information. Down-regulation of the fight-or-flight
response affords people the ability to protect others from threats (Brown & Brown, 2006; Henry
& Wang, 1998). Such a system may enable people to process potentially threatening information
rather than protecting the self.
Although writing about an important value induced feelings of love in men as well as
women in Study 1, Study 2 did not include enough men to test whether gender moderates the
effect of self-affirmation on acceptance of self-threatening information. Feelings of love might
foster acceptance more in women than men, due to sex differences in oxytocin processing
(Carter, 2007). Future research should examine gender differences.
Self-affirmation theory has spawned a host of manipulations, from values-affirmation to
positive feedback, each of which presumably affirms self-integrity (McQueen & Klein, 2006).
Different manipulations may operate via different mechanisms. The present studies identify one
mediator of a values-affirmation manipulation in the health domain, feeling love; the operative
mechanism may differ by the specific manipulation and the domain of the threat. Future research
identifying the active ingredients of various self-affirmation manipulations will ultimately
14

VALUES AND DEFENSIVENESS
advance self-affirmation theory, clarifying why these different manipulations reduce
defensiveness.
Conclusion
Self-affirmation theory has inspired a great deal of research showing that self-affirmation
can reduce defensive responses to self-threatening information. Yet, the mechanism for this
effect has been mysterious. The present studies show that one self-affirmation manipulation,
writing about important values, has powerful effects on positive, other-directed feelings,
especially love, and these feelings can explain why this manipulation reduces defensiveness.
These studies raise the prospect that reminding people what they love or care about may enable
people to transcend the self and foster learning under difficult circumstances.
15

VALUES AND DEFENSIVENESS
References
Brown, S. L., & Brown, R. M. (2006). Selective investment theory: Recasting the functional
significance of close relationships. Psychological Inquiry, 17, 1-29.
Carter, C. S. (2007). Sex differences in oxytocin and vasopressin: Implications for autism
spectrum disorders? Behavioural Brain Research, 176, 170-186.
Cohen, G. L., Aronson, J., & Steele, C. M. (2000). When beliefs yield to evidence: Reducing
biased evaluation by affirming the self. Personality and Social Psychology Bulletin, 26,
1151-1164.
Cooper, J., & Fazio, R. H. (1984). A new look at dissonance theory. Advances in experimental
social psychology, 17, 229-266.
Dillard, A. J., McCaul, K. D., & Randall, A. (2006, May). Positive mood fails to enhance
processing and openness to health information. Paper presented at the Association for
Psychological Science, New York.
Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psychology, 2,
300-319.
Gonzaga, G. C., Turner, R. A., Keltner, D., Campos, B., & Altemus, M. (2006). Romantic love
and sexual desire in close relationships. Emotion, 6, 163-179.
Haidt, J. (2003). The moral emotions. In R. J. Davidson, K. R. Scherer & H. H. Goldsmith
(Eds.), Handbook of affective sciences (pp. 852-870). Oxford: Oxford University Press.
Harris, P. R., & Napper, L. (2005). Self-affirmation and the biased processing of threatening
health-risk information. Personality and Social Psychology Bulletin, 31, 1250-1263.
Henry, J. P., & Wang, S. (1998). Effects of early stress on adult affiliative behavior.
Psychoneuroendocrinology, 23, 863-875.
16

VALUES AND DEFENSIVENESS
Koole, S. L., Smeets, K., van Knippenberg, A., & Dijksterhuis, A. (1999). The cessation of
rumination through self-affirmation. Journal of Personality and Social Psychology, 77,
111-125.
Kumashiro, M., & Sedikides, C. (2005). Taking on board liability-focused information: Close
positive relationships as a self-bolstering resource. Psychological Science, 16, 732-739.
McQueen, A., & Klein, W. M. P. (2006). Experimental Manipulations of Self-Affirmation: A
Systematic Review. Self and Identity, 5, 289-354.
Pronin, E., Gilovich, T., & Ross, L. (2004). Objectivity in the Eye of the Beholder: Divergent
Perceptions of Bias in Self Versus Others. Psychological Review, 111, 781-799.
Schmeichel, B. J., & Martens, A. (2005). Self-Affirmation and mortality salience: Affirming
values reduces worldview defense and death-thought accessibility. Personality and
Social Psychology Bulletin, 31, 658-667.
Sherman, D. K., & Cohen, J. L. (2006). The psychology of self-defense: Self-affirmation theory.
In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 38, pp. 183-
242). San Diego: Academic Press.
Sherman, D. K., Nelson, L. D., & Steele, C. M. (2000). Do messages about health risks threaten
the self? Increasing the acceptance of threatening health messages via self-affirmation.
Personality and Social Psychology Bulletin, 26, 1046-1058.
Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In
L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 261-302).
New York: Academic Press.
Steele, C. M., & Liu, T. J. (1983). Dissonance processes as self-affirmation. Journal of
Personality and Social Psychology, 45, 5-19.
17

VALUES AND DEFENSIVENESS
Tesser, A. (2000). On the confluence of self-esteem maintenance mechanisms. Personality and
Social Psychology Review, 4, 290-299.
Zak, P. J., Kurzban, R., & Matzner, W. T. (2004). The neurobiology of trust. Annual
Proceedings of the New York Academy of Sciences, 1032, 224-227.
18

VALUES AND DEFENSIVENESS
Authors’ Note
Jennifer Crocker and Yu Niiya, Department of Psychology, University of Michigan.
Dominik Mischkowski, Universitaet Konstanz, Germany.
This research was supported by National Institute of Mental Health grant R01 MH58869
and National Science Foundation grant BCS-0446567.
We thank Tim Cavnar, Andrew Crocker, Paul Denning, and Victoria Parver for
assistance with data collection and entry, and David Sherman, Geoffrey Cohen, and Steven
Spencer for their comments on a previous draft.
Address correspondence to Jennifer Crocker, Department of Psychology, University of
Michigan, 530 Church St., Ann Arbor, Michigan 48109, jcrocker@umich.edu
19

VALUES AND DEFENSIVENESS
Table 1
Means and Standard Deviations of Love by Value-Affirmation Conditions and by Value Choice
Most Important Values M SD M SD t
Social life 4.43 .73 2.10 1.37
t(64) = 8.87, p
rep
> .99
Religion/ morality 4.63 .50 2.23 1.38
t(36) = 6.63, p
rep
> .99
Science/ pursuit of knowledge 2.67 1.51 1.14 .38
t(11) = 2.60, p
rep
> .90
Other (business, art/ music,
government / politics)
3.78 1.48 1.64 .81
t(18) = 4.12, p
rep
> .99
Values- Affirmation
Control
20

VALUES AND DEFENSIVENESS
Figure Captions
Figure 1. Ratings of feelings by condition in Study 1. Vertical lines depict standard errors of the
means.
Figure 2. Ratings of feelings by condition in Study 2. Vertical lines depict standard errors of the
means.
Figure 3. Distribution of loving by condition in Study 2.
Figure 4. Mediation model among smokers: Loving mediating the effect of values-affirmation
on acceptance of threatening information. Note. Statistics in parentheses represent beta
coefficients and prep values when values-affirmation and loving are entered simultaneously in the
regression.
21

VALUES AND DEFENSIVENESS
22

VALUES AND DEFENSIVENESS
23

VALUES AND DEFENSIVENESS
24

VALUES AND DEFENSIVENESS
25