Content uploaded by Dara Greenwood
Author content
All content in this area was uploaded by Dara Greenwood
Content may be subject to copyright.
DOI ./humor--Humor ; aop
Christopher R. Long and Dara N. Greenwood
Joking in the face of death: A terror
management approach to humor production
Abstract: Terror management theory has spawned a body of experimental re-
search documenting a multitude of defensive responses to mortality salience
manipulations (e.g., rigid adherence to dominant cultural values, self-esteem
bolstering). Another substantive body of work suggests that humor functions as a
natural and oen eective means of down-regulating stressful or traumatic expe-
riences. Integrating a terror management paradigm with a cartoon captioning
task, the present study nds that participants subliminally primed with death
wrote funnier captions than those primed with pain, as judged by outside raters.
Interestingly, a reverse pattern was obtained for participants’ own ratings of their
captions; explicitly death-primed participants rated themselves more successful
at generating humorous captions than their pain-primed counterparts, while no
signicant dierence emerged between the two subliminal priming conditions.
Findings contribute new insights to recent research suggesting that death re-
minders may sometimes facilitate creativity and open-mindedness.
Keywords: humor, terror management, mortality, creativity
Christopher R. Long: Ouachita Baptist University. E-mail: longc@obu.edu
Dara N. Greenwood: Vassar College.
Joking in the face of death: A terror
management approach to humor production
Terror Management Theory (TMT, e.g., Pyszczynski, Greenberg and Solomon
1997) posits that human awareness (whether conscious or unconscious) of the
inevitability of death can lead to potentially paralyzing anxiety. To manage or pre-
empt this anxiety, individuals may turn to cultural and psychological defenses
that ostensibly oer symbolic ways to transcend death. Indeed, decades of em-
pirical research have shown that priming individuals with their own mortality
leads to various negative outcomes that seem predicated on a narrowing of the
social psychological lens: increased adherence to nationalistic attitudes (e.g.,
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
Landeau et al. 2004), derogation of counter-attitudinal outgroup members (e.g.,
See and Petty 2006), and endorsement of materialistic goals (e.g., Kasser and
Sheldon 2000). Presumably, cultural status confers a sense of value in a rule-
bound and meaningful social universe, both of which provide relief from the
simultaneous certainty of death and the uncertainty of when, how, or why one’s
death will occur. Interestingly, although increased rigidity of world views seems
born of negative vs. positive emotions, a recent study found that mortality sa-
lience facilitated unconscious accessibility of and attention to positive emotional
content, which researchers argue signies activation of an automatic coping
strategy (De Wall and Baumeister 2007). A few additional studies have focused on
the conditions under which mortality salience may either lead to or interact with
creative thinking (Routledge and Arndt 2009; Routledge, Arndt, Vess and Shel-
don 2008). Thus, it is important to consider alternative responses to reminders of
death that may implicate a creative, positive emotional mindset. The present
study is the rst to investigate the impact of death reminders on the creative task
of humor production.
Humor production may be particularly relevant to staving o death anxiety,
not only because it typically is a culture-bound phenomenon, and hence useful
for rearming one’s place in society, but humor has also been identied as a
psychologically useful coping mechanism that enables individuals to remain re-
silient in the face of aversive life circumstances (Abel 2002; in the case of smiling/
laughing during bereavement; Keltner and Bonanno 1997; Martin et al. 1993;
Nezu, Nezu and Blissett, 1988). Considered a “high adaptive” defense mechanism
by the latest edition of the Diagnostic and Statistics Manual of Mental Disorders
(American Psychiatric Association, DSM-IV-TR 2000: 75), the capacity to utilize
humor in the face of negative experiences lends itself to both positive outcomes
and a classication of “trait resilience” by researchers in positive psychology
(e.g., Tugade and Fredrickson 2007: 319). Other scholars have underscored situa-
tional forces that pull for humor as a coping mechanism, citing the value and fre-
quency of humor production among individuals who are chronically confronted
with reminders of mortality, such as medical sta (van Wormer and Boes 2009) or
prisoners of war (Henman 2001). Still others have reected on the evolutionary
value of “laughing at death” in response to the “fundamental terrorizing incon-
gruity of mortality awareness in creatures who dream of immortality” (Elgee
2003, p. 479).
In the humor-as-coping literature, Victor Frankl’s (1988 [1946]) documenta-
tion of how he and others survived the chronic emotional and physical torment of
the Holocaust is oen referenced. In his preface to Frankl’s Man’s Search for
Meaning, Gordon Allport notes that the unimaginable horrors of life in a concen-
tration camp were made tolerable by, among other life-saving coping mechanisms,
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
“a grim sense of humor” (9), and Frankl himself alludes to humor as “another of
the soul’s weapons in the ght for self-preservation” (63). It is signicant to con-
ceptualize the role of humor within Frankl’s original contribution of logotherapy
– literally “meaning therapy” – a process by individuals can learn to infuse the
random chaos and suering of everyday life with signicance. From this per-
spective, humor may be conceived of, in part, as a route by which a sense of
meaning or control is achieved via adaptive distancing from an otherwise over-
whelming situation.
In one of only a handful of studies to experimentally examine the eect of
humor generation on stress and negative aect, Newman and Stone (1996) ran-
domly assigned participants to generate humorous vs. serious narrative while
viewing a stressful, silent lm segment (depicting an industrial accident). Results
showed that individuals in the humorous narrative condition reported signi-
cantly lower negative aect, showed lower physiological stress indicators, and
evidenced faster stress recovery just aer the task compared to those in the seri-
ous narrative condition. The authors highlight the potential health implications
of humor production as a stress-reduction technique. Similarly, Giulini, McRae
and Gross (2008) found that instructing individuals to reappraise the amusing
lm so as to make it more amusing lead to increased perceptions, behaviors and
physiological reactivity associated with amusement compared to simply watch-
ing the amusing lm without instruction. The authors note that such ndings
clarify the value of being able to consciously “up-regulate” positive emotional
experiences in the face of stress and provide empirical support for the role of
humor in physical and emotional health (718).
To the extent that down-regulating death anxiety is one of the more primitive
tasks facing human beings, and to the extent that humor oen appears to miti-
gate stress, it seems plausible that humor generation may be (adaptively) facili-
tated, rather than inhibited, by mortality salience primes. Although it may seem
counterintuitive in light of the amassed research on terror management theory
suggesting the perspective-narrowing impact of death primes, there are a few
recent studies that provide converging evidence for our assertion. Specically,
Routledge and Arndt (2009) found that instructing individuals to engage in a cre-
ative task (designing a t-shirt) following an explicit mortality salience manipu-
lation led to increased exploration intentions as well as increased interest in
“worldview-challenging lms” (e.g., a lm challenging American ethnocen-
trism). Their third study found that simply priming individuals with the idea that
creativity is culturally valued and associated with “helping people nd success in
America” increased their interest in lms that challenged dominant cultural
views on religion. The authors conclude that their ndings “extend previous ter-
ror management reactions beyond the dogmatic armation of existing beliefs
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
and advance the possibility that people can maintain psychological security by
exploring other cultural worldviews” (502).
A related study looks at creativity as an outcome variable in itself (Routledge,
Arndt, Vess and Sheldon 2008). Explicitly death-primed individuals who were
given a task to design a rock concert to make money for themselves did so less
creatively (as judged by outside raters) than did their pain-primed counterparts
when given similar instructions. However, when told the goal was to make money
for a charitable organization, death-primed participants did not dier from pain-
primed participants in their creativity. The authors note that although the trend
was non-signicant, the means suggested that “a connectedness-oriented activity
may even promote creativity in response to existential threat” (335).
Both of the above-described studies suggest that individuals may need to be
instructed to engage in either a creative or communal task in order for death re-
minders to elicit or at the least not diminish cognitively exibility. However, a
more basic test of the emotional mechanism triggered by death reminders sug-
gests that positive emotional responses may be activated automatically as a cop-
ing mechanism against existential terror. De Wall and Baumeister (2007) reason
that: “Clutching happy thoughts may serve the function (central to terror man-
agement theory) of preventing the conscious mind from being paralyzed by the
terror of death” (984). In support of this prediction, a series of studies found that
explicitly death-primed individuals showed enhanced accessibility of positive af-
fective information. Individuals primed with thoughts of their own death were
more likely to make positive word stem completions and to make more positive
emotional associations than semantic associations to a presented target word
(e.g., associating “mouth” with “smile” instead of “cheek”) compared to their
pain-primed counterparts.
In sum, there is some evidence that individuals may marshal positive emo-
tional and cognitive resources in response to death anxiety, either spontaneously
or under certain instructional conditions. Our study builds on this research by
integrating literature documenting the adaptive benets of humor with literature
suggesting that creativity and positive thinking may serve as a buer against
death anxiety. Specically, we explore whether the ability to generate humor is
facilitated under mortality salience. We further ask whether the modality of prime
is a factor.
.The present study
The current study was undertaken to determine whether individuals are more
adept at devising humorous material when primed with thoughts of death com-
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
pared to thoughts of pain. Pain is a typical terror management control condition
because it is aversive yet qualitatively dierent from death. A second purpose of
the present study was to determine whether prime modality, in this case, primes
presented subliminally (via computer) or explicitly (via written exercise) would
impact humor generation. We assessed humor generation by asking participants
generate captions for a series of caption-less cartoons following a ller task. Out-
side raters, blind to study condition and hypotheses, subsequently rated the cap-
tions for perceived humorousness. In addition, we asked the participants them-
selves to report on their perceived success and subjective ease with which they
generated the captions.
Based on the research outlined above, we predicted that individuals primed
with death would generate more humorous cartoons (as judged by outside raters)
than individuals primed with pain. However, we le open the question of whether
subliminal or explicit death reminders would be more likely to facilitate humor
generation. In addition, we were interested in determining whether prime (death
vs. pain) and/or modality (subliminal vs. explicit) eects would emerge with re-
spect to individuals’ perceptions of their own captions. That is, we wanted to
clarify whether individuals would perceive their own captions as more successful
and/or easy to write as a function of how they were primed.
This study oers the rst empirical test of whether death reminders result in
more successful humor generation, which would contribute to the small but com-
pelling body of work that suggests that mortality salience spontaneously acti-
vates adaptive, creative thinking. This study also oers the rst test of whether
quality of humor generation would vary as a function of prime modality (e.g.,
subliminal vs. explicit). Arndt, Greenberg, Pyszczynski and Solomon (1997) found
evidence that subliminal reminders of death aroused stronger defenses than
those that were activated explicitly (e.g., greater endorsement of a pro-US essay).
The authors suggest that when it comes to death anxiety, it is “precisely when
people are unaware of this fear that it has the strongest impact on their behavior”
(384). Of note, a recent meta-analysis of research on terror management theory
found no dierences in prime modality on numerous outcome variables; how-
ever, only 4% of the studies sampled (277 experiments) utilized the subliminal
priming technique (Burke, Martens and Faucher 2010). Also of note, we utilized a
delay (e.g., ller task) in both versions of the priming manipulation to preserve
continuity across conditions. Although the TMT literature suggests that a delay is
only needed aer an explicit reminder of death to facilitate non-conscious defen-
sive processes, we reasoned that the only concern with respect to the subliminal
priming condition would be a potential fade out eect. We kept this possibility in
mind when performing analyses, anticipating that we might focus in on the rst
caption attempt relative to all others.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
Method
.Participants
One-hundred twenty-three college students (79% female) at a small private uni-
versity were recruited to participate in a study examining “Cognition and Humor”
in return for course credit. When asked during debrieng, two participants indi-
cated that they may have seen additional words onscreen during the subliminal
priming task (without memory for the content of these words), and four partici-
pants indicated that they had eye or reading diculties that caused them prob-
lems during the subliminal priming task; these six participants were excluded
from the analyses reported below. Thus, 28 participants completed the sub liminal
death priming task, 26 completed the subliminal pain priming task, 32 completed
the written death priming task, and 31 completed the written pain priming task.
.Materials
There were four versions of the priming tasks, which varied according to modality
and content. With respect to modality, two of them were subliminal priming
tasks, and two were written priming tasks. For each modality, one task primed
death whereas the other primed pain. As described below, each participant com-
pleted one of the priming tasks followed by the ller task and then the caption
generation task. More detail is provided below.
..Subliminal priming tasks
These tasks generally followed the procedures outlined by Arndt et al. (1997). Par-
ticipants assigned to either of the subliminal priming tasks were seated in front of
a computer and told that their rst task would be a “word-relation” task. Aer the
experimenter le the room, onscreen instructions indicated that for each trial of
the word-relation task, “two words will be ashed in sequence, and you must
decide whether or not you feel like the words are related to each other. We are
interested in your ‘gut-level responses’ so please respond rapidly in categorizing
each pair, but don’t respond so fast that you make many errors.” Participants
were informed that they were to indicate whether or not they thought the words
seemed to be related by pressing the ‘A’ key if the words seemed unrelated or the
‘L’ key if the words seemed related. Participants were then shown an example of
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
possible responses: “What if the word ‘FAJITA’ is ashed rst and is followed by
‘SNEAKER’? In this case, you would press the ‘A’ key if these two words seem to
you to be unrelated. However, you would press the ‘L’ key if these words seem to
you to be related.”
Participants were then asked to complete ve practice trials followed by a
short break and then een experimental trials. For each of the practice and ex-
perimental trials, two dierent words selected randomly from a list of stimuli
were displayed one aer the other in the center of the screen. Using Inquisit
(2006) stimulus presentation soware, the rst word of each pair was displayed
for approximately 430 ms, and the second word of each pair was displayed until
the participant responded by pressing either the ‘A’ or ‘L’ key on the computer
keyboard. The stimulus words were chosen from a list of words selected because
they were unrelated to death or pain and were relatively neutral in aective tone.
For example, the list of stimuli included CALCULATOR, ACCOUNTANT, LETTUCE,
LIGHTSWITCH, and SHOELACE.
For each of the practice and experimental trials, subliminal primes were also
displayed. Specically, depending on whether a participant had been assigned
the death or the pain subliminal priming task, either the word DEAD or PAIN was
displayed onscreen for 33 ms immediately aer the display of rst stimulus word
of each pair and was then replaced by the second stimulus word, which served as
a post-stimulus mask.
Aer the nal trial, onscreen instructions directed participants to open an
envelope that contained the pencil-and-paper questionnaire that contained the
ller task, the cartoon caption generation task, and a debrieng questionnaire
and to begin working on these tasks.
..Written priming tasks
Participants assigned to either of the written priming tasks were seated at a table
and told that their rst task would be to complete an “Attitudes toward Life
Events” questionnaire, which assessed “people’s beliefs about what certain ex-
periences feel like.” Depending on whether a participant had been assigned the
death or the pain written priming task, each participant then responded to either
of two pairs of writing prompts. These open-ended writing prompts used explicit
priming to activate thoughts of death or dental pain and have been widely used in
mortality salience research. Specically, the death primes prompted participants
to “Please briey describe the thoughts and emotions that the thought of your
own death arouses in you” and to “Jot down, as specically as you can, what you
think will happen to you as you physically die and once you are physically dead.”
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
The dental pain primes prompted participants to “Please briey describe the
emotions that the thought of dental pain arouses in you” and to “Jot down, as
specically as you can, what you think will happen to you as you physically expe-
rience dental pain.” Space was provided below each prompt for participants to
indicate their responses.
When seating participants for the written priming task, the experimenter in-
structed them to continue to the next task immediately upon completing the sec-
ond writing prompt. To ensure that participants in all four conditions completed
the tasks in the same order, all participants received oral and written instructions
to be sure to complete each item on a page before proceeding to the next page
and, once they had proceeded to a subsequent page, not to turn back to a previ-
ous page.
..Filler task
For all participants, the priming task was followed by a word search task labeled
as a “Search Organization Task” which required participants to locate 14 “num-
ber” words (e.g., “eleven” or “twenty”) in a matrix of letters. This task required
approximately ve minutes to complete. Mortality salience researchers typically
nd that a brief delay between the priming task and completion of a study’s de-
pendent measures accentuates the eects of the death prime (cf., Burke et al.
2010).
..Caption generation task
The nal task for all participants was described as a “Humor Generation Task.”
This task presented each participant with four caption-less cartoon drawings
drawn from a sample of The New Yorker magazine’s weekly Cartoon Caption Con-
test (devised and administered by The New Yorker cartoonist and cartoon editor
Robert Manko). The four selected cartoons were taken from a sample of cartoons
prescreened to eliminate overt images or themes of death, dentistry, and violence.
In addition, the four selected cartoons were chosen because, as compared to the
other prescreened cartoons, they were rated by a panel of four undergraduates as
neutral in mood, moderately funny, and moderately dicult to caption. The four
selected cartoons depicted, respectively, (a) a woman and man in a hotel room
looking at an open suitcase full of water and swimming sh (Vey 2005); (b) a
man, a woman, and a dog sitting in a living room and each appearing to be read-
ing a book (Shanahan 2005); (c) a young girl playing violin for an older man in a
disheveled room featuring scattered papers and a broken lamp (Diee 2006); and
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
(d) a business executive and the contents of his oce oating in the air, seem-
ingly liberated from the constraints of gravity (Ziegler 2005). The instructions for
the caption generation task directed participants to “write a humorous caption”
and that their “goal should be to write down the funniest caption that comes to
your mind” and that they should not turn the page until they have written a cap-
tion. Each participant completed the captions in one of four randomly assigned
orders, with each cartoon appearing in a dierent ordinal position in each of the
four orders. Cartoons were presented one per page, and below each cartoon was
a blank in which to write a caption.
Aer generating the captions, participants were asked, separately for each
cartoon, how easy it was for them to generate a caption for the cartoon and how
successful they felt they were at generating a humorous caption for the cartoon.
These self-ratings were made on nine-point scales (where, e.g., 1=extremely
dicult and 9=extremely easy). The items corresponding to each cartoon were
presented on a separate page, with the pages ordered to mirror the order in which
the captions were generated and a copy of the relevant cartoon reprinted at the
top of each page.
Although we allowed participants four attempts to generate captions, we also
were mindful that the priming eects, if any, would most likely manifest in the
rst attempt. We were also mindful that any subsequent attempts at humor gen-
eration would necessarily be confounded with the rst attempt (particularly if
humor is acting as a defense against death reminders) and would thus render
interpretation of the second through fourth captions dicult. That is, if the rst
attempt were most sensitive to the manipulation and hence also most relevant
and successful in its defensive function, subsequent attempts would necessarily
be diluted with respect to both prime impact and defensive functioning. Results
reported below are therefore based only on the rst caption attempt.
.Procedure
Each participant completed the study tasks while alone in a small, sound-proof
room. First, each participant completed one of four randomly-assigned versions
of the priming task (subliminal death, subliminal pain, written death, or written
pain). Following the priming task and the ller task, each participant completed
the caption generation task. Most participants completed their assigned tasks in
fewer than 15 minutes. Aer completing the caption generation task, participants
were thanked and debriefed.
To generate ratings of the humorousness of the generated captions, six
undergraduate psychology students were recruited to serve as raters. Adapting
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
Amabile’s (1983) consensual assessment technique, each rater rated each par-
ticipant’s captions on a seven-point scale, where 1=extremely unfunny and
7=extremely funny. Following a brief training session in which each rater was
instructed to rate the (randomized) captions according to his or her own sense of
humor, each rater worked alone on rating all captions.
Results
.Caption ratings
Interrater agreement was calculated for the six raters’ ratings of the captions’
humorousness and was acceptable, intraclass correlation (2, 6)=.65. For each
caption, a mean humorousness rating was calculated by computing the mean of
all six raters’ ratings. The mean of the mean humorousness ratings was 3.13
(SD=.82), slightly below the midpoint of the seven-point rating scale. The mean
ratings ranged from 1.50, for the caption with the lowest average humor rating, to
5.17, for the caption with the highest average humor rating. By way of illustration,
an example of a caption rated as especially humorous for the cartoon of the sh
in the suitcase was “When I said be sure not to forget the sh, I meant for you to
feed them” (M=4.33). For the same cartoon, a caption rated as relatively unfunny
was “I guess my suitcase went for a swim” (M=2.50).
To ensure that the selected cartoons did not dier in the extent to which they
elicited humorous captions, we performed an analysis of variance (ANOVA) ana-
lyzing humor ratings as a function of which of the four cartoons were randomly
captioned rst by participants. No dierences in funniness ratings emerged as a
function of which of the four cartoons participants were captioning, F(3, 112)=1.67,
p=.18.
.Priming eects on perceived humorousness
A 2×2 ANOVA examining the eects of prime content (death vs. pain) and prime
modality (subliminal vs. written) on the captions’ humorousness yielded no main
eects for prime content or prime modality, F(1, 112)=.02, p=.89, and F(1, 112)<
.01, p>.99, respectively. However, there was a signicant Content×Modality
1 As noted above, all reported analyses were only based on the rst caption attempt. However,
for all analyses implicating prime content or prime modality, when we examined each sub-
sequent caption attempt separately (or all four in aggregate), no priming eects emerged.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
interaction for caption humorousness, F(1, 112)=8.90, p<.01. As depicted in Fig-
ure 1, analysis of simple main eects claried the nature of this interaction.
Among participants who completed the subliminal priming tasks, the death
prime (M=3.34) yielded more humorous caption ratings than did the pain prime
(M=2.92), F(1, 112)=3.77, p=.05. On the other hand, among participants who
completed the written priming tasks, the pain prime (M=3.36) yielded more
humorous caption ratings than did the death prime (M=2.90), F(1, 112)=5.25,
p=.02.
.Priming eects on self-rated task performance
A 2×2 ANOVA examining the eects of prime content (death vs. pain) and prime
modality (subliminal vs. written) on participants’ ratings of how easy it was for
them to generate a caption yielded no main eects for prime content or prime
modality and no interaction, F(1, 113)=.19, p=.66; F(1, 113)=.11, p=.74; and
F(1, 113)=2.65, p=.11, respectively.
A 2×2 ANOVA on participants’ ratings of how successful they were at gener-
ating funny captions yielded no main eect for prime content, F(1, 113)=.02,
Fig. 1: Mean humorousness ratings as a function of prime content and prime modality. Standard
errors are represented in the gure by the error bars attached to each column.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
p=.89. However, there was a main eect for prime modality, with participants
who completed the subliminal priming tasks rating themselves as more success-
ful (M=4.36) than did participants who completed the writing priming tasks
(M=3.36), F(1, 113)=6.91, p=.01. There was also a signicant Content×Modality
interaction for self-rated success, F(1, 113)=5.65, p=.02. Analysis of simple main
eects claried the nature of this interaction. For the subliminal priming tasks,
those in the death prime condition (M=4.14) and the pain prime condition
(M=4.58) did not dier in self-rated success, F(1, 113)=.03, p=.86. However, for
the written priming tasks, those who were in the death prime condition (M=4.05)
rated themselves as more successful at generating a funny caption than did those
in the pain prime condition (M=2.68), F(1, 113)=5.25, p<.01.
Discussion
The present study is the rst to investigate the role that humor generation plays in
the context of terror management theory. Moreover, it is the rst to test how mo-
dality of prime (subliminal vs. explicit) interacts with a creative task. Results
showed no main eects for prime or modality, but did show a signicant inter-
action between the two. Specically, those primed with subliminal reminders of
death generated cartoon captions judged to be more humorous than those primed
with subliminal reminders of pain. However, the reverse pattern was obtained for
those primed with an explicit writing task; in this case, those who wrote about
dental pain generated more humorous captions than those who wrote about their
own death.
These ndings suggest that creative humor production is inhibited by ex plicit
mortality primes but facilitated by subliminal mortality primes. The latter nd-
ings are in step with research which suggests that subliminal or subtle reminders
of death are most likely to increase death-accessibility and, accordingly, increase
activation of symbolic, experiential terror management strategies (Arndt et al.
1997; Greenberg et al. 1994; Pyszczynski, Greenberg and Solomon 1999). Our
study nds evidence of a type of “broaden-and-build” cognitively exible re-
sponse to subliminal death reminders (Fredrickson 1998). As noted earlier,
however, only a small minority of terror management experiments make use of
subliminal death reminders. More work is needed to clarify the conditions under
which death reminders do in fact activate creativity vs. rigidity of mind and the
associated positive outcomes (e.g., more open-mindedness for dierent political
ideologies) that some work has already found occurs in response to creative exer-
cises (Routledge and Arndt 2009).
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
It is somewhat surprising that explicit reminders of death inhibited humor
generation, given research that nds such cues trigger automatic attunement to
positive emotional information (De Wall and Baumeister 2007), or, under certain
conditions (prosocial cues), creative problem solving (Routledge et al. 2008). This
inhibition eect is also at odds with the literature documenting spontaneous use
of humor in response to chronic and explicit death reminders (e.g., in times of
war, Henman 2001; in the hospital emergency room, van Wormer and Boes 2009).
It is possible that the explicit reminders of death activated a less positive initial
response to the cartoon images (all of which were fairly surreal), which may have
then inhibited creative humor production; this would t with research showing
that explicit death primes increase negative responses to modern art, presumably
because such art is viewed as threateningly meaningless (Landeau et al. 2006).
Interestingly, whereas individuals who were subliminally primed with death
or pain judged their captions to be more successful than those in the explicit
prime conditions, individuals explicitly primed with death judged their cartoons
to be signicantly more successful than those explicitly primed with pain. There
was thus a disconnect between subjective and objective ratings of success at the
cartoon task; individuals in the explicit death prime condition were found to have
generated signicantly less humorous captions than those primed with pain.
Findings might be partly explained by explicit death reminders activating a self-
esteem bolstering defense relative to explicit pain reminders (e.g., Harmon-Jones
et al. 1997). It is not clear why no signicant dierence between pain and death
primes emerged in the subliminal conditions, however. More work is needed to
clarify these distinctions.
It bears mentioning that while our divergent ndings with respect to prime
modality are somewhat anomalous within the limited studies of terror manage-
ment in which subliminal processes are examined, divergence between implicit
vs. explicit primes and/or assessments are the norm in other areas of research
such as prejudice, in which dual process theories abound (e.g., research utilizing
the implicit association test). To the extent that humor production is activated in
response to unconscious defensive strategies, and to the extent that subliminally
primed death anxiety may activate more authentically unconscious defense strat-
egies (vs. those activated via mere distraction from an explicit death prime), we
may be observing a particular synchrony between prime modality and defensive
strategy. This nding speaks to the possibility that successful humor generation
(vs. the perception of success) may be particularly sensitive to unconscious inu-
ences. More work is needed to continue probing these provocative distinctions.
Our study is not without limitations. While our single-item measures (e.g.,
self and cartoon ratings) provided the most parsimonious tests of our hypo-
thesesand research questions, future research should continue to explore other
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
operational denitions of successful humor generation. Similarly, utilizing The
New Yorker caption contest provided a convenient and ecologically valid outcome
variable but it represented only one particular test of humor production. Future
work should replicate across a number of dierent humor exercises. It is also not
entirely clear from our study whether generating a caption implicitly primed the
idea of amusing others, in addition to or instead of the self. Given that prosocial
motivations for humor production increase creativity when explicitly primed in a
lab setting (Routledge et al. 2008), future work should attempt to tease apart in-
dividual vs. collective goals.
Future research should also explore the distinction between humor genera-
tion and humor appreciation as they relate to TMT. Humor generation may be
more specic to creative thinking than humor appreciation; the latter may actu-
ally be more linked to bolstering rigid cultural worldviews as a coping mecha-
nism. For example, it is possible that mortality salience increases liking for humor
that disparages outgroups. Interestingly, recent empirical evidence suggests that
(explicitly) death-primed individuals appreciate death-themed humor more than
punishment-themed humor and marginally more than control primed individu-
als (Hackney, 2011). However, the punishment-themed comic may not generalize
to meaningful social outgroups, as it focused on the humiliation associated with
being a telemarketer. This research on humor appreciation does raise the possi-
bility that “grim reaper” style cartoons, for example, may serve a particularly
death-defying function. More work is needed to determine whether enjoyment of
death-relevant humor stimuli actually reduces death anxiety or merely reects a
tendency to appreciate prime congruent material.
Finally, although other research has found positive outcomes associated with
creative thinking as well as the stress-mitigating eects of humor and amuse-
ment, our work was only designed to test the link between death reminders and
successful humor generation. It will be important to clarify the ecacy of humor
generation as a coping technique in the context of mortality salience experiences,
as well as the conditions under which individuals’ capacity for humor generation
under adverse circumstances (e.g., death reminders) may confer emotional, cog-
nitive, and/or social benets.
Acknowledgements
Many thanks are due to Robert Manko for his interest in and enthusiasm for this
project and for his creatively challenging Cartoon Caption Contest. We are also
grateful to Jamie Barrett, Stephanie Ellison, Courtney Gibson, Jessica Hensarling,
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
Lauren Landes, Michael Marsden, Aimee Mathis, Kim Parker, Gabri Sanders,
Catherine Senko, and Sara Terlecki for rating cartoon captions or assisting with
data collection. Both authors made equal and complementary contributions to
this manuscript.
References
Abel, Millicent H. 2002. Humor, stress, and coping strategies. Humor: International Journal of
Humor Research 15(4). 365–381.
Arndt, Jamie, Je Greenberg, Tom Pyszczynski and Sheldon Solomon. 1997. Subliminal
exposure to death-related stimuli increases defense of the cultural worldview.
Psychological Science 8(5). 379–385.
American Psychiatric Association. 2000.Diagnostic and statistical manual of mental
disorders(Revised 4th edn.). Washington, DC: Author.
Burke, Brian L., Andy Martens and Erik H. Faucher. 2010. Two decades of terror management
theory: A meta-analysis of mortality salience research. Personality and Social Psychology
Review 14(2). 155–195.
De Wall, C. Nathan and Roy F. Baumeister. 2007. From terror to joy: Automatic tuning to positive
aective information following mortality salience. Psychological Science 18(11). 984–990.
Diee, Matthew. 2006, May 1. Untitled cartoon. New Yorker 82(11). 98.
Elgee, Neil J. 2003. Laughing at Death. Psychoanalytic Review 90(4). 475–497.
Frankl, Viktor E. 1988 [1946]. Man’s search for meaning: An introduction to logotherapy.
Washington Square Press: New York.
Fredrickson, Barbara L. 1998. What good are positive emotions? Review of General Psychology
2(3). 300–319.
Giuliani, Nicole R, Kateri McRae and James J. Gross. 2008. The up- and down-regulation of
amusement: Experiential, behavioral, and automatic consequences. Emotion 8(5).
714–719.
Greenberg, Je, Tom Pyszczynski, Sheldon Solomon, Linda Simon and Michael Breus. 1994.
Role of consciousness and accessibility of death-related thoughts in mortality salience
eects. Journal of Personality and Social Psychology 67(4). 627–637.
Hackney, C. H. (2011). The Eect of mortality salience on the evaluation of humorous material.
Journal of Social Psychology, 151, 51–62.
Harmon-Jones, E., Simon, L., Greenberg, J., Pyszczynski, T., Solomon, S., & McGregor, H. (1997).
Terror Management Theory and self-esteem: Evidence that increased self-esteem reduces
mortality salience eects. Journal of Personality and Social Psychology, 72, 24–36.
Henman, Linda D. 2001. Humor as a coping mechanism: Lessons from POWs. Humor:
International Journal of Humor Research 14(1). 83–94.
Kasser, Tim and Kennon M. Sheldon. 2000. Of wealth and death: Materialism, mortality
salience, and consumption behavior. Psychological Science 11(4). 348–351.
Inquisit 2.0.60616 [Computer soware]. 2006. Seattle, WA: Millisecond Soware.
Keltner, Dacher and George A. Bonanno. 1997. A study of laughter and dissociation: Distinct
correlates of laughter and smiling during bereavement. Journal of Personality and Social
Psychology 73(4). 687–702.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Christopher R. Long and Dara N. Greenwood
Landeau, Mark J., Sheldon Solomon, Je Greenberg, Florette Cohen, Tom Pyszczynski, Jamie
Arndt, Claude H. Miller, Daniel M. Ogilvie and Alison Cook. 2004. Deliver us from evil: The
eects of mortality salience and reminders of 9/11 on support for President George W.
Bush. Personality and Social Psychology Bulletin 30(9). 1136–1150.
Landau, Mark J., Je Greenberg, Sheldon Solomon, Tom Pyszczynski and Andy Martens, A.
2006. Windows into nothingness: Terror management, meaninglessness, and negative
reactions to modern art. Journal of Personality and Social Psychology 90(6). 879–892.
Lykins, Emily L. B., Suzanne C. Segerstrom, Alyssa J. Averill, Daniel R. Evans and Margaret E.
Kemeny. 2007. Goal shis following reminders of mortality: Reconciling posttraumatic
growth and terror management theory. Personality and Social Psychology Bulletin 33(8).
1088–1099.
Martin, Rod A., Nicholas A. Kuiper, L. Joan Olinger and Kathryn A. Dance. 1993. Humor, coping
with stress, self-concept, and psychological well-being. Humor: International Journal of
Humor Research 6(1). 89–104.
Nezu, Arthur M., Christine M. Nezu and Sonia E. Blissett. 1988. Sense of humor as a moderator
of the relation between stressful events and psychological distress: A Prospective
analysis. Journal of Personality and Social Psychology 54(3). 520–525.
Newman, Michelle Gayle and Arthur A. Stone. 1996. Does humor moderate the eects of
experimentally-induced stress. Annals of Behavioral Medicine 18(2). 101–109.
Pyszczynski, Tom, Je Greenberg and Sheldon Solomon. 1997. Why do we need what we need?
A terror management perspective on the roots of human social motivation. Psychological
Inquiry 8(1). 1–20.
Pyszczynski, Tom, Je Greenberg and Sheldon Solomon. 1999. A dual-process model of defense
against conscious and unconscious death-related thoughts: An extension of terror
management theory. Psychological Review 106(4). 835–845.
Routledge, Clay D. and Jamie Arndt. 2009. Creative terror management: Creativity as a
facilitator of cultural exploration aer mortality salience. Personality and Social
Psychology Bulletin 35(4). 493–505.
Routledge, Clay, Jamie Arndt, Matthew Vess and Kennon M. Sheldon. 2008. The life and death
of creativity: The eects of mortality salience on self versus social-directed creative
expression. Motivation and Emotion 32(4). 331–338.
See, Ya Hui Michelle and Richard E. Petty. 2006. Eects of mortality salience on evaluation of
ingroup and outgroup sources: The impact of pro-versus counterattitudinal positions.
Personality and Social Psychology Bulletin 32(3). 405–416.
Shanahan, Danny. 2005, September 19. Untitled cartoon. New Yorker 81(28). 100.
Tugade, Michele M. and Barbara L. Fredrickson. 2007. Regulation of positive emotions: Emotion
regulation strategies that promote resilience. Journal of Happiness Studies 8(3). 311–333.
van Wormer, Katherine and Mary Boes. (2009). Humor in the emergency room: A Social work
perspective. Health and Social Work 22(2). 87–92.
Vey, P. C. 2005, December 26. Untitled cartoon. New Yorker 81(42). 152.
Ziegler, Jack. 2005. September 12. Untitled cartoon. New Yorker 81(27). 104.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Terror management and humor production
Bionote
Christopher R. Long is an Associate Professor of Psychology at Ouachita Baptist
University.
Dara Greenwood is an Assistant Professor of Psychology at Vassar College.
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
Brought to you by | Universidad de Magallanes
Authenticated | 143.229.168.197
Download Date | 4/29/13 6:16 PM
