ArticlePDF Available

When Self-Destructive Thoughts Flash Through the Mind: Failure to Meet Standards Affects the Accessibility of Suicide-Related Thoughts

Authors:

Abstract

When individuals realize that they fail to attain important standards or expectations, they may be motivated to escape the self, which could lead thoughts of suicide to become more accessible. Six studies examined this hypothesis, mainly derived from escape theory (Baumeister, 1990). The results indicated that whenever individuals realize that they fail to attain an important standard, they experience increased accessibility of suicide-related thoughts (Studies 1-6). In line with the idea that such effects reflect motivations to escape from negative self-awareness, they were especially pronounced when associated with high levels of self-consciousness and escapist motivations (Study 1) and with a large discrepancy between self and standards (Studies 2-4). Moreover, failure to attain standards increased suicide-thought accessibility along with the desire for an altered state of consciousness (Study 5). Finally, increases in suicide-thought accessibility after failure were associated with simultaneous increases in accessibility of general concepts related to escape (Study 6). Implications of these findings for escape and terror management theories are discussed.
When Self-Destructive Thoughts Flash Through the Mind:
Failure to Meet Standards Affects the Accessibility of Suicide-Related Thoughts
Armand Chatard
University of Geneva
Leila Selimbegovic´
University of Poitiers
When individuals realize that they fail to attain important standards or expectations, they may be
motivated to escape the self, which could lead thoughts of suicide to become more accessible. Six studies
examined this hypothesis, mainly derived from escape theory (Baumeister, 1990). The results indicated
that whenever individuals realize that they fail to attain an important standard, they experience increased
accessibility of suicide-related thoughts (Studies 1– 6). In line with the idea that such effects reflect
motivations to escape from negative self-awareness, they were especially pronounced when associated
with high levels of self-consciousness and escapist motivations (Study 1) and with a large discrepancy
between self and standards (Studies 2– 4). Moreover, failure to attain standards increased suicide-thought
accessibility along with the desire for an altered state of consciousness (Study 5). Finally, increases in
suicide-thought accessibility after failure were associated with simultaneous increases in accessibility of
general concepts related to escape (Study 6). Implications of these findings for escape and terror
management theories are discussed.
Keywords: escape theory, self-awareness, standards, failure, accessibility of suicide-related thoughts
My paper has been rejected. I’m going to throw myself into the lake.
—Anonymous student
Like the student quoted here, most of us have experienced major
failures or setbacks. Fortunately, we are somewhat resilient to
failure and do not attempt to end our lives each time we fail to
meet important standards or expectations. However, this anecdote
suggests that there may be a direct causal link between failure and
self-destructive thoughts. That is, when individuals realize that
they fall short of standards, they could experience increased ac-
cessibility of suicide-related thoughts. The present research exam-
ines this intriguing possibility. As we show, even if this hypothesis
has never been proposed before, a review of the literature suggests
that it is plausible. In particular, it is consistent with an analysis of
suicide in terms of motivation to escape the self (Baumeister,
1990, 1991a). The main hypothesis examined here is important, we
believe, because it can shed light on some relatively simple but
essential issues: When do normal, mentally healthy individuals
come to think of suicide? And what kind of situations can trigger
self-destructive thoughts?
The Problem of Suicide in Psychology
According to the World Health Organization (2009), almost one
million people die by suicide every year. Suicide accounts for
more than all deaths from wars and homicides combined. In the
past 45 years, suicide rates have increased by 60% worldwide. The
prescription of new medications, such as antidepressants, has
failed to reduce suicide rates (van Praag, 2002). Furthermore, and
rather alarmingly, suicide rates tend to increase in times of eco-
nomic crisis (Baudelot & Establet, 2006). The psychological
causes and antecedents of suicidal behavior thus deserve immedi-
ate attention.
Important scientific advances in understanding suicidal behavior
have been made. For example, many researchers now consider that
suicidal thoughts (i.e., cognitions) are the primary antecedents of
suicidal behaviors (e.g., Nock et al., 2008). Many researchers thus
believe that by understanding the origin of suicidal thoughts, we
can better understand, predict, and treat suicidal behavior (e.g.,
Wenzel, Brown, & Beck, 2009). In spite of this consensus, how-
ever, researchers have not been particularly successful in predict-
ing and preventing suicide (O’Connor, 2003; O’Connor & Sheehy,
2000; Wenzel & Beck, 2008). There are many reasons for this, but
part of the problem is the lack of comprehensive models and
theoretical frameworks within which specific hypotheses might be
developed and tested (Cornette, Abramson, & Bardone, 2000;
O’Connor, 2003; Wenzel & Beck, 2008). A related issue is that
most research conducted so far has relied on correlational designs
(O’Connor, 2003; Wenzel & Beck, 2008), which are limited in
their ability to reveal causal relations. Finally, as Nock and Banaji
(2007) recently noticed, previous research has relied almost ex-
clusively on individuals’ self-reports to assess suicidal thoughts.
This article was published Online First February 7, 2011.
Armand Chatard, Department of Psychology, University of Geneva,
Switzerland; Leila Selimbegovic´, Center for Research on Cognition and
Learning, University of Poitiers, France.
This research was supported by a grant from the Swiss National Foun-
dation for Scientific Research (FNRS Grant 100014-120626/1) awarded to
Armand Chatard. We express our gratitude to Paul N’Dri Konan, Mouna
Bakouri, Tatjana Safarikova, Pascal Maring, and Laurence Schaffhauser
for their help in data collection, as well as to Jamie Arndt, Lotte Van
Dillen, Pascal Huguet, Martial Van Der Linden, and Ann Beaton for having
read and commented on previous versions of this article.
Correspondence concerning this article should be addressed to Armand
Chatard, Faculte´ de Psychologie et des Sciences de l’E
´ducation, Uni Mail,
40 Bd du Pont d’Arve, CH-1205 Geneva, Switzerland. E-mail:
Armand.Chatard@unige.ch
Journal of Personality and Social Psychology © 2011 American Psychological Association
2011, Vol. 100, No. 4, 587–605 0022-3514/11/$12.00 DOI: 10.1037/a0022461
587
This can be problematic, they argued, because individuals may be
unwilling or unable to accurately report their own thoughts. A
number of researchers have hence suggested that progress may
now come from efforts to explore suicidal thoughts through im-
plicit measures (for a distinction between implicit and explicit
measures, see De Houwer, Teige-Mocigemba, Spruyt, & Moors,
2009). Nock and Banaji (2007) suggested that implicit measures
could provide important information on the associations between
the self and concepts related to suicide as well as on the factors that
affect the cognitive accessibility of suicide-related thoughts (see
also Cornette et al., 2000).
In summary, research should be more integrated into compre-
hensive theories, should be less correlational in its methodology,
and should rely more on implicit measures of suicidal thoughts.
These three goals served as guidelines for the present research. We
were mainly interested here in the cognitive accessibility of
suicide-related thoughts.
The Role of Suicide-Thought Accessibility
Social cognition researchers have long argued that construct
accessibility contributes to orienting, shaping, and directing human
behaviors (Bargh & Pratto, 1986; Dodgson & Wood, 1998; Green-
wald & Banaji, 1995; Higgins, Bond, Klein, & Strauman, 1986;
Higgins, King, & Mavin, 1982; Strauman & Higgins, 1987). In the
same way, most theories of suicide assume that repeated increases
in suicide-thought accessibility play a role in the suicidal process.
Consider, for example, the characterization of the processes that
precede suicide proposed by Shneidman (1996), a widely acknowl-
edged expert in this domain:
Suicide is the result of an interior dialogue. The mind scans its options;
the topic of suicide comes up, the mind rejects it, scans again; there is
suicide, it is rejected again, and then finally the mind accepts suicide as
a solution, then plans it, and fixes it as the only answer. (p. 15)
From this viewpoint, suicide-related thoughts can be and are
repeatedly rejected when they first come to mind, and a suicide
attempt occurs only when this is no longer the case. In a similar
vein, several authors have argued that automatic activation of
suicide-related thoughts is a phenomenon that needs to be consid-
ered in research on suicide (see, e.g., Abramson, Metalsky, &
Alloy, 1989; Cornette et al., 2000; Wenzel et al., 2009; Williams
& Pollock, 2001). However, to our knowledge, only two groups of
researchers have examined cognitive accessibility of concepts re-
lated to suicide.
Becker, Strohbach, and Rinck (1999) used a modified Stroop
task. Participants were presented with single words in various ink
colors and instructed to name the ink color as quickly as possible,
regardless of the word’s meaning (for a similar procedure, see
Williams & Broadbent, 1986). Some words were suicide-related
(e.g., hang), some suicide-unrelated but negative (e.g., envy), some
positive (e.g., talent), and some neutral (e.g., square). If suicide-
related concepts are particularly accessible, then recognition of
words’ meanings should be likely to interfere with recognition of
words’ colors. In this case, more time would be necessary to name
the color of suicide-related words than to name the color of neutral,
positive, and negative words. Such attentional bias was demon-
strated among individuals who had recently attempted suicide but
not in a control group. This bias was not associated with an
increase in the accessibility of negative emotional content, but was
positively correlated with self-reported suicidal ideation.
Similarly, Nock and Banaji (2007) recently developed the Self-
Injury Implicit Association Test to measure how strongly individ-
uals associate self-injury with themselves. The strength of this
association is measured through reaction times on a computerized
test (for a detailed description of the procedure for this test, see
Greenwald, McGhee, & Schwartz, 1998). The results indicated
that adolescents who had recently engaged in self-injury showed a
positive association between self-injurious concepts (e.g., cutting)
and self-referent words (e.g., me), whereas adolescents who were
not self-injurious showed a negative association. In addition,
scores on the Self-Injury Implicit Association Test significantly
predicted recent suicidal ideation and attempts, as well as suicidal
ideation at a 6-month follow-up. These findings suggest that there
are individual differences in how individuals process suicide-
relevant information, reflecting personal history of suicidal behav-
ior.
In the present research, we used experimental designs to study
suicide-thought accessibility in a theoretically integrated manner.
In so doing, we relied primarily on escape theory (Baumeister,
1990, 1991a), an important theory of suicidal behavior, but we also
integrated insights from several other relevant theories.
Escape Theory
Escape theory is an integrative framework proposing a causal
chain that leads to suicide. Influenced by the work of French
sociologists Durkheim (1897), Halbwachs (1930), and Baechler
(1975), escape theory integrates important notions from personal-
ity and social psychology, such as self-discrepancy, self-
awareness, and attribution, among others. Its main assumption is
that individuals do not attempt suicide because they desire death
but rather because they are motivated to escape from self. If they
could be someone else, be somewhere else, have another life, not
be tormented by existential problems, they would not attempt to
end their lives. Thus, escape theory suggests that suicide is best
conceptualized as being motivated by the desire to escape the self.
In this framework, suicide involves an orderly progression
through six distinct stages, and a suicide attempt is expected if
alternative means to escape from the six-stage cycle are not found.
The initial step is a major disappointment or failure that is due to
events falling severely short of expectations or standards. This may
occur because standards are too high, because events are very
negative, or both. It is the discrepancy between the standard and
reality that is crucial (see also Higgins, 1987). The second stage
involves self-blame: Responsibility for failure is attributed to the
self. The following steps successively involve high self-awareness,
negative feelings and affect, a state of cognitive deconstruction,
and finally several consequences of cognitive deconstruction that
may increase the appeal of suicide, such as a lack of inhibitions,
rejection of meaning, and irrationality.
A variety of indirect evidence provides support for escape
theory (for a comprehensive review, see Baumeister, 1990). In
particular, different observations tend to corroborate the view that
a major disappointment or defeat in attaining important standards
plays a major role in precipitating suicidal behavior (see also
Williams, 1997, 2001; Williams & Pollock, 2001). Baumeister and
his colleagues also found evidence for some specific mechanisms
588 CHATARD AND SELIMBEGOVIC
´
(Twenge, Catanese, & Baumeister, 2002, 2003) and applied escape
theory to a number of other self-destructive behaviors, including
alcoholism, masochism, and binge eating (Baumeister, 1988;
Heatherton & Baumeister, 1991). Escape theory is a prominent
theory of suicidal behavior (e.g., Cornette et al., 2000; O’Connor,
2003). However, because of the difficulty of studying self-
destructive behavior, direct evidence for the theory is rather lim-
ited.
In the present research, we aimed to contribute to escape theory
by specifying when suicidal thoughts emerge in the six-stage
cycle, an issue insufficiently addressed by the theory. In the initial
formulation of escape theory, suicidal thoughts are assumed to
emerge in the final stages of the suicidal cycle, in a state of
irrationality and disinhibition that makes drastic measures seem
more acceptable. As illustrated by our introductory quotation,
however, the relation between failure and suicidal thoughts could
be much more direct and straightforward than previously expected.
Escape Theory Revisited
Our theoretical proposition can be considered as an elaboration
on escape theory. Baumeister (1990) predicted that individuals
attempt suicide when they fail to attain important standards. Trans-
posing this reasoning from behavior to cognition, we argue that
individuals come to think of suicide when they realize that they fail
to attain important standards. Indeed, our proposition can be sum-
marized in three main points: (a) failure to attain standards in-
creases motivation to escape from negative self-awareness; (b)
when individuals are motivated to escape the self, the means to
reach that goal are activated; and (c) suicide is a very efficient
means of escaping from self. These three points are briefly devel-
oped next.
According to Duval and Wicklund (1972), self-awareness is a
state of self-focused attention in which individuals mentally com-
pare their current self-perception (the real self) with the person
they ultimately want to become (personal standards, or the ideal
self). When comparisons with standards yield undesirable results,
self-awareness is postulated to motivate either discrepancy reduc-
tion (a fight response) or avoidance of the self-aware state (a flight
response). The adopted behavioral response depends on a number
of factors, such as importance and permanence of the discrepancy
(for a review, see Silvia & Duval, 2001; see also Morin, 2002,
2003). Generally speaking, a large discrepancy that cannot be
quickly reduced is most conducive to escapist behaviors.
In the 1970s, researchers used mirrors and cameras to explore
motivations to escape negative self-awareness. In some experi-
ments, participants were quicker to leave a room after receiving a
negative (rather than positive) evaluation, especially if there was a
mirror in the room (Duval & Wicklund, 1972). Similarly, partic-
ipants were particularly inclined to escape self-awareness (again,
this meant leaving a room with a mirror) when they were led to
believe, after a negative evaluation, that their failure would be
permanent (Steenbarger & Aderman, 1979). In another study, male
participants showed a desire to avoid self-related cues after being
negatively evaluated by an attractive woman (Gibbons & Wick-
lund, 1982). In summary, prior research has indicated that negative
comparisons between self and standards often induce motivation to
escape negative self-awareness.
With regard to the second point, the idea that motivation to
escape the self can trigger means to escape is consistent with goal
systems theory (Kruglanski et al., 2002). In this framework, mo-
tivation and cognition work together rather than as separate entities
such that when a goal is activated, means to that goal become
automatically accessible. Research tends to confirm this central
assumption (see Kruglanski et al., 2002). Moreover, it suggests
that the means with the greatest utility to reach the desired goal is
the most likely to be activated, even if the presence of competing
goals can hinder this activation (Shah & Kruglanski, 2002). Thus,
research on goal systems theory suggests that the motivation to
escape from self-awareness can trigger the means to attain this
specific goal.
Third, suicide has long been considered as a means of escape
from aversive self-awareness (for a review, see Baumeister, 1990).
Among different means of escape (television, spirituality, alcohol-
ism, etc.), it is certainly not the most pleasant one, but it can be
considered as the most radical. In effect, with suicide, one no
longer needs to face reality, reduce the discrepancy, or attain
unattainable goals or standards. The problem of negative self-
awareness is definitely eliminated. Thus, from a purely rational
viewpoint (Baechler, 1975), suicide is a perfectly efficient way to
attain the goal of escape. In the perspective of goal systems theory
(Kruglanski et al., 2002), the means of suicide has great utility in
reaching the goal of escape.
These three main points suggest that when individuals realize
that they fail to attain an important standard, they may be moti-
vated to escape negative self-awareness, which could lead thoughts
of suicide to become particularly accessible.
Death- and Suicide-Thought Accessibility
Because death and suicide are semantically related, an increase
in the accessibility of suicide-related thoughts can also generate a
parallel increase in death-thought accessibility (through spreading
of activation). However, our reasoning based on escape theory
suggests that when individuals fail to attain standards, they would
think about suicide as a means of escape, rather than as something
terrifying (i.e., associated with death). Thus, failure would affect
suicide-thought accessibility first. In other words, the spreading of
activation would go from suicide- to death-related thoughts rather
than the other way around. This suggests that failure would affect
suicide- more than death-thought accessibility immediately after
confrontation with failure, and possible effects of failure on
suicide-thought accessibility would not be accounted for by death-
thought accessibility.
Different predictions can be drawn from terror management
theory (TMT; Solomon, Greenberg, & Pyszczynski, 1991). TMT
is not a theory of suicidal behavior, but it is a comprehensive
framework that seeks to understand, explain, and predict how
people manage the fear of death. It posits that psychological
responses to reminders of death (mortality salience) involve prox-
imal and distal defensive mechanisms (Pyszczynski, Greenberg, &
Solomon, 1999). Proximal defenses are initiated when the idea of
death is in focal attention and entail denial and suppression of
death-related thoughts. Distal defenses emerge when death-related
thoughts are highly accessible but are not in focal attention (i.e.,
after a delay) and entail a tripartite security system consisting of
cultural worldviews, self-esteem, and close relationships (Hart,
589
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
Shaver, & Goldenberg, 2005). According to TMT, these psycho-
logical structures serve as buffers against death anxiety by provid-
ing feelings of continuity, stability, permanence, and literal or
symbolic immortality.
Considerable research supports the basic tenets of TMT (for a
review, see Pyszczynski, Greenberg, Solomon, Arndt, & Schimel,
2004). In particular, studies have indicated that mortality salience,
as compared with control conditions, elicits proximal reactions,
such as suppression of death-related thoughts, but also elicits distal
reactions, such as worldview defense and self-esteem striving.
More recently, studies have also suggested that death-related
thoughts tend to become more accessible when cultural world-
views, self-esteem, or close relationships are threatened. For ex-
ample, in some experiments, individuals who identified highly
with their nation showed increased accessibility of death-related
thoughts after reading an essay that criticized their nation
(Schimel, Hayes, Williams, & Jahrig, 2007). In the same way,
negative feedback on students’ intelligence increased death-
thought accessibility among students for whom intelligence was an
important source of self-esteem (Hayes, Schimel, Faucher, &
Williams, 2008). The mere imagination of a separation with a
closely related person increased death-thought accessibility among
individuals with an insecure attachment style (Mikulincer, Florian,
Birnbaum, & Malishkevich, 2002). In the TMT literature, in-
creases in death-thought accessibility are interpreted as indicating
a relatively unconscious fear of one’s death (see Hayes, Schimel,
Arndt, & Faucher, 2010).
TMT suggests that failure to attain an important standard can
undermine the existential protection that self-esteem confers, thus
leading to increases in death-thought accessibility (reflecting fear
of death). However, because suicide is clearly not compatible with
the fear of death (e.g., Heine, Proulx, & Vohs, 2006; Muraven &
Baumeister, 1997), TMT would not predict increases in suicide-
thought accessibility after failure (even if death and suicide are
semantically related). Indeed, such effects would be clearly incom-
patible with the notion of fear of death, central in TMT. Thus,
escape theory and TMT lead us to formulate different predictions.
TMT suggests that failure to attain an important standard would
increase death- but not suicide-thought accessibility, whereas es-
cape theory suggests that failure would increase suicide- more than
death-thought accessibility. Because suicide-thought accessibility
has never been assessed in previous TMT studies, it remains to be
seen which of these two predictions (if any) is supported by
experimental evidence.
The Present Research
In the current research, we were mainly interested in the cog-
nitive accessibility of suicide-related thoughts (an understudied
issue), but death-thought accessibility was also assessed in most of
our studies to disentangle effects of failure on these two kinds of
thoughts. We tested the hypothesis derived from escape theory that
failure (i.e., negative comparisons between self and standards) can
lead to increases in suicide-thought accessibility.
1
We expected
this effect to reflect motivations to escape from negative self-
awareness, rather than the fear of death. Thus, it should be mod-
erated by escapist motivations and self-consciousness, such that it
should be especially pronounced when associated with high levels
of these constructs. It should also depend on the discrepancy
between self and standards, a large discrepancy being the most
conducive to suicide-thought accessibility after failure. In addition,
this effect should be accompanied by increased accessibility of
more general concepts related to escape. It should not, however, be
accompanied by increased death-thought accessibility (the fear of
death), and priming the idea of death should not increase suicide-
thought accessibility. Six studies were conducted to examine these
predictions.
In Study 1, we examined whether priming failure could increase
suicide-thought accessibility compared with different control con-
ditions, including a death-related prime condition. We also ex-
plored whether individual differences in self-consciousness and
escapist motivations moderate effects of failure on suicide-thought
accessibility. In Studies 2 through 4, we examined the assumption
that the discrepancy between self and standards moderates effects
of failure on suicide-thought accessibility. In Study 5, priming
failure was expected to increase both suicide-thought accessibility
and the desire for an altered state of consciousness. Finally, in
Study 6, we examined whether failure to attain an unattainable
standard might increase suicide-thought accessibility, and whether
this effect is accompanied by a parallel increase in accessibility of
more general concepts related to escape.
Study 1: Self-Consciousness, Escapist Motivations,
and Suicide-Thought Accessibility
In Study 1, we tested the hypothesis that failure to attain an
important standard can cause thoughts of suicide to become more
accessible, compared with conditions involving no prime or a
death-related prime (Hypothesis 1). In this study, we also at-
tempted to identify individual factors and motivations associated
with increased accessibility of suicide-related thoughts after fail-
ure. Our theoretical reasoning entails that this effect would reflect
motivations to escape from negative self-awareness. Hence, this
effect should be especially pronounced among participants with
relatively high levels of escapist motivations and high levels of
self-consciousness (Hypothesis 2). Such moderation by self-
consciousness (or chronic self-awareness) and escapist motiva-
tions would be very consistent with our theoretical reasoning.
Previous terror management research has shown that following
mortality salience, there is an effortful suppression of death-related
thoughts, reducing death-thought accessibility (Greenberg, Arndt,
Schimel, Pyszczynski, & Solomon, 2001). Research also suggests
that this basic defensive strategy is especially pronounced when it
is associated with high levels of self-awareness (Arndt, Greenberg,
Simon, Pyszczynski, & Solomon, 1998). Thus, in line with TMT,
we predicted that a death-related prime, compared with a control
condition, would reduce death-thought accessibility among partic-
ipants high in self-consciousness (Hypothesis 3). A death-related
prime should increase death-thought accessibility only among par-
ticipants with low levels of self-consciousness. Such moderation
by self-consciousness would be consistent with terror management
theorizing and research.
In summary, we expected that different conditions and motiva-
tions would lead to increases (or decreases) in death- and suicide-
1
Ethical approval for the present studies was obtained from the Univer-
sity of Geneva Research Ethics Committee, and participants gave their
written consent.
590 CHATARD AND SELIMBEGOVIC
´
thought accessibility. In line with TMT, we expected a death-
related prime (compared with a control condition involving no
prime) to provoke typical proximal reactions predicted by and
documented in the terror management literature (i.e., a suppression
of death-related thoughts), especially among participants high in
self-consciousness. In line with our predictions derived from es-
cape theory, we also expected that priming failure (compared with
a control condition involving no prime and a condition involving
a death-related prime) would lead to increased accessibility of
suicide-related thoughts, especially among participants high in
both self-consciousness and escapist motivations.
Methods
Participants. Eighty-four students (42 female, 40 male, and
two gender-unspecified participants) from the University of Ge-
neva took part in this study (M
age
26.12 years, SD 6.37).
Procedure and materials. Participants were tested individ-
ually by a research assistant. They received a booklet (in French)
including all the materials. They first completed measures of
self-consciousness, escapist motivations, and desirability of con-
trol. This last measure was included to test a hypothesis alternative
to our predictions. According to this hypothesis, increased acces-
sibility of suicide-related thoughts after failure would reflect mo-
tivations to restore a sense of control, rather than motivations to
escape negative self-awareness. This possibility was suggested by
recent research indicating that the desire to restore a generalized
perception of control plays a major role in how people respond to
existential threats (Fritsche, Jonas, & Fankha¨nel, 2008). Next,
participants were randomly assigned to one of three conditions: a
control condition including no prime, a death-related prime con-
dition, and a failure-related prime condition. Finally, participants
completed a word completion task assessing death- and suicide-
thought accessibility.
Self-consciousness. Individual differences in private self-
consciousness (nine items; e.g., “I’m always trying to figure my-
self out”) and public self-consciousness (seven items; e.g., “I’m
usually aware of my appearance”) were assessed using Scheier and
Carver’s (1985) Self-Consciousness Scale (French version by Pel-
letier & Vallerand, 1990). Participants responded to each statement
on a 4-point scale (0 not like me at all, 3a lot like me). We
averaged private and public self-consciousness scores to indicate
global self-consciousness, ␣⫽.81. High scores on this measure
indicate high self-consciousness (or chronic self-awareness).
Escapist motivations. We used nine items of the COPE
Inventory (Carver, Scheier, & Weintraub, 1989) reflecting escapist
tendencies in the face of stress. Participants were asked to indicate
how they deal with negative and stressful life events in general by
rating a number of possible strategies on 4-point Likert scales (1
I usually don’t do this at all, 4I usually do this a lot). We
selected items assessing renouncement or disengagement (four
items: “I admit to myself that I can’t deal with it, and quit trying”;
“I just give up trying to reach my goal”; “I give up the attempt to
get what I want”; “I reduce the amount of effort I’m putting into
solving the problem”), mental avoidance or escape (four items: “I
turn to work or other substitute activities to take my mind off
things”; “I daydream about things other than this”; “I sleep more
than usual”; “I go to movies or watch TV to think about it less”),
and alcohol/drug substance use (one item: “I drink alcohol or take
drugs in order to think about it less”). These items showed good
interitem correlations and satisfactory internal reliability, ␣⫽.70.
They were thus averaged to form a global score of escapist
motivations. Participants scoring high on this measure show a high
inclination to escape when confronted with negative and stressful
life events.
Desirability of control. Individual differences in motivation
to control the events in one’s life were assessed with the Desir-
ability of Control Scale (Burger & Cooper, 1979). Representative
items of this 20-item instrument are as follows: “I prefer a job
where I have a lot of control over what I do and when I do it” and
“I try to avoid situations where someone else tells me what to do.”
Participants responded to each statement on a 7-point scale (1
this statement doesn’t apply to me at all,7this statement always
applies to me). This scale showed good internal reliability, ␣⫽.72
(three items were excluded because of low interitem correlations).
High scores indicate high motivation to exert control in one’s life.
Experimental conditions. In the death-related prime condi-
tion, participants were asked to describe, as precisely as possible,
the emotions and feelings that the idea of living during war elicited
in them. They also indicated what they would do if they were
confronted with such a situation one day. In the failure-related
prime condition, participants were asked to describe, as precisely
as possible, the emotions and feelings that the idea of failing to
find a job after college (and becoming poor as a consequence)
elicited in them. They also indicated what they would do if they
were confronted with such a situation one day. Participants in the
priming conditions were given 3 min to complete the task. In the
control condition, participants did not complete any tasks.
Construct accessibility. Participants next completed a word
completion task, adapted from previous TMT studies (see Hayes et
al., 2010), to assess accessibility of concepts related to death and
suicide. Participants were asked to complete 24 word fragments
(e.g., LI_ _ _), presented in the same order to all participants, such
as to form a word (e.g., LIVRE,which means “book” in French).
Intertwined in the list were four fragments that could be completed
either as neutral or as death-related (target words were tombeau
grave,mort [dead/death], cercueil [coffin], and de´ce`s [decease]),
and four fragments that could be completed as neutral or as
suicide-related (target words were suicide [suicide], pendre [to
hang], corde [rope], and veine [wrist]). The words meaning rope,
to hang, and wrist were used because they refer to the most
common methods of suicide in Europe (Va¨rnik et al., 2009). As in
previous research (e.g., Maxfield et al., 2007), to control for
individual differences in the number of word fragments completed,
we computed percentages of death- and suicide-related comple-
tions by dividing the number of death- and suicide-related words
completed by the total number of word fragments completed and
multiplying the result by 100. After the experiment, participants
were thanked and debriefed.
Results
Data were screened for outliers (following McClelland, 2000),
but no atypical observations were found in this study (for descrip-
tive statistics, see Table 1). Table 2 presents means and standard
deviations for the proportions of death- and suicide-related com-
pletions within each condition.
591
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
To test Hypotheses 1 and 2, we conducted a regression analysis
on the proportion of suicide-related completions. We created two
contrast-coded variables: one contrasting the failure condition with
the other two conditions (no prime and death-related prime con-
ditions) and the other contrasting the no prime condition with the
death-related prime condition. We next examined main effects of
the two contrast-coded variables, of self-consciousness, and of
escapist motivations, as well as all two-way and three-way inter-
actions (product terms) among these variables.
Regressions conducted on suicide-thought accessibility revealed
a main effect of the contrast opposing the failure condition and the
other two conditions after controlling for all other variables, ␤⫽
.24, t(69) 2.30, p.05. The contrast opposing the no prime
condition and the death-related prime condition was not signifi-
cant, ␤⫽–.05, t(69) –.47, ns. Similarly, there were no main
effects for self-consciousness, ␤⫽.18, t(69) 1.55, ns,or
escapist motivations, ␤⫽.18, t(69) 1.65, ns. There were no
significant two-way interactions ( ps.10), but there was a
significant three-way interaction involving the contrast opposing
the failure condition and the other two conditions, self-
consciousness, and escapist motivations, ␤⫽.25, t(69) 2.09,
p.05.
To decompose this interaction, we computed the effect of the
contrast opposing the failure condition and the other two condi-
tions at plus or minus one standard deviation from the means of
self-consciousness and escapist motivations (following Aiken &
West, 1991). These analyses revealed that the difference between
the failure condition and the other two conditions was most pro-
nounced at high levels of both self-consciousness and escapist
motivations, ␤⫽.59, t(69) 2.90, p.01. This difference was
not significant at low levels of self-consciousness and/or escapist
motivations (ps.10). Thus, in line with Hypotheses 1 and 2, the
failure prime increased suicide-thought accessibility compared
with the other two conditions, and this effect was the most pro-
nounced among persons scoring high in both self-consciousness
and escapist motivations.
In examining Hypothesis 3, we ran the same regression analysis
on the proportion of death-related completions. In this analysis,
only the two-way interaction between self-consciousness and the
contrast opposing the no prime condition and the death-related
prime condition reached significance, ␤⫽–.39, t(69) –3.59,
p.001. This interaction indicated that the death-related prime
condition, compared with the no prime condition, increased death-
related completions among participants scoring low in self-
consciousness (computed at –1 SD), ␤⫽.47, t(69) 2.91, p
.01, but decreased death-related completions among participants
scoring high in self-consciousness (computed at 1 SD), ␤⫽–.37,
t(69) –2.41, p.05. These findings fit Hypothesis 3. A
death-related prime reduced death-thought accessibility among
participants high in self-consciousness (in line with the notion of
suppression).
The significant three-way interaction found on suicide-related
completions remained statistically significant, ␤⫽.24, t(69)
2.01, p.05, after controlling for death-related completions. In
the same way, the significant two-way interaction found on death-
related completions remained statistically significant, ␤⫽–.42,
t(69) –3.86, p.001, after controlling for suicide-related
completions. Thus, effects found on suicide-related completions
were not mediated by death-thought accessibility, and vice versa.
A regression analysis on the difference between death- and
suicide-thought accessibility was conducted to test whether the
effects obtained on these two kinds of thought differed. Two
significant interactions were revealed. There was a two-way inter-
action between self-consciousness and the contrast opposing the
no prime and the death-related prime condition, ␤⫽.39, t(69)
3.52, p.001. There was also a three-way interaction involving
self-consciousness, escapist motivations, and the contrast opposing
the failure condition and the other two conditions, ␤⫽.27, t(69)
2.30, p.05. Thus, effects found on death-thought accessibility
were specific to death-related thoughts, and effects found on
suicide-thought accessibility were specific to suicide-related
thoughts.
Table 1
Descriptive Statistics in Study 1
Variable M(SD)
Correlation
Self-consciousness
Escapist
motivations
Control
motivations
Death-related
completions
Suicide-related
completions
Self-consciousness 2.11 (0.37) .18
.17
.09 .16
Escapist motivations 1.96 (0.52) .01 .11 .15
Control motivations 4.96 (0.70) .05 .06
Death-related completions 5.56 (4.82) .16
Suicide-related completions 4.72 (4.20)
p.10.
Table 2
Means and Standard Deviations for Death- and Suicide-Related
Completions in Study 1
Completions
Condition
No prime
Failure
prime
Death-
related
prime
Suicide-related 4.18 (4.11) 6.25 (4.75) 3.54 (3.00)
Death-related 5.19 (5.21) 6.32 (4.24) 5.12 (5.07)
Note. Standard deviations are in parentheses. In an analysis of variance,
the effect of condition was significant on suicide-related completions, F(2,
78) 3.14, p.05,
2
.07, but not on death-related completions, F(2,
78) 1, ns,
2
.01.
592 CHATARD AND SELIMBEGOVIC
´
Finally, we examined whether control motivations moderated
the effect of our contrast-coded variables on death- and suicide-
related completions. Death-thought accessibility was regressed on
the two contrast-coded variables, control motivations, and the
product terms between each contrast-coded variable and control
motivations. In this analysis, we found no significant effects ( ps
.10). On suicide-thought accessibility, we found only a significant
effect of the contrast opposing the failure condition and the other
two conditions, ␤⫽.26, t(75) 2.26, p.05. Thus, control
motivations did not moderate the effect of experimental condition
on death- and suicide-thought accessibility.
Discussion
The results of Study 1 were in line with our predictions derived
from escape theory. Overall, priming failure increased suicide-
thought accessibility compared with control conditions involving
no prime or a death-related prime. Consistent with the idea that this
effect reflects a desire to escape from self, it was the most pro-
nounced among participants scoring high in both self-
consciousness and escapist motivations. In addition, in line with
the idea that failure affects suicide-thought accessibility first, these
effects were more pronounced on suicide- than on death-thought
accessibility. Indeed, they were specific to suicide-thought acces-
sibility. In this study, we found no evidence for the prediction
derived from TMT that failure increases death-thought accessibil-
ity. However, in line with TMT, priming the idea of war (vs. no
prime) reduced death-thought accessibility among participants
scoring high in self-consciousness. It seems that when these indi-
viduals are confronted with the idea of war, they reflect on them-
selves, which induces typical defensive reactions predicted by
TMT (i.e., a suppression of death-related thoughts). In contrast,
priming the idea of war (vs. no prime) increased death-thought
accessibility among participants low in self-consciousness. Fi-
nally, Study 1 results did not support the view that increased
accessibility of suicide-related thoughts after failure reflects a
desire to restore a sense of control.
Study 2: The Dark Side of High Cultural Standards
Study 1 provided experimental evidence in favor of our predic-
tions that failure to meet standards is associated with increased
accessibility of suicide-related thoughts and that this effect reflects
a desire to escape negative self-awareness. To further substantiate
this claim, in Studies 2 through 4, we examined whether the
discrepancy between self and standards moderates effects of fail-
ure on suicide-thought accessibility. According to escape theory,
when the discrepancy between self and standards becomes larger,
the desire to escape the self becomes stronger. This suggests that
failure would increase suicide-thought accessibility more when it
is associated with a high rather than a low discrepancy between
self and standards.
In line with this idea, in Study 2 we tested the hypothesis that the
same failure (unemployment and poverty) would have different
implications in a country with high standards of living (high gross
domestic product, high employment rates, etc.) than in a country
with lower standards. The discrepancy between self and standards
is larger when individuals fail to reach high rather than low
standards. The ideas of unemployment and poverty may thus
provoke a greater increase in suicide-thought accessibility in a
country with high standards of living, where almost everyone is
employed, than in a country with lower standards of living, where
unemployment and poverty are rather common.
To test this hypothesis, we conducted a study with students from
Switzerland and Coˆ te d’Ivoire. These two countries strongly differ
on a number of relevant economic indicators. For instance, the
gross domestic product per capita in 2008 was estimated at
$40,900 in Switzerland and at $1,700 in Coˆ te d’Ivoire (Central
Intelligence Agency, 2009). The same year, the unemployment
rate in Switzerland was estimated at 3% and at 40% to 50% in Coˆte
d’Ivoire (Central Intelligence Agency, 2009). Given these differ-
ences, and somewhat counterintuitively, we expected ideas of
unemployment and poverty to provoke a stronger increase in
suicide-thought accessibility among Swiss than among Ivorian
students.
Methods
Participants. Participants were 44 students from the Univer-
sity of Geneva, Switzerland, and 60 students from the University
of Abidjan, Coˆ te d’Ivoire. There were 22 male and 22 female
students in Switzerland (M
age
24.41, SD 4.39) and 50 male
and 10 female students in Coˆ te d’Ivoire (M
age
25.86, SD
3.21).
Procedure and materials. The Swiss sample was recruited
by a research assistant on the campus of the University of Geneva.
The sample from Coˆ te d’Ivoire was recruited by a research assis-
tant on the campus of the University of Abidjan. Participants
received a booklet (in French) containing all the necessary mate-
rials and instructions for the experiment. They rated the national
standards of their country before being primed or not with unem-
ployment/poverty. Finally, they completed a thought accessibility
measure, similar to the one used in Study 1.
National standards. Participants were provided with a list of
10 propositions concerning standards of living in their countries,
for example, “Standards of living are high in Switzerland (vs. Coˆte
d’Ivoire)”; “Switzerland (vs. Coˆ te d’Ivoire) is a wealthy country”;
“It is easy to find a job in Switzerland (vs. Coˆ te d’Ivoire).” They
indicated their agreement with each statement on a 7-point Likert
scale (1 completely disagree,7completely agree). Scale
reliability was judged satisfactory in Switzerland (␣⫽.72) and in
Coˆ te d’Ivoire (␣⫽.70).
Experimental conditions. Participants were randomly as-
signed to one of two conditions, similar to the no prime and the
failure condition used in Study 1. In the failure prime condition,
participants were asked to respond to the following statements:
“Please describe, in a few words, the feelings and thoughts that the
idea of failing to find a job after University arises in you” and
“Imagine and describe, as precisely as possible, what would hap-
pen to you if you became poor in the long term.” For each
question, a blank space for answers was provided. In the no prime
condition, this task was simply omitted.
Construct accessibility. Next, participants completed a word
completion task similar to the one used in Study 1. There were four
fragments that could be completed either as neutral or as death-
related (target words were grave,dead/death, coffin, and decease)
and four fragments that could be completed either as neutral or as
suicide-related (target words were suicide, to hang, rope, and
593
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
wrist). To control for possible effects of experimental condition on
negatively valenced thoughts, in this study we also included four
fragments that could be completed as neutral or as sadness-related
(targets words were e´chec [failure], de´prime [depression], triste
[sad], and chagrin [grief]). Finally, 12 fragments could be com-
pleted only as neutral. Proportions of death-, suicide-, and sadness-
related completions were calculated as in Study 1. After the
experiment, participants were thanked and debriefed.
Results
The death- and sadness-thought accessibility scores contained
one atypical observation each (with a zscore greater than 3 SD);
these observations were excluded from the analyses. As expected,
cultural standards were perceived as being higher in Switzerland
(M4.98, SD 0.70) than in Coˆ te d’Ivoire (M3.30, SD
0.89), F(1, 102) 106.28, p.001,
2
.51. Thus, participants
were quite aware of the low or high cultural standards of their
countries.
To test our hypothesis, we conducted a 2 (country: Switzerland
vs. Coˆ te d’Ivoire) 2 (condition: prime vs. no prime) 3 (word
fragments: death vs. sadness vs. suicide) analysis of variance
(ANOVA) on the proportion of word fragments completed, with
repeated measures on the last variable. In this analysis, there was
a main effect of the repeated variable, F(2, 96) 9.65, p.001,
2
.09. Overall, participants completed more fragments as
related to death (M7.55, SD 4.96) and suicide (M7.51,
SD 4.88) than as related to sadness (M5.14, SD 4.06).
There was also a significant three-way interaction, F(2, 96)
3.38, p.04,
2
.03 (see Table 3 for the means). To examine
the nature of this interaction, we performed separate ANOVAs on
death-, sadness-, and suicide-related completions.
Country had a significant effect on the proportion of death-
related completions, F(1, 98) 8.41, p.01,
2
.08, indicating
that participants from Coˆ te d’Ivoire (M8.73, SD 4.46) had
more death-related completions than participants from Switzerland
(M5.93, SD 5.14). There was no effect of prime, F(1, 98)
1, ns, and no interaction, F(1, 98) 1.23, ns.
There were no significant effects of condition, F(1, 98) 1, ns,
or of country, F(1, 98) 1.56, ns, on the proportion of sadness-
related completions, and there was no significant interaction, F(1,
98) 1, ns.
There were no main effects of country, F(1, 99) 1, ns,orof
prime, F(1, 99) 2.50, p.11,
2
.02, on the proportion of
suicide-related completions. However, there was a significant in-
teraction between country and prime, F(1, 99) 5.61, p.02,
2
.05 (see Table 3). Planned comparisons indicated that, as
expected, Swiss participants provided more suicide-related com-
pletions when primed with unemployment/poverty than in the no
prime condition, t(99) 2.61, p.02. In contrast, the prime had
no effect on participants from Coˆ te d’Ivoire, t(99) – 0.60, ns.
There were no other significant differences ( ps.05).
In an analysis of covariance, the interaction between country
and condition on suicide-related completions remained statistically
significant, F(1, 95) 6.75, p.02,
2
.06, after controlling
for death- and sadness-related completions.
Discussion
The results of Study 2 were consistent with our expectations.
Unemployment and poverty seem to have different psychological
implications for people from countries with high and low standards
of living. Participants from Switzerland, a country in which un-
employment and poverty are relatively rare, showed greater acces-
sibility of suicide-related concepts when primed with unemploy-
ment and poverty than when they were not primed (replicating
thereby the results found in Study 1). In contrast, participants from
Coˆ te d’Ivoire, a country in which unemployment and poverty are
much more common, were unaffected by the prime. These findings
are in line with the contention that high cultural standards at the
national level may have a dark side. In line with escape theory,
high standards seem to create a burden on the self and a high
inclination to escape in case of failure.
As in Study 1, failure had only small effects on death-related
completions, and effects of failure were specific to suicide-related
completions. Participants from Coˆ te d’Ivoire showed greater ac-
cessibility of death-related thoughts than participants from Swit-
zerland, but this effect was not qualified by a significant interac-
tion with the condition. The recent civil war in Coˆ te d’Ivoire may
account for this difference. Whatever the reason, these findings
provided initial support for our hypothesis that the discrepancy
between self and standards moderates effects of failure on suicide-
thought accessibility. We further examined this hypothesis in
Study 3.
Study 3: One Man’s Meat Is Another Man’s Poison
In Study 3, we examined the idea that it is the discrepancy
between self and standards, rather than the negativity of the prime
per se, that is crucial in increasing suicide-thought accessibility
after failure. In previous studies, we used only negative primes to
induce failure. Thus, it is possible that only such primes lead to
increases in suicide-thought accessibility. Our reasoning based on
escape theory suggests, however, that even a positive prime could
increase suicide-thought accessibility to the extent that it renders
salient a discrepancy from an important standard. To examine this
possibility, we confronted Swiss participants with their country’s
very high standard for life satisfaction. In the experimental con-
dition, participants read an essay indicating that more than 90% of
Swiss citizens are satisfied with their living conditions. This essay,
which reported only positive information (i.e., high levels of
Table 3
Means and Standard Deviations for the Proportion of Death-,
Sadness-, and Suicide-Related Completions in Study 2
Completions
Switzerland Coˆte d’Ivoire
No prime
Failure
prime No prime
Failure
prime
Death-related 5.44 (5.22) 6.48 (5.26) 9.26 (3.97) 8.20 (4.91)
Sadness-related 4.84 (4.84) 4.32 (4.21) 5.22 (3.82) 5.89 (3.57)
Suicide-related 5.39 (5.12) 9.60 (4.07) 8.11 (4.89) 7.02 (4.53)
Note. Standard deviations are in parentheses. An analysis of variance
revealed a significant interaction between country and condition on
suicide-related completions but not on death- or sadness-related comple-
tions (see text).
594 CHATARD AND SELIMBEGOVIC
´
satisfaction in various domains), was not included in the control
condition. Even if the general tenor of the essay was positive, it
represented a failure for participants who did not feel particularly
happy. When these individuals realize their failure to attain their
country’s high national standard for life satisfaction, they may be
motivated to escape the self, which could cause thoughts of suicide
to become more accessible. We thus expected confrontation with
the high national standard to induce an increase in suicide-thought
accessibility among Swiss participants relatively low (vs. high) in
subjective happiness.
Methods
Participants. Participants were 80 University of Geneva un-
dergraduates (M
age
22.64, SD 5.51; 69 female and 11 males;
all Swiss native), who took part in the study collectively at an
educational science course.
Procedure and materials. Participants first completed a
measure of subjective happiness. Next, they read (or not) an essay
reporting levels of life satisfaction in their country and completed
a construct accessibility measure.
Subjective happiness. We assessed subjective happiness with
the scale developed and validated by Lyubomirsky and Lepper
(1999). This measure consists of four items (␣⫽.77), and the
associated 7-point Likert response scales vary in their anchoring
point labels. An illustrative item is “In general, I consider myself
. . .” with the associated response scale anchored 1 not a very
happy person and 7 a very happy person. High scores indicate
high subjective happiness (M4.76, SD 1.12).
Experimental condition. Next, participants in the experimen-
tal condition were exposed to an essay entitled, “Life Conditions in
Switzerland: More Than 90% Satisfied.” This essay reported the
results of a survey indicating that more than 90% of Swiss citizens
were highly satisfied with their working conditions, their balance
between professional and private life, and their housing conditions.
Inspired by a real newspaper article, the essay concluded that
“Swiss citizens have all good reasons to be happy.” In the control
condition, the essay was omitted. Participants were randomly
assigned to one of the two conditions.
Construct accessibility. The accessibility of suicide-, death-,
and sadness-related concepts was assessed and calculated as in
Study 2. After the experiment, participants were thanked and
debriefed.
Results
Suicide-thought accessibility scores contained one atypical ob-
servation (with a zscore greater than 3 SD), which was excluded
from the analyses. To test our hypotheses, we ran regression
analyses to predict our dependent variables from condition (control
coded –1 and experimental coded 1), subjective happiness (con-
tinuous variable, centered), and their product term.
For the proportion of sadness-related completions, there was
only a significant effect of happiness, ␤⫽–.26, t(76) –2.33, p
.03. As might have been expected, the happier the participants
were, the fewer fragments they completed as sadness-related.
For the proportion of death-related completions, there were no
significant effects of condition, ␤⫽.08, t(76) 0.70, ns,of
happiness, ␤⫽.03, t(76) .29, ns, or of their product term, ␤⫽
.04, t(76) 0.36, ns.
For the proportion of suicide-related completions, there were no
effects of condition, ␤⫽.15, t(75) 1.39, ns, or of happiness,
␤⫽–.13, t(75) –1.13, ns. However, the interaction was signif-
icant, ␤⫽–.24, t(75) –2.16, p.04 (Figure 1). Simple slope
tests indicated that, as expected, the effect of prime was significant
among relatively unhappy participants (computed at –1 SD), ␤⫽
.40, t(75) 2.49, p.02, but not among relatively happy
participants (computed at 1 SD), ␤⫽–.09, t(75) – 0.59, ns.
Viewed differently, happiness and the proportion of suicide-related
completions were unrelated in the control condition, ␤⫽.11,
t(75) 0.78, ns, but were negatively related in the priming
condition, ␤⫽–.37, t(75) –2.17, p.04.
The interaction between condition and subjective happiness on
suicide-related completions remained statistically significant, ␤⫽
–.24, t(73) –2.14, p.04, after controlling for sadness- and
death-related completions. In the same way, a regression analysis
was run to predict scores of difference between death- and suicide-
related completions from our key variables, to verify whether
effects on these two indicators of thought accessibility were dif-
ferent. In this analysis, we found a nearly significant interaction
between condition and subjective happiness, ␤⫽–.19, t(75)
–1.71, p.10, indicating that our variables had a greater impact
on suicide- than on death-related completions.
Discussion
Results of this study confirmed that relatively unhappy Swiss
citizens show increased suicide-thought accessibility when con-
fronted with the high national standard of their country in terms of
life satisfaction. In contrast, exposure to this standard did not affect
relatively happy individuals. These findings provide a scientific
account of the familiar contention that “one man’s meat is another
man’s poison.”
The present findings were also consistent with the idea that even
a positive prime can trigger suicide-related thoughts to the extent
that it renders salient a discrepancy between self and standards.
These findings could not be accounted for by increased accessi-
Figure 1. Suicide-related completions as a function of subjective happi-
ness and essay on life satisfaction (Study 3).
595
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
bility of death- and sadness-related thoughts, and they were quite
specific to suicide-related thoughts. We found no significant ef-
fects on death-thought accessibility and found only a main effect
of subjective happiness on sadness-thought accessibility. In Study
4, we set out to conceptually replicate these effects in a different
culture.
Study 4: Disappointment With Capitalism and
Suicide-Thought Accessibility
Study 4 was conducted in the Czech Republic, a country that has
recently undergone substantial and successful changes in its tran-
sition from communism to capitalism (Pehe, 1994). The Czech
Republic is often depicted in the media as a model of successful
transition. Of course, the transition generated high expectations for
improvement in life conditions. However, because transition could
not bring equal improvement to everyone, some individuals are
disappointed by the new capitalist system. In line with our theo-
retical reasoning, when those individuals are confronted with the
idea that the transition is a success (a positive prime), they should
become highly aware of the fact that events fell short of their
expectations and should therefore experience increased suicide-
thought accessibility. This effect should not appear among those
not disappointed with capitalism, as transition brought them the
improvement that they had expected.
In summary, this study could provide converging evidence from
a country with a different cultural background and in a different
language that a positive prime can lead to increases in suicide-
thought accessibility when it renders salient a discrepancy between
self (personal attitude toward capitalism) and an important stan-
dard (the success of the transition).
Methods
Participants. This study was conducted on a heterogeneous
sample of adults (N65). Participants were Czech (27 men, 37
women, and one gender-unspecified participant). Their mean age
was 54.19 (SD 7.78).
Procedure and materials. Participants were recruited by a
research student who contacted her family members, acquain-
tances, neighbors, and her parents’ social network. Participants
were all over 40 years old and had thus known both communist
and capitalist systems. They were given a booklet (in Czech)
containing all the necessary materials and instructions for the
study. Participants rated their disappointment in the capitalist
system, before being exposed (vs. not) to a procapitalist essay
stressing the positive consequences of the transition. Finally, they
completed a construct accessibility measure similar to the one used
in previous experiments.
Disappointment by capitalism. Participants were provided
with a list of propositions concerning the transition from commu-
nism to capitalism. They indicated their agreement with each
proposition on a 7-point Likert scale (1 completely disagree,
7completely agree). Embedded in this list were three proposi-
tions tapping disappointment with capitalism: “I had high expec-
tations for capitalism, but today I’m disappointed”; “Today, I’m
disappointed because capitalism did not keep its promises”; “I’ve
grown disillusioned with capitalism.” Responses on these items
were averaged to form a global score of disappointment with
capitalism (␣⫽.79, M2.90, SD 1.37).
Experimental conditions. In the priming condition, partici-
pants were exposed to an essay, inspired by articles published in
Czech magazines or on the Internet, entitled “For a New Czech
Republic” and ostensibly written by a historian. This essay stressed
the positive consequences of the transition. Participants read the
following:
Since the first hours of its independence, our country has undergone
a major national construction. We have broken all links to the com-
munist history. This was not easy. We have suffered instability,
political and economic crises, and many forms of conflicts. However,
we have successfully faced all these difficulties. We have successfully
changed from communism to capitalism. Nowadays, the Czech Re-
public is one of the most successful transition economies. Since
March 2004, we are members of the European Union, and we will
soon be part of the Euro zone. Henceforth, we make important
advances in all aspects of life (economic, political, social, etc.).
In the control condition, the essay was omitted. Participants were
randomly assigned to one of the two conditions.
Construct accessibility. Participants were presented with 20
word fragments. Six could be completed either as neutral or as
suicide-related, and five could be completed as neutral or as
death-related. The possible suicide-related words were sebevrazˇda
[suicide], povesit se [to hang oneself], provaz [rope], zˇila [wrist],
zastrˇelit se [to shoot oneself], and otra` vit se [to poison oneself].
The possible death-related words were smrt [death], mrtvi [dead],
hrob grave,rakev [coffin], and hrbitov [cemetery]. Other frag-
ments could be completed only as neutral. The proportions of
suicide- and death-related completions were calculated as in pre-
vious studies. After the experiment, participants were thanked and
debriefed.
Results
To test our hypothesis, we ran regression analyses to predict
death- and suicide-related completions from condition (essay
coded 1 and no essay coded –1), disappointment with capitalism
(continuous variable, mean centered), and their product term.
On death-related completions, the regression analysis revealed
no significant effects of condition, ␤⫽.02, t(61) 0.16, ns,of
disappointment with capitalism, ␤⫽.08, t(61) 0.66, ns,orof
the interaction term, ␤⫽–.02, t(61) – 0.16, ns.
On the proportion of suicide-related completions, the effect of
condition was significant, ␤⫽.36, t(61) 2.94, p.01. Overall,
participants completed a greater proportion of fragments as
suicide-related when they were exposed to the essay (M9.68,
SD 5.84) than when they were not (M7.06, SD 4.53). In
addition, disappointment with capitalism was positively related to
the proportion of suicide-related completions, ␤⫽.34, t(61)
2.67, p.02. These main effects were qualified by a significant
interaction, ␤⫽.31, t(61) 2.55, p.02 (Figure 2). Consistent
with our reasoning, simple slope tests revealed that exposure to the
essay increased suicide-related completions in individuals highly
disappointed with capitalism (computed at 1 SD), ␤⫽.70, t(61)
3.60, p.001, but not in those less disappointed with capitalism
(computed at –1 SD), ␤⫽.02, t(61) 0.11, ns. Viewed differ-
ently, disappointment with capitalism was positively related to
suicide-related completions when participants read the essay, ␤⫽
.66, t(61) 3.24, p.01, but not when they did not, ␤⫽.01,
t(61) 0.09, ns.
596 CHATARD AND SELIMBEGOVIC
´
The interaction between disappointment with capitalism and
condition on suicide-related completions remained statistically sig-
nificant, ␤⫽.30, t(61) 2.50, p.02, after controlling for
death-related completions. In the same way, there was a significant
interaction between disappointment and condition on difference
scores between suicide- and death-related completions, ␤⫽.27,
t(61) 2.12, p.04, indicating that our variables had a greater
impact on suicide- than on death-related completions.
Discussion
The results confirmed our expectations. Individuals highly dis-
appointed with capitalism, compared with those less disappointed,
showed greater accessibility of suicide-related thoughts when ex-
posed to an essay stressing the success of the transition to capi-
talism, that is, when they were confronted with the discrepancy
between their disappointment and the cultural success of the tran-
sition. Once again, reminding participants of a globally positive
outcome (i.e., the success of the transition in their country) led to
an increase in suicide-thought accessibility in some participants.
This counterintuitive finding is consistent with escape theory. The
current results provided evidence for our reasoning in a different
culture, on an older population, and in a different language.
Taken together, the findings of Studies 2 through 4 were con-
sistent with the idea that suicide-thought accessibility is especially
pronounced when individuals perceive a large discrepancy be-
tween self and standards. In Study 5, we examined whether failure,
compared with another aversive condition, could increase suicide-
thought accessibility and the desire for an altered state of con-
sciousness. To achieve this aim, we studied a small sample of
marijuana smokers.
Study 5: Failure and the Desire for an Altered
State of Consciousness
People use marijuana or cannabis for several reasons: Some
seem to take pleasure from it, some use it as a medicine for
physical or psychological pains, some smoke to stimulate their
creativity, and so forth. Researchers have argued that some drugs
(e.g., alcohol) can reduce negative self-awareness (Hull, 1981), but
there is no evidence that marijuana can serve such a function.
Thus, in this study, we do not assume that marijuana serves the
purpose of escaping self-awareness. However, smoking marijuana
causes an alteration of brain activity, resulting in an altered state of
consciousness (altered perceptions, judgments, movement coordi-
nation, etc.). Research suggests that the quest for an altered state of
consciousness can be considered as a coping strategy. For exam-
ple, the COPE Inventory (Carver et al., 1989), used in Study 1,
includes one factor called substance use (e.g., “I’ve been using
alcohol or other drugs to think about it less”). Responses on this
item are strongly related to responses on other items that nonam-
biguously assess escapist behaviors (e.g., “I’ve been doing some-
thing to think about it less, such as going to movies, watching TV,
reading, daydreaming, sleeping, or shopping”). Thus, even if mar-
ijuana use is not necessarily motivated by the desire to escape from
negative self-awareness, there are reasons to believe that some-
times individuals desire smoking marijuana to avoid dealing with
their problem.
In Study 5, we considered the desire to smoke marijuana broadly
as an indication of the desire for an altered state of consciousness.
We expected that priming failure (vs. anger) among a sample of
marijuana smokers would increase both the desire to escape the
self (suicide-thought accessibility) and the desire for an altered
state of consciousness (the behavioral intention to smoke mari-
juana). In other words, we expected priming failure to elicit more
escapist reactions than priming anger. Such findings could provide
evidence that failure to attain important standards affects not only
accessibility of suicide-related concepts but also relevant behav-
ioral intentions.
Methods
Participants. Participants in this study were 29 members of
an Internet forum related to marijuana use and growing (one
woman and 28 men), who volunteered to complete a questionnaire
in English. Their age ranged from 22 to 52 (M33.2 years, SD
7.3), and they were all regular marijuana smokers, of various
nationalities and professions. They were recruited by posts on the
forum, inviting them to complete an anonymous questionnaire.
Procedure and materials. After reading instructions for the
study, participants were primed either with the idea of anger or
with the idea of failure. Next, they completed a construct acces-
sibility measure, an explicit measure of affect, and a measure of
the intention to smoke marijuana.
Experimental condition. In the control condition, participants
were asked to recall the occasion in which they had experienced
the greatest anger in their lives. In this way, the control condition
pertained to a negative and stressful experience but one that was
unrelated to failure. In the failure condition, participants recalled
their greatest personal failure. In the two conditions, participants
were asked to describe the negative event in a few statements and
to indicate the repercussions that it has had on their lives.
Construct accessibility. Participants completed a word com-
pletion task similar to those used in previous studies. The word
completion task, containing 30 word fragments, was called Cre-
ative Task. Among the fragments, six could be completed as
Figure 2. Suicide-related completions as a function of disappointment with
capitalism and essay on positive consequences of transition (Study 4).
597
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
neutral or as death-related (target words were coffin,dead,killed,
grave,skull, and buried), and five could be completed as neutral or
suicide-related (target words were attempt,hang/hung,rope, sui-
cide, and wrist). The remaining 19 fragments could be completed
only as neutral. Death- and suicide-thought accessibility were
assessed as in previous studies.
Affect. To examine possible effects of our conditions on
affect, participants completed the 10-item form of the Positive and
Negative Affect Schedule (Kercher, 1992; for the original scale see
Watson, Clark, & Tellegen, 1988). They indicated how inspired,
scared, alert, excited, upset, nervous, enthusiastic, determined,
afraid, and distressed they felt on 7-point Likert scales (1
completely disagree, 7completely agree). We computed aver-
age positive (␣⫽.86) and negative emotions (␣⫽.97).
Intention to smoke marijuana. Participants reported the
amount of time that they thought would pass before they smoked
marijuana the next time (in minutes). Low scores on this measure
indicated high intentions to smoke marijuana.
Results
Five observations were detected as outliers on the measure of
intention to smoke marijuana (with zscores greater than 3 SD). To
avoid excluding too many observations, we recoded these obser-
vations to 15 min (i.e., the highest value with an acceptable Cook’s
distance). The results were very close when the five extreme
observations were excluded rather than recoded. To avoid redun-
dancy, we report hereinafter only findings observed with the
transformed scores.
A 2 (condition: anger vs. failure) 2 (type of thoughts: suicide
vs. death) ANOVA on construct accessibility, with repeated mea-
sures on the last variable, revealed only a significant effect of the
type of thoughts, F(1, 26) 4.88, p.04,
2
.16. Overall,
participants showed greater accessibility of suicide- than of death-
related thoughts. However, this effect was qualified by a marginal
interaction with the condition, F(1, 26) 4.01, p.06,
2
.13,
indicating that experimental condition did not have the same effect
on death- and suicide-thought accessibility. Death-related comple-
tions were not statistically different in the failure condition (M
5.39, SD 4.28) than in the anger condition (M5.75, SD
5.19), F(1, 26) 1, ns,
2
.00. However, as expected, suicide-
related completions were significantly more frequent in the failure
(M11.28, SD 7.41) than in the anger condition (M6.03,
SD 5.21), F(1, 26) 4.69, p.04,
2
.15.
A 2 (condition: anger vs. failure) 2 (affect: positive vs.
negative) ANOVA on affect, with repeated measures on the last
variable, revealed only a significant effect of the repeated variable,
F(1, 26) 17.73, p.001,
2
.47. Overall, participants
reported more positive than negative affect (M3.62, SD 1.14,
and M2.06, SD 1.26, respectively), but this effect was not
qualified by a significant interaction with the condition, F(1, 26)
1.15, ns,
2
.05.
An ANOVA on the intention to smoke marijuana revealed a
strong effect of condition, F(1, 27) 20.97, p.001,
2
.44.
In the anger condition, participants intended to smoke marijuana in
about 10 min (M10.73, SD 5.49). Those in the failure
condition expected to smoke in 2.50 min on average (SD 4.01).
Discussion
Results of this study were consistent with our expectations. In
line with the idea that individuals are particularly motivated to
escape the self when they fail to attain important standards,
suicide-related thoughts were more accessible and desire to smoke
marijuana was more urgent when participants were reminded of
their greatest personal failure rather than of their greatest anger.
Priming failure did not seem to induce more negative emotions
than priming anger. However, priming failure, rather than anger,
considerably increased intention to smoke marijuana. This is con-
sistent with the idea that consciousness is problematic after failure
and that individuals can desire smoking marijuana in order to
avoid (at least momentarily) dealing with their failure.
In Study 6, we set out to extend previous findings by testing the
hypothesis that failure to attain important standards increases not
only suicide-thought accessibility but also the accessibility of more
general concepts related to escape and relief.
Study 6: The Thin-Ideal Body or When
Standards Are Unattainable
Western societies impose very high and often unattainable stan-
dards of beauty for women. In Study 6, we tested whether failure
to attain the thin-ideal body standard (compared with larger mod-
els) might cause thoughts of suicide to become more accessible.
Research supports the sociocultural perspective that mass media
promulgates criteria for slenderness that elicit body dissatisfaction
in women (for a review, see Groesz, Levine, & Murnen, 2002).
Brief exposure to media images depicting the thin-ideal body
(compared with larger models) has been linked to women’s body
dissatisfaction, internalization of the thin ideal, and eating disor-
ders (for a recent meta-analysis of more than 70 studies, see Grabe,
Ward, & Hyde, 2008). This prompted an examination of the links
between exposure to the cultural standard of the thin-ideal body
and suicide-thought accessibility. In line with our theoretical rea-
soning, we expected women who did not feel particularly satisfied
with their body (i.e., who felt that they did not meet the cultural
standard of thinness) to show increased suicide-thought accessi-
bility after brief exposure to images of the thin-ideal body. We did
not expect this effect to appear (or expected it to be reduced)
among women who were relatively satisfied with their body.
In Study 6, we also explored whether increased accessibility of
suicide-related thoughts after failure to attain cultural standards is
associated with a simultaneous increase in the accessibility of
more general concepts related to escape. In Studies 2 and 3, we
found no effects of our manipulations on accessibility of nega-
tively valenced thoughts (sadness, depression, etc.), suggesting
that failure to attain standards does not affect negative/positive
thoughts in general. Our theoretical reasoning suggests, however,
that failure to attain important standards would increase not only
suicide-thought accessibility but also the accessibility of more
general concepts related to escape. Such a pattern of findings
would be consistent with the logic of goal system theory that
activation of a goal (escape) automatically triggers means to that
goal (suicide).
We did not assess death-thought accessibility in this study, in
part because it became clear from our previous studies that failure
to attain standards has only small effects on this kind of thought
598 CHATARD AND SELIMBEGOVIC
´
(this issue is taken up further in the General Discussion section),
and in part because we wanted to ensure that failure to attain
standards affects suicide-thought accessibility even when accessi-
bility of death-related thoughts is not simultaneously assessed.
To provide converging evidence for our reasoning using a different
method, we used a lexical decision task, rather than a word comple-
tion task, to assess accessibility of concepts related to suicide and
escape. We thus predicted that images of the thin-ideal body (vs.
images of larger models) would facilitate identification of suicide- and
escape-related words in a lexical decision task and that this effect
would be moderated by participants’ body dissatisfaction.
Methods
Participants. Participants were 43 female students (M
age
24.41 years, SD 7.40) from the University of Geneva who took
part in an experiment on personality and body image.
Procedure and materials. The experiment was implemented
on a computer using E-prime software. Participants first completed
measures of body mass index and body dissatisfaction. Next, they
were exposed to a thin-ideal body model or to a larger model. Finally,
they completed a lexical decision task to assess construct accessibility.
Body mass index and body dissatisfaction. Participants were
first asked to indicate their height and weight. This allowed us to
compute their body mass index. According to this objective indi-
cator (M21.53, SD 2.91), 83.3% of participants had normal
weight, 9.5% were underweight, 4.8% were overweight, and only
2.4% were obese. Research has indicated that women’s subjective
perceptions of their bodies are often better predictors of their
behavior than the body mass index. Thus, participants were also
asked to complete the Body Dissatisfaction Scale (Garner, Olm-
sted, & Polivy, 1983). They indicated their agreement with nine
propositions (e.g., “I think that my thighs are too large”; “I like the
shape of my buttocks”; “I think that my hips are too big”), using
7-point Likert scales (1 completely disagree,7completely
agree). High scores on this measure (␣⫽.84) indicate feelings of
dissatisfaction with one’s weight and one’s body. Body mass index
and body dissatisfaction scores were strongly positively related,
r(42) .70, p.001.
Experimental condition. Participants were randomly exposed
to one of six media images depicting a female top model and were
asked to form an impression of this person. All images represented
the same model (in different positions and wearing different sets of
underwear). However, the images were slightly modified in Pho-
toshop, such that the model looked underweight (in the thin-ideal
condition) or overweight (in the large model condition).
2
The only
difference between conditions was the underweight versus over-
weight appearance of the model.
Construct accessibility. Participants completed a lexical de-
cision task, which involved distinguishing between words and
nonwords presented on the computer screen. They were instructed
to press a key labeled word if they saw a word and a key labeled
nonword if they saw a nonword. The task was composed of 48
trials, presented in a computerized totally random order to each
participant. There were 32 nonwords and 16 words (eight neutral
words, four suicide-related words, and four escape-related words).
The suicide-related words were the same as those used in our
previous studies: suicide [suicide], pendre [to hang], corde [rope],
and veine [wrist]. The escape-related words were e´chapper [to
escape], calme [quiet], paix [peace], and sereine [serene]. The
neutral words were vent [wind], souvent [often], ballon [ball], livre
[book], bonsoir [good evening], train [train], haie [hedge], and
poche [pocket]. Participants were instructed to proceed as quickly
as possible, while trying to answer each trial correctly. Unbe-
knownst to participants, the computer recorded the speed with
which they responded to each of the 48 trials (in milliseconds). At
the end of the study, participants were probed for suspicion and
fully debriefed.
Results
In line with recommended procedures for this type of data, we
first conducted some minor transformations on our reaction time
data (Bargh & Chartrand, 2000). Incorrect responses (fewer than
1%) were excluded from the analysis, and responses greater than
2,000 ms (about 3%) were recoded to 2,000 ms. In addition, one
participant was excluded because she was detected as an outlier
(with zscores on reaction time data greater than 3 SD). After these
transformations, the assumptions for data analysis were acceptably
met. We thus computed mean reaction times for suicide, escape,
and neutral words for each participant.
To test our main hypothesis, we regressed reaction time for each
kind of words on condition (large model coded –1 and thin-ideal
model coded 1), body dissatisfaction scores (continuous variable,
mean centered), and the interaction term.
For reaction time for neutral words, the analysis revealed no
effect of condition, ␤⫽–.20, t(38) –1.25, ns, no effect of body
dissatisfaction, ␤⫽–.08, t(38) – 0.48, ns, and no interaction,
␤⫽–.02, t(38) – 0.12, ns.
The results were quite different on reaction time for suicide-
related words. There was a nearly significant effect of condition,
␤⫽–.26, t(38) –1.84, p.08. Reaction time for suicide-related
words were somewhat lower in the thin-ideal condition (M
761.96, SD 224.26) than in the large model condition (M
903.95, SD 295.88). There was no main effect of body dissat-
isfaction, ␤⫽–.14, t(38) – 0.97, ns. However, the predicted
interaction was significant, ␤⫽–.35, t(38) –2.42, p.02 (see
Figure 3). An analysis of simple slopes revealed that, as expected,
participants relatively high in body dissatisfaction (computed at 1
SD from the mean) showed lower reaction times for suicide-related
words in the thin-ideal condition than in the large model condition,
␤⫽–.60, t(38) –3.09, p.005. This difference was not
significant among participants relatively low in body dissatisfac-
tion (computed at –1 SD from the mean), ␤⫽.07, t(38) 0.35,
ns. Indeed, body dissatisfaction scores were not significantly related
to reaction times for suicide-related words in the large model condi-
tion, ␤⫽.21, t(38) 0.98, ns, but body dissatisfaction scores were
negatively related to reaction times for suicide-related words in the
thin-ideal condition, ␤⫽–.49, t(38) –2.50, p.02.
The interaction between body dissatisfaction scores and condi-
tion on reaction times for suicide-related words remained statisti-
cally significant, ␤⫽–.33, t(38) –3.20, p.005, after con-
trolling for reaction times for neutral words. In the same way, there
was a significant interaction between body dissatisfaction scores
and condition on difference scores between reaction times for
2
Stimulus materials are available on request from Armand Chatard.
599
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
suicide-related and neutral words, ␤⫽–.42, t(38) –2.88, p
.01, indicating that our variables had a greater impact on suicide-
than on neutral-related words (Figure 3).
For reaction times for escape-related words, there were no main
effects of condition, ␤⫽–.17, t(38) –1.16, ns, or of body
dissatisfaction, ␤⫽.03, t(38) 0.19, ns. However, there was a
nearly significant interaction, ␤⫽–.29, t(38) –1.92, p.07
(Figure 3). This interaction indicated that participants relatively
high in body dissatisfaction showed lower reaction times for
escape-related words in the thin-ideal condition than in the large
model condition, ␤⫽–.46, t(38) –2.23, p.04. This difference
was not significant among participants relatively low in body
dissatisfaction, ␤⫽.10, t(38) 0.48, ns. Body dissatisfaction
scores were not significantly related to reaction times for escape-
related words in the large model condition, ␤⫽.32, t(38) 1.44,
ns, or in the thin-ideal condition, ␤⫽–.26, t(38) –1.27, ns.
Thus, the interaction on escape-related words was mainly triggered
by reactions of participants with relatively high scores on body
dissatisfaction. This interaction remained statistically significant,
␤⫽–.28, t(38) –2.46, p.02, after controlling for reaction
times for neutral words.
Discussion
Study 6 results were largely consistent with our expectations. A
brief exposure to the cultural standard of the thin-ideal body,
compared with exposure to a larger model, increased accessibility
of concepts related to suicide among women relatively dissatisfied
with their body (i.e., who felt they did not meet the cultural
standard). This effect was not found among women relatively
satisfied with their bodies. Analyses of escape-related completions
suggested that increased accessibility of suicide-related concepts
after exposure to the thin-ideal body model was concomitant to
increased accessibility of more general concepts related to escape
and relief. Such a finding is in line with our reasoning that failure to
attain important standards increases motivations to escape the self.
General Discussion
Six studies provided support for the central hypothesis of this
article that suicide-related thoughts arise as a result of a motivation
to escape from negative self-awareness, when individuals realize
that they fail to attain an important standard. This hypothesis,
consistent with escape theory, was examined on different stan-
dards, among different populations, and with different methods of
measuring construct accessibility. The results suggest that when
individuals realize that they fail to attain an important standard,
they are often motivated to escape the self. Thereby, they experi-
ence an immediate increase in the accessibility of suicide-related
thoughts. Of course, suicide-related thoughts can increase in ac-
cessibility without reaching consciousness. Even when they do,
they do not necessarily reflect explicit suicidal thoughts (see our
introductory quotation). However, our findings are important be-
cause an increase in suicide-thought accessibility is arguably a
necessary condition for explicit suicidal thoughts and behavior to
occur.
We believe that the present findings offer new insights into
when, why, and how suicide-related thoughts come to mind. Our
results also have important theoretical implications for escape
theory and TMT. In spite of their limitations, that should be kept
in mind when interpreting the results, we think that our findings
offer new avenues of research for the study of factors that affect
accessibility of suicide-related thoughts.
When, Why, and How Suicide-Related Thoughts
Come to Mind
Our results indicate that suicide-thought accessibility tends to
increase whenever individuals realize that they fail to attain an
important standard. Different factors moderate this effect, but we
consistently found that negative comparisons between self and
standards were sufficient to lead some participants to experience
an immediate increase in suicide-thought accessibility. The mere
Figure 3. Reaction time for neutral, suicide-related, and escape-related words as a function of body dissatis-
faction and the model to which participants were exposed (Study 6). RT reaction time.
600 CHATARD AND SELIMBEGOVIC
´
comparison with a standard did not increase suicide-thought ac-
cessibility among participants who met the standard. In the same
way, participants who were discrepant from the standard (rela-
tively unhappy Swiss participants in Study 3) did not show in-
creased accessibility of suicide-related thoughts unless they were
confronted with the standard. Thus, it seems that it is the conjunc-
tion of the two factors that is crucial in generating increases in
suicide-thought accessibility. When individuals mentally compare
the self with important standards and become aware of a discrep-
ancy, they experience a state of negative self-awareness (Duval &
Wicklund, 1972), which seems to be a minimal but sufficient
condition to generate increases in suicide-thought accessibility.
Our results suggest that this might occur because individuals fall
severely short of standards but also because standards are too high
(Studies 2 and 6). This might even occur after a positive prime
because it is the discrepancy between self and standards that is
crucial in precipitating suicide-related thoughts, rather than the
negativity of the prime in itself (Studies 3 and 4). Our results also
suggest that suicide-thought accessibility increases after failure
because in such situations individuals are motivated to escape from
negative self-awareness. In line with this proposal, individuals
high in self-consciousness and escapist motivations appeared es-
pecially vulnerable to such effects (Study 1). Priming failure
increased suicide-thought accessibility only (or mainly) when it
was associated with a large discrepancy between self and standards
(Studies 2– 4 and 6). Finally, increased accessibility of suicide-
related thoughts after failure was associated with increased acces-
sibility of more general concepts related to escape and relief
(Study 6).
Finally, our results suggest that suicide-thought accessibility
increases immediately and quite spontaneously after negative com-
parisons between self and standards. In line with this view, failure
to attain standards led to immediate increases in the proportion of
word fragments completed as suicide-related rather than as neutral
in a word completion task (Studies 1–5) and reduced reaction
times for suicide-related words in a lexical decision task (Study 6).
The fact that we found converging evidence for our hypotheses
when using reaction time data is consistent with our argument that
increased accessibility of suicide-related thoughts after failure is a
normal, immediate, and quite spontaneous reaction.
In spite of the recognition that self-defeating behaviors are
common even in normal individuals (Baumeister & Scher, 1988),
to the best of our knowledge, the current research represents the
first attempt to examine suicide-thought accessibility in a theoret-
ically integrated manner relying on experimental designs. Consid-
erable research has explored how people respond to psychological
threats, especially self-esteem threats (e.g., Heine et al., 2006).
Nonetheless, most of this work has focused on what might be
considered fight responses, such as efforts to restore a sense of
self-worth or meaning. Very little research has examined how
people attempt to escape from such aversive situations (flight
responses). From research on self-awareness, we know that the
desire to escape the self after failure can have some behavioral
implications, such as behavioral avoidance (Silvia & Duval, 2001).
The present research complements this line of work by document-
ing some important cognitive consequences of the desire to escape
the self after failure.
Escape Theory Complemented and Extended
In the present studies, we tested one implication of escape
theory (Baumeister, 1990), the idea of a direct causal link between
failure to attain standards and suicide-thought accessibility. Escape
theory implied such a link but never tested it. The present research
shows that this causal relation is well supported by experimental
evidence. The concept of suicide appears to be strongly connected
with the concept of failure, such that the idea of suicide often
comes to mind immediately after failure. The main contribution of
the present research is to specify when (implicit) thoughts of
suicide emerge in the suicidal cycle proposed by escape theory. If
explicit thoughts emerge at the end of the suicidal cycle, the very
idea of suicide tends to increase in accessibility immediately after
failure.
In the present studies, we also tested more specific predictions
derived from escape theory. Of interest, the same conditions that
lead to suicidal behavior, according to escape theory, also seem to
lead to increases in suicide-thought accessibility, according to the
present findings. For example, Baumeister (1990) argued that high
standards can have a dark side, because of the high expectations
that they create and the ensuing substantial stress if those expec-
tations are not met. In line with this contention, our studies suggest
that the pursuit of high standards at the national level can have a
dark side for those who do not attain them, by giving rise to
suicide-related thoughts (Studies 2 and 3). Escape theory also
predicts that the desire to escape the self is particularly pronounced
in periods of transition and incertitude (Baumeister, 1991b). In-
deed, transitions often bring high expectations for a better life. If
and when these expectations turn into disappointment, individuals
can feel a desire to escape and thus experience an immediate
increase in the accessibility of suicide-related thoughts. Study 4
results were consistent with this reasoning. They suggest that the
gap between high expectations and disappointment by capitalism
in former communist countries can play a role in the very high
suicide rates observed in these countries.
In summary, our findings largely confirm the predictive validity
of escape theory and suggest that the same conditions that lead to
increased accessibility of suicide-related thoughts can also ulti-
mately lead to suicidal behavior.
Reconciling TMT With Suicide
Our findings are at odds with predictions derived from TMT. On
the one hand, we found little evidence that failure to attain stan-
dards affected death-thought accessibility. On the other hand, we
consistently found that failure to attain standards increased
suicide-thought accessibility, a finding not particularly compatible
with TMT.
There does not seem to be a problem with our measure of
death-thought accessibility, because it was affected by other ma-
nipulations (a death-related prime in Study 1) in a theoretically
predictable way. Moreover, the same measure has been success-
fully used in many other studies to assess death-thought accessi-
bility. Rather, we believe there are three main reasons for the lack
of significant effects of failure on death-thought accessibility.
First, our studies were not designed to test TMT predictions. To do
this, it would have certainly been preferable to preselect only
participants for whom the standard was an important source of
601
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
self-esteem (see Hayes et al., 2008). Second, it is quite possible
that some of our participants responded defensively to our manip-
ulations, such that they suppressed death-related thoughts. After
all, participants often respond defensively to self-esteem threats
(e.g., Bongers, Dijksterhuis, & Spears, 2009). Finally, it is possible
that death-thought accessibility would have increased later. It is
possible, for instance, that suicide-related thoughts are rejected
when death-related thoughts become accessible (the fear of death
could serve as a safeguard mechanism that prevents thoughts of
suicide from influencing action). Because in this research we were
interested only in reactions occurring in the immediate aftermath
of the confrontation with failure, such an increase may have passed
undetected. In summary, there are several possible reasons for the
lack of significant effects on death-thought accessibility, and we
do not see this as really problematic for TMT.
More problematic, however, is that failure to attain standards
increased suicide-thought accessibility. TMT would not have pre-
dicted such effects. Clearly, suicide is not compatible with the
notion of fear of death, central in TMT (Muraven & Baumeister,
1997). As Heine et al. (2006) noticed, “If people sometimes choose
death over meaninglessness, it scarcely seems possible that death
can be considered their greatest fear” (p. 105). TMT is a very
important and promising theoretical framework that has proved
very useful in predicting a variety of human behaviors. Indeed, few
behaviors are inconsistent with TMT, except perhaps suicide. If
our findings are not compatible with TMT in its actual formula-
tion, we believe that it is possible to reconcile TMT with suicide
and accessorily with our findings. To this end, we propose to
introduce the notion of the fear of life in this theoretical frame-
work.
Indeed, the existentialists have long argued that the awareness of
death entails not one but two great fears: Todesangst [the fear of
death] and Lebensangst [the fear of life] (e.g., Heidegger, 1962;
Rank, 1952). The latter is the fear of not being who one wants to
be, not meeting one’s standards, not fulfilling one’s aspirations in
this world (see Becker, 1973). Indeed, it may be misleading to
reduce all human fears to the sole fear of death (Lerner, 1997). The
fact that certain persons deliberately choose to kill themselves
suggests that the fear of life might sometimes surpass the fear of
death. We propose that the fear of life and the fear of death coexist
but that they dominate under different conditions. The fear of death
would dominate when mortality is salient, whereas the fear of life
would dominate when failure is salient. The fear of death leads to
the typical reactions documented in the TMT literature, whereas
the fear of life promotes escapist thoughts and behaviors. Our
findings are consistent with this reasoning.
In summary, even if our findings are not consistent with TMT in
its actual formulation, they suggest one interesting possibility for
reconciling TMT with suicide.
Limitations
One potential limitation of the present studies is its exclusive
emphasis on failure as a determinant of suicide-thought accessi-
bility. In line with escape theory, many negative events (getting
fired, getting a divorce, etc.) might be interpreted as a failure.
However, our reasoning about the fear of life suggests that other
factors than failure can provoke increases in suicide-thought ac-
cessibility. Indeed, the same factors that protect from the fear of
death (self-esteem, cultural worldviews, and close relationships)
may also protect from the fear of life. If self-esteem, close rela-
tionships, and cultural values protect from the fear of life, then a
disruption in any of these three psychological structures may lead
to increases in suicide-thought accessibility, and future research
should address this question.
Our aim in the present research was not to examine all factors
that could lead to increases in suicide-thought accessibility but to
examine whether one factor that is supposed to play a major role
in suicidal behavior (failure) affects the accessibility of suicide-
related thoughts. We excluded several alternative explanations, but
there are probably others that might be relevant. For example,
failure to conform to cultural standards can induce fear of social
exclusion. Social exclusion has been shown to foster a decon-
structed state of mind characteristic of individuals likely to attempt
suicide (Baumeister, 1990; Twenge et al., 2003; see also Joiner,
2005). If we agree that some of our results can be interpreted in
these terms (Studies 3 and 4), we do not believe that social
exclusion provides an adequate and sufficient explanation. In
particular, Study 6 seems to provide results inconsistent with such
an interpretation. Cultural standards are often very high and some-
times even unattainable (the thin-ideal body). Most people do not
reach this standard. In this context, it seems relatively unlikely that
women would fear social exclusion. However, we acknowledge
that failure and social exclusion are intimately related, and future
research may seek to disentangle their respective contributions to
the present results.
Behavioral Implications
One final question that arises in relation to our findings concerns
the potential behavioral implications of increased suicide-thought
accessibility and especially its link to suicidal behavior. When the
idea of suicide reaches conscious awareness, in most cases it is
likely to be rejected, and individuals might not even notice it as a
suicidal thought. However, we believe that the likelihood of a
suicide attempt increases to the extent that individuals experience
repeated increases in suicide-thought accessibility (i.e., when fail-
ure automatically activates the idea of suicide). Such repeated
increases in suicide-thought accessibility can lead individuals to
believe that suicide is the only available means of escape. Yet,
other factors that we believe could facilitate acting out on suicide-
related thoughts are self-blame (Baumeister, 1990), the importance
and permanence of the discrepancy, and the perception of one’s
inability to reduce or eliminate the discrepancy (e.g., Silvia &
Duval, 2001). In addition, the link between thought accessibility
and behavior can be moderated by a censorship of potentially
costly behaviors (Macrae & Johnston, 1998) or attitudes toward
the behavior in question (Cesario, Plaks, & Higgins, 2006). Recent
research also suggests that the interpretation of accessible con-
structs depends on the situational context and may or may not lead
to corresponding action (Jefferis & Fazio, 2008).
Building on goals systems theory (Kruglanski et al., 2002), it
could be hypothesized that availability of and reliance on other
means of escape could postpone a suicide attempt. In relation to
this, one of the potential behavioral consequences of suicide-
thought accessibility is engagement in escape-related behaviors
other than suicide, such as watching television, spirituality, work-
ing, and alcohol or drug consumption, among others. Results of
602 CHATARD AND SELIMBEGOVIC
´
our Study 5 are consistent with this idea, because suicide-thought
accessibility was related to a pressing wish to alter the state of
consciousness by smoking marijuana.
Finally, protective and precipitating factors identified in relevant
suicide research are to be taken into consideration. In particular,
social support, marriage, and parenthood should decrease the prob-
ability of a suicide attempt, whereas impulsivity and the availabil-
ity of means and occasions for suicide could precipitate it.
Concluding Thoughts
The accessibility of suicide-related thoughts is influenced and
determined by a sense of failure to attain standards. The set of
studies presented here, taken together, underscores a phenomenon
that has rarely been examined before and provides a very consis-
tent pattern of results that cannot be easily explained by other
theoretical perspectives than the one adopted here. They have
important implications for escape theory and TMT. The present
research is a first but an important step toward understanding
circumstances that cause increases in suicide-thought accessibility.
References
Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness
depression: A theory-based subtype of depression. Psychological Re-
view, 96, 358 –372. doi:10.1037/0033-295X.96.2.358
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Newbury Park, CA: Sage.
Arndt, J., Greenberg, J., Simon, L., Pyszczynski, T., & Solomon, S. (1998).
Terror management and self-awareness: Evidence that mortality salience
provokes avoidance of the self-focused state. Personality and Social
Psychology Bulletin, 24, 1216 –1227. doi:10.1177/01461672982411008
Baechler, J. (1975). Les suicides [Suicides]. Paris, France: Calmann-Le´vy.
Bargh, J. A., & Chartrand, T. L. (2000). The mind in the middle: A
practical guide to priming and automaticity research. In H. T. Reis &
C. M. Judd (Eds.), Handbook of research methods in social and per-
sonality psychology (pp. 253–285). New York, NY: Cambridge Univer-
sity Press.
Bargh, J. A., & Pratto, F. (1986). Individual construct accessibility and
perceptual selection. Journal of Experimental Social Psychology, 22,
293–311. doi:10.1016/0022-1031(86)90016-8
Baudelot, C., & Establet, R. (2006). Suicide: L’envers de notre monde
[Suicide: The reverse side of our world]. Paris, France: Seuil.
Baumeister, R. F. (1988). Masochism as escape from self. Journal of Sex
Research, 25, 28 –59. doi:10.1080/00224498809551444
Baumeister, R. F. (1990). Suicide as escape from self. Psychological
Review, 97, 90 –113. doi:10.1037/0033-295X.97.1.90
Baumeister, R. F. (1991a). Escaping the self: Alcoholism, spirituality,
masochism, and other flights from the burden of selfhood. New York,
NY: Basic Books.
Baumeister, R. F. (1991b). Meanings of life. New York, NY: Guilford
Press.
Baumeister, R. F., & Scher, S. J. (1988). Self-defeating behavior patterns
among normal individuals: Review and analysis of common self-
destructive tendencies. Psychological Bulletin, 104, 3–22. doi:10.1037/
0033-2909.104.1.3
Becker, E. (1973). The denial of death. New York, NY: Free Press.
Becker, E. S., Strohbach, D., & Rinck, M. (1999). A specific attentional
bias in suicide attempters. Journal of Nervous and Mental Disease, 187,
730 –735. doi:10.1097/00005053-199912000-00004
Bongers, K. C. A., Dijksterhuis, A., & Spears, R. (2009). Self-esteem
regulation after automatic goal-pursuit success and failure. Journal of
Experimental Social Psychology, 45, 468 – 477. doi:10.1016/j.j-
esp.2008.12.007
Burger, J. M., & Cooper, H. M. (1979). The desirability of control.
Motivation and Emotion, 3, 381–393. doi:10.1007/BF00994052
Carver, C. S., Scheier, M., & Weintraub, J. (1989). Assessing coping
strategies: A theoretically based approach. Journal of Personality and
Social Psychology, 56, 267–283. doi:10.1037/0022-3514.56.2.267
Central Intelligence Agency. (2009). The 2008 world factbook. Retrieved
from https://www.cia.gov/library/publications/the-world-factbook/
index.html
Cesario, J., Plaks. J. E., & Higgins, E. T. (2006). Automatic social behavior
as motivated preparation to interact. Journal of Personality and Social
Psychology, 90, 893–910. doi:10.1037/0022-3514.90.6.893
Cornette, M. M., Abramson, L. Y., & Bardone, A. M. (2000). Toward an
integrated theory of suicidal behaviors: Merging the hopelessness, self-
discrepancy, and escape theories. In T. E. Joiner & M. D. Rudd (Eds.),
Suicide science: Expanding boundaries (pp. 43– 66). Boston, MA: Klu-
wer Academic.
De Houwer, J., Teige-Mocigemba, S., Spruyt, A., & Moors, A. (2009).
Implicit measures: A normative analysis and review. Psychological
Bulletin, 135, 347–368. doi:10.1037/a0014211
Dodgson, P. G., & Wood, J. V. (1998). Self-esteem and the cognitive
accessibility of strengths and weaknesses after failure. Journal of Per-
sonality and Social Psychology, 75, 178 –197. doi:10.1037/0022-
3514.75.1.178
Durkheim, E. (1897). Le suicide [Suicide]. Paris, France: Alcan.
Duval, S., & Wicklund, R. A. (1972). A theory of objective self awareness.
Oxford, England: Academic Press.
Fritsche, I., Jonas, E., & Fankha¨nel, T. (2008). The role of control moti-
vation in mortality salience effects on ingroup support and defense.
Journal of Personality and Social Psychology, 95, 524 –541. doi:
10.1037/a0012666
Garner, D. M., Olmsted, M. P., & Polivy, J. (1983). Development and
validation of a multidimensional Eating Disorder Inventory for anorexia
nervosa and bulimia. International Journal of Eating Disorders, 2, 15–34.
doi:10.1002/1098-108X(198321)2:215::AID-EAT22600202033.0
.CO;2-6
Gibbons, F. X., & Wicklund, R. A. (1982). Self-focused attention and
helping behavior. Journal of Personality and Social Psychology, 43,
462– 474. doi:10.1037/0022-3514.43.3.462
Grabe, S., Ward, L. M., & Hyde, J. S. (2008). The role of the media in body
image concerns among women: A meta-analysis of experimental and
correlational studies. Psychological Bulletin, 134, 460 – 476. doi:
10.1037/0033-2909.134.3.460
Greenberg, J., Arndt, J., Schimel, J., Pyszczynski, T., & Solomon, S.
(2001). Clarifying the function of mortality-salience induced worldview
defense: Renewed suppression or reduced accessibility of death-related
thoughts? Journal of Experimental Social Psychology, 37, 70 –76. doi:
10.1006/jesp.2000.1434
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition:
Attitudes, self-esteem, and stereotypes. Journal of Personality and So-
cial Psychology, 102, 4 –27.
Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. (1998). Measuring
individual differences in implicit cognition: The implicit association test.
Journal of Personality and Social Psychology, 74, 1464 –1480. doi:
10.1037/0022-3514.74.6.1464
Groesz, L. M., Levine, M. P., & Murnen, S. K. (2002). The effect of
experimental presentation of thin media images on body satisfaction: A
meta-analytic review. International Journal of Eating Disorders, 31,
1–16. doi:10.1002/eat.10005
Halbwachs, M. (1930). Les causes du suicide [The causes of suicide].
Paris, France: Alcan.
Hart, J., Shaver, P. R., & Goldenberg, J. L. (2005). Attachment, self-
esteem, worldview, and terror management: Evidence for a tripartite
603
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
security system. Journal of Personality and Social Psychology, 88,
999 –1013. doi:10.1037/0022-3514.88.6.999
Hayes, J., Schimel, J., Arndt, J., & Faucher, E. H. (2010). A theoretical and
empirical review of the death-thought accessibility concept in terror
management research. Psychological Bulletin, 136, 699 –739. doi:
10.1037/a0020524
Hayes, J., Schimel, J., Faucher, E. H., & Williams, T. (2008). Evidence for
the DTA hypothesis II: Threatening self-esteem increases death-thought
accessibility. Journal of Experimental Social Psychology, 44, 600 – 613.
doi:10.1016/j.jesp.2008.01.004
Heatherton, T. F., & Baumeister, R. F. (1991). Binge eating as escape from
self-awareness. Psychological Bulletin, 110, 86 –108. doi:10.1037/0033-
2909.110.1.86
Heidegger, M. (1962). Being and time. New York, NY: Harper & Row.
Heine, S. J., Proulx, T., & Vohs, K. D. (2006). The meaning maintenance
model: On the coherence of human motivations. Personality and Social
Psychology Review, 10, 88 –110. doi:10.1207/s15327957pspr1002_1
Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect.
Psychological Review, 94, 319 –340. doi:10.1037/0033-295X.94.3.319
Higgins, E. T., Bond, R., Klein, R., & Strauman, T. J. (1986). Self-
discrepancies and emotional vulnerability: How magnitude, type and
accessibility of discrepancy influence affect. Journal of Personality and
Social Psychology, 51, 5–15. doi:10.1037/0022-3514.51.1.5
Higgins, E. T., King, G. A., & Mavin, G. H. (1982). Individual construct
accessibility and subjective impressions and recall. Journal of Person-
ality and Social Psychology, 43, 35– 47. doi:10.1037/0022-3514.43.1.35
Hull, J. G. (1981). A self-awareness model of the causes and effects of
alcohol consumption. Journal of Abnormal Psychology, 90, 586 – 600.
doi:10.1037/0021-843X.90.6.586
Jefferis, V. E., & Fazio, R. H. (2008). Accessibility as input: The use of
construct accessibility as information to guide behavior. Journal of
Experimental Social Psychology, 44, 1144 –1150. doi:10.1016/
j.jesp.2008.02.002
Joiner, T. (2005). Why people die by suicide. Cambridge, MA: Harvard
University Press.
Kercher, K. (1992). Assessing subjective well-being in the old-old. The
PANAS as a measure of orthogonal dimensions of positive and
negative affect. Research on Aging, 14, 131–168. doi:10.1177/
0164027592142001
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y.,
& Sleeth-Keppler, D. (2002). A theory of goal systems. In M. P. Zanna
(Ed.), Advances in experimental social psychology (Vol. 34, pp. 331–
378). San Diego, CA: Academic Press.
Lerner, M. J. (1997). What does the belief in a just world protect us from:
The dread of death or the fear of undeserved suffering? Psychological
Inquiry, 8, 29 –32. doi:10.1207/s15327965pli0801_5
Lyubomirsky, S., & Lepper, H. (1999). A measure of subjective happiness:
Preliminary reliability and construct validation. Social Indicators Re-
search, 46, 137–155. doi:10.1023/A:1006824100041
Macrae, C. N., & Johnston, L. (1998). Help, I need somebody: Automatic
action and inaction. Social Cognition, 16, 400 – 417.
Maxfield, M., Pyszczynski, T., Kluck, B., Cox, C. R., Greenberg, J.,
Solomon, S., & Weise, D. (2007). Age-related differences in responses
to thoughts of one’s own death: Mortality salience and judgments of
moral transgressions. Psychology and Aging, 22, 341–353. doi:10.1037/
0882-7974.22.2.341
McClelland, G. H. (2000). Nasty data: Unruly, ill-mannered observations
can ruin your analysis. In H. T. Reis & C. M. Judd (Eds.), Handbook of
research methods in social and personality psychology (pp. 393– 411).
Cambridge, England: Cambridge University Press.
Mikulincer, M., Florian, V., Birnbaum, G., & Malishkevich, S. (2002). The
death-anxiety buffering function of close relationships: Exploring the
effects of separation reminders on death-thought accessibility. Person-
ality and Social Psychology Bulletin, 28, 287–299. doi:10.1177/
0146167202286001
Morin, A. (2002, December). Self-awareness review Part 1: Do you “self-
reflect” or “self-ruminate”? Science & Consciousness Review, 1. Re-
trieved from http://www2.mtroyal.ab.ca/%7Eamorin/Rumination.pdf
Morin, A. (2003, January). Self-awareness review Part 2: Changing or
escaping the self. Science & Consciousness Review, 1. Retrieved from
http://www2.mtroyal.ab.ca/%7Eamorin/Self-eval.pdf
Muraven, M., & Baumeister, R. F. (1997). Suicide, sex, terror, paralysis,
and other pitfalls of reductionist self-preservation theory. Psychological
Inquiry, 8, 36 – 40. doi:10.1207/s15327965pli0801_7
Nock, M. K., & Banaji, M. R. (2007). Prediction of suicide ideation and
attempts among adolescents using a brief performance-based test. Jour-
nal of Consulting and Clinical Psychology, 75, 707–715. doi:10.1037/
0022-006X.75.5.707
Nock, M. K., Borges, G., Bromet, E. J., Alonso, J., Angermeyer, M.,
Beautrais, A., . . . Williams, D. (2008). Cross-national prevalence and
risk factors for suicidal ideation, plans and attempts. British Journal of
Psychiatry, 192, 98 –105. doi:10.1192/bjp.bp.107.040113
O’Connor, R. C. (2003). Suicidal behaviour as a cry of pain: Test of a
psychological model. Archives of Suicide Research, 7, 297–308. doi:
10.1080/713848941
O’Connor, R. C., & Sheehy, N. P. (2000). Understanding suicidal behav-
ior. Leicester, England: BPS Blackwell.
Pehe, J. (1994). The Czech Republic: A successful transition. Radio Free
Europe/Radio Liberty, 3, 70 –75.
Pelletier, L., & Vallerand, R. J. (1990). L’Echelle Re´vise´e de Conscience
de Soi: Une traduction et une validation canadienne-franc¸aise du Re-
vised Self-Consciousness Scale [Revised Self-Consciousness Scale:
Canadian–French translation and validation]. Revue Canadienne des
Sciences du Comportement, 22, 191–206. doi:10.1037/h0078983
Pyszczynski, T., Greenberg, J., & Solomon, S. (1999). A dual-process
model of defense against conscious and unconscious death-related
thoughts: An extension of terror management theory. Psychological
Review, 106, 835– 845. doi:10.1037/0033-295X.106.4.835
Pyszczynski, T., Greenberg, J., Solomon, S., Arndt, J., & Schimel, J.
(2004). Why do people need self-esteem? A theoretical and empirical
review. Psychological Bulletin, 130, 435– 468. doi:10.1037/0033-
2909.130.3.435
Rank, O. (1952). The trauma of birth. New York, NY: Brunner. (Original
work published 1924)
Scheier, M. F., & Carver, C. S. (1985). The Self-Consciousness Scale: A
revised version for use with general populations. Journal of Applied Social
Psychology, 15, 687– 699. doi:10.1111/j.1559-1816.1985.tb02268.x
Schimel, J., Hayes, J., Williams, T., & Jahrig, J. (2007). Is death really the
worm at the core? Converging evidence that worldview threat increases
death-thought accessibility. Journal of Personality and Social Psychol-
ogy, 92, 789 – 803. doi:10.1037/0022-3514.92.5.789
Shah, J. Y., & Kruglanski, A. W. (2002). Priming against your will: How
goal pursuit is affected by accessible alternatives. Journal of Experi-
mental Social Psychology, 38, 368 –383. doi:10.1016/S0022-
1031(02)00005-7
Shneidman, E. S. (1996). The suicidal mind. New York, NY: Oxford
University Press.
Silvia, P. J., & Duval, T. S. (2001). Objective self-awareness theory:
Recent progress and enduring problems. Personality and Social Psy-
chology Review, 5, 230 –241. doi:10.1207/S15327957PSPR0503_4
Solomon, S., Greenberg, J., & Pyszczynski, T. (1991). Terror management
theory of self-esteem. In C. R. Snyder & D. R. Forsyth (Eds.), Handbook
of social and clinical psychology: The health perspective (pp. 21– 40).
Elmsford, NY: Pergamon Press.
Steenbarger, B. N., & Aderman, D. (1979). Objective self-awareness as a
nonaversive state: Effect of anticipating discrepancy reduction. Journal
of Personality, 47, 330 –339. doi:10.1111/j.1467-6494.1979.tb00206.x
604 CHATARD AND SELIMBEGOVIC
´
Strauman, T. J., & Higgins, E. T. (1987). Automatic activation of self-
discrepancies and emotional syndromes: When cognitive structures in-
fluence affect. Journal of Personality and Social Psychology, 53, 1004 –
1014. doi:10.1037/0022-3514.53.6.1004
Twenge, J. M., Catanese, K. R., & Baumeister, R. F. (2002). Social
exclusion causes self-defeating behavior. Journal of Personality and
Social Psychology, 83, 606 – 615. doi:10.1037/0022-3514.83.3.606
Twenge, J. M., Catanese, K. R., & Baumeister, R. F. (2003). Social
exclusion and the deconstructed state: Time perception, meaningless-
ness, lethargy, lack of emotion, and self-awareness. Journal of Person-
ality and Social Psychology, 85, 409 – 423. doi:10.1037/0022-
3514.85.3.409
van Praag, H. M. (2002). Why has the antidepressant era not shown a
significant drop in suicide rates? Crisis, 23, 77– 82. doi:10.1027//0227-
5910.23.2.77
Va¨rnik, A., Ko˜lves, K., Allik, J., Arensman, E., Aromaa, E., van Auden-
hove, C., . . . Hegerl, U. (2009). Gender issues in suicide rates, trends
and methods among youths aged 15–24 in 15 European countries.
Journal of Affective Disorders, 113, 216 –226. doi:10.1016/
j.jad.2008.06.004
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and vali-
dation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality and Social Psychology, 54, 1063–1070.
doi:10.1037/0022-3514.54.6.1063
Wenzel, A., & Beck, A. T. (2008). A cognitive model of suicidal behavior:
Theory and treatment. Applied and Preventive Psychology, 12, 189 –201.
doi:10.1016/j.appsy.2008.05.001
Wenzel, A., Brown, G. K., & Beck, A. T. (2009). Cognitive therapy for
suicidal patients: Scientific and clinical applications. Washington, DC:
American Psychological Association. doi:10.1037/11862-000
Williams, J. M. G. (1997). The cry of pain. London, England: Penguin.
Williams, J. M. G. (2001). Suicide and attempted suicide. Understanding
the cry of pain. London, England: Penguin.
Williams, J. M. G., & Broadbent, K. (1986). Distraction by emotional
stimuli: Use of a Stroop task with suicide attempters. British Journal of
Clinical Psychology, 25, 101–110.
Williams, J. M. G., & Pollock, L. R. (2001). Psychological aspects of the
suicidal process. In K. van Heeringen (Ed.), Understanding suicidal
behaviour (pp. 76 –93). Chichester, England: Wiley.
World Health Organization. (2009). Suicide prevention (SUPRE). Re-
trieved from http://www.who.int/mental_health/prevention/suicide/
suicideprevent/en/index.html
Received March 29, 2010
Revision received November 2, 2010
Accepted November 10, 2010
Instructions to Authors
For Instructions to Authors, please consult the February 2011 issue of the volume or visit www.apa.org/
pubs/journals/psp and click on the “Instructions to authors” tab in the Journal Info box.
605
STANDARDS, FAILURE, AND SUICIDE-THOUGHT ACCESSIBILITY
... Scores on the obsession and compulsion subscales range from 0 to 20, but only the total Y-BOCS score is interpreted. Total scores can be split into five categories, based on severity of symptoms: subclinical (0 to 7), mild (8)(9)(10)(11)(12)(13)(14)(15), moderate (16)(17)(18)(19)(20)(21)(22)(23), severe (24)(25)(26)(27)(28)(29)(30)(31), and extreme case of OCD (32)(33)(34)(35)(36)(37)(38)(39)(40). The present sample included only patients with moderate (n = 6, 6.3%), severe (n = 65, 68.4%), or extreme (n = 24, 25.2%) scores. ...
... Similar findings were found on multiple SA, even though the effects were less pronounced and significant. The present data are in line with previous research and theorizing, thereby suggesting that the primary motivation for suicide is intense psychological distress that eventually becomes intolerable and which cannot be escaped from by other means (e.g., treatment) than suicide (29,30). Our data are also consistent with previous research that indicates that the 5-HTTLPR polymorphism in the serotonin transporter gene may contribute to the risk of SB under stress (15,17,31). ...
Article
Full-text available
Background The severity of symptoms represents an important source of distress in patients with a psychiatric disease. However, the extent to which this endogenous stress factor interacts with genetic vulnerability factors for predicting suicide risks remains unclear. Methods We evaluated whether the severity of symptoms interacts with a genetic vulnerability factor (the serotonin transporter gene-linked promoter region variation) in predicting the frequency of lifetime suicide attempts in patients with a psychiatric disease. Symptom severity and 5-HTTLPR polymorphism were collected from a sample of 95 patients with obsessive-compulsive disorder (OCD). Lifetime suicide attempt was the primary outcome, and antecedent of multiple suicide attempts was the secondary outcome. Results The gene-by-symptoms interaction was associated with an excess risk of suicide attempts (OR = 4.39, 95CI[1.44, 13.38], p < 0.009) and of multiple suicide attempts (OR = 4.18, 95CI[1.04, 16.77], p = 0.043). Symptom severity (moderate, severe, or extreme) was associated with an approximately five-fold increase in the odds of a lifetime suicide attempt in patients carrying one or two copies of the short allele of 5-HTTLPR. No such relationship was found for patients carrying the long allele. Conclusion This study provides preliminary evidence for the gene-by-stress interaction on suicide attempt when stress is operationalized as symptom severity. Progress in suicide research may come from efforts to investigate the gene-by-symptoms interaction hypothesis in a variety of diseases.
... Proposed by Baumeister (1990) and expanded by Chatard and Selimbegović (2011), the escape theory of suicide posits that there is a consecutive chain of events/experiences that cumulatively lead individuals closer to suicidal behavior, ultimately framed as an "escape from the self." These sequential stages of suicide risk proceed as follows: (1) failure to meet one's self-expectations; (2) perceived fault for failure; (3) increased self-consciousness and perceived negative evaluation from others; (4) general increase of negative thoughts and emotions; (5) avoidance of meaningful thought and long-term goals; and (6) desire for suicide. ...
Article
Full-text available
Objective: Suicide prevention campaigns commonly employ brief informational materials aimed at emerging adults. Are such programs helpful, and do design characteristics yield differences in user outcomes? Literature is reviewed from the interpersonal theory of suicide, escape theory, and terror management theory, to inform our experimental design. Method: Participants (n = 977 MTurk emerging adults) reported demographics and suicide histories and were randomized to one of nine experimental cells with varying video and journaling conditions to approximate suicide prevention materials. Participants were surveyed on perceptions of the materials' risk reduction effectiveness, indicated their suicidality risk factors (e.g., hopelessness, depressiveness, purposelessness, and non-belongingness), and conducted an implicit association test of suicidality. Results: Suicide risk factors did not differ between experimental and control conditions, but certain conditions were rated as more effective (i.e., essay conditions prompting reflection, and the video condition featuring a personal/affective narrative). While there was no actual comparative reduction of risk, there was a perception that certain designs were more helpful. Conclusion: Real-world suicide prevention campaigns often feel justified despite lacking impact. Effective suicide risk reduction requires greater time investment and deeper personal connection than brief campaigns can offer, as well as systemic changes from a public health policy perspective.
... When people fail to meet the demands of life, the selfawareness of these failures is painful. To escape the painful-self people engage in immersing activities (Baumeister, 1990;Chatard & Selimbegovic, 2011;Heatherton & Baumeister, 1991). When people fail to meet the demands of life, they internalize these failures by self-blaming. ...
Article
E-shopping engagement has become a significant user activity in cyberspace. Prior scholarship has found self-escapism a crucial user motivation to engage in e-shopping. The current research identifies dimensions of self-escapism motivation to engage in e-shopping and their influence on shoppers’ attitude, e-cart use, and abandonment. Data were collected from 305 shoppers in Pakistan via an offline survey. Results reveal self-suppression and self-expansion as dimensions of shoppers’ self-escapism motivation to engage in e-shopping. These dimensions of self-escapism motivation determine shoppers’ positive affective attitude towards e-shopping. Nonetheless, they have a differing influence on shoppers’ e-cart use and abandonment. Keywords: Self-Escapism Motivation, E-Shopping, Affective Attitude, E-Cart Use; E-Cart Abandonment.
... negative thoughts (Bessenoff, 2006;Balcetis et al., 2013;Chatard & Selimbegović, 2011;Dittmar & Howard, 2004;Halliwell & Dittmar, 2004;Harrison, 2001;Harrison & Cantor, 1997;Heinberg & Thompson, 1995;Stice et al., 1994;Thornton & Maurice, 1997; for meta-analyses, see Grabe et al., 2008;Groesz et al., 2002;Want, 2009). Therefore, considering that thinness is a necessary characteristic of female beauty seems to play a significant role in a number of problematic thought and behavior patterns. ...
Article
Full-text available
The present research aimed to investigate whether retraining thin-beautiful associations could modify implicit beliefs relating thinness to beauty, increase me-beauty associations, and decrease explicit body anxiety in young women. In Experiment 1 (N = 180 women), participants were repeatedly exposed to beauty-related words paired with thin-related or large-related words on 50% (control) versus 75% (retraining) of trials. Implicit belief was assessed with a Relational Responding Task. In Experiment 2 (N = 195 women), me-beauty associations were assessed with a single-category Implicit Association Task, and body anxiety with a self-report measure. The implicit association measure remained unaffected by the retraining task. However, after retraining, women displayed weaker thin-is-beautiful implicit belief, Cohen’s 'd' = 0.44, 'CI95'[0.14, 0.74], and less body anxiety, Cohen’s 'd' = 0.34, 'CI95'[0.06, 0.63] than in the control condition. These results suggest that retraining thin-beautiful associations could reduce thin-is-beautiful implicit beliefs and decrease explicit body anxiety among women.
... Negative thoughts or attitudes about the self have been shown to increase implicit identification with self-destructive behaviors such as suicide. For example, Chatard and Selimbegović (2011) demonstrated that implicit identification with suicide, as measured by a word completion task, increased when participants were asked to think about to think about what would happen if they failed to find a job after college. Similarly, Tang, Wu, and Miao (2013) showed that implicit identification with death and suicide on an implicit association test increased when medical students were asked to think about a situation where they failed to become a physician. ...
Article
Non-suicidal self-injury (NSSI) is the intentional destruction of bodily tissue in the absence of suicidal motives. Individuals who self-injure often report doing so in order to punish the self or express self-hatred. Self-criticism, or thoughts marked by shame, self-consciousness, and inferiority, is associated with higher rates of NSSI while self-compassion, or the tendency to be sympathetic and caring with oneself, is negatively associated with NSSI. The aim of the present study was to determine whether experimentally-induced self-criticism would increase and self-compassion would decrease implicit identification with NSSI. Participants were randomly assigned to either a self-criticism induction, a self-compassion induction, or a neutral, control condition and completed a measure of strength of the automatic associations that a person holds between themselves and self-harming behaviors before and after the experimental induction. Study hypotheses were partially supported. Results showed that participants in the self-criticism induction experienced an increase in their implicit associations with NSSI while implicit associations in the self-compassion condition did not significantly change. These results highlight the importance of self-criticism in NSSI. Future research should examine increases in self-criticism as a potential precursor of NSSI in longitudinal samples.
Article
Online shopping is a well-recognized business and user activity. Moved by utilitarian and hedonic motives, millions of users visit online stores daily. Literature indicates users’ self-escapism as a crucial hedonic motivation of online shopping engagement. Yet, its influence has rarely been studied on users’ online shopping cart use and buying behavior. The current study assesses the dimensionality of users’ self-escapism motivation of online shopping engagement and its direct and mediated influence on their online shopping cart use and buying behavior. The data were gathered from 308 users of daraz.com and aliexpress.com in Pakistan using an offline survey. The findings indicate users’ self-escapism motivation of online shopping engagement as a two-dimensional construct consisting of self-suppression and self-expansion factors. Users’ self-escapism motivation of online shopping engagement directly but partially influences their online shopping cart use. The flow state and affective attitude toward online shopping play a significant mediating role during users’ online shopping engagement. In addition, users’ self-escapism motivated online shopping cart use influences their online buying behavior. The current study provides important insights about those users who engage in online shopping to regulate the self. Retailers can apply these insights to make the online stores desired spaces for such users and influence them to buy online.
Article
Objective: Converging evidence from basic science and experimental suicide research suggest that the anticipated consequences of suicide may have direct causal effects on suicidal behavior and accordingly represent a promising intervention target. Raising doubt about individuals' desirable anticipated consequences of suicide may be one means of disrupting this target. We tested this possibility across two complementary experimental studies. Method: Study 1 tested the effects of raising doubt about desirable anticipated consequences on virtual reality (VR) suicide in the lab, randomizing 413 participants across four conditions. In Study 2, 226 suicidal adults were randomized to an anticipated consequence manipulation or control condition then re-assessed at 2- and 8-weeks post-baseline. Results: In Study 1, anticipating that engaging in VR suicide would guarantee a desirable outcome significantly increased the VR suicide rate; conversely, raising doubt about the desirable anticipated consequences significantly reduced the VR suicide rate. In Study 2, raising doubt about the anticipated consequences of attempting suicide by firearm significantly reduced the perceived lethality of firearms as well as self-predicted likelihood of future suicide attempts, with effects sustained at 2-week follow-up. Conclusions: Findings suggest that raising doubt about desirable anticipated consequences of suicide merits further research as one potential approach to inhibit suicidal behavior.
Article
Objective Adolescence marks an important time to detect suicide risk, as suicidal ideation often emerges during this developmental period. Death-themed cognitions represent a promising domain of risk factors for suicidal ideation, but they have been understudied among adolescents. To address this knowledge gap, the present study examines the association between adolescents’ attitudes and beliefs about death, hereafter referred to as death conceptualizations, and suicidal ideation. Methods Seventy-four adolescents (12–19 years) completed a self-report measure assessing three types of death conceptualizations: Death Avoidance (i.e., suppression of death-related thoughts), Neutral Acceptance (i.e., belief that death is a natural part of life), and Escape Acceptance (i.e., belief that death is a viable escape from pain). Suicidal ideation was assessed at baseline, as well as 3 and 6 months later. Results At baseline, suicidal adolescents endorsed higher Escape Acceptance and lower Death Avoidance. Suicidal and nonsuicidal adolescents did not differ in their degree of Neutral Acceptance. Death conceptualizations, especially Escape Acceptance, also predicted future suicidal ideation. Conclusions Initial findings suggest that suicidal adolescents are more likely to believe that death is a viable escape from pain and that this death-related cognition is a risk factor for suicidal ideation.
The Benefits and Barriers model of NSSI postulates that engagement in NSSI is positively reinforced by potent benefits, however there are a host of barriers to engagement, any one of which is salient enough to prevent engagement. It is possible that individual differences in sensation seeking, a trait that describes optimal level of positive reinforcement, may alter the balance between the benefits and barriers of engagement in NSSI. There are significant associations between engagement in NSSI and sensation seeking in college undergraduates, a population with disproportionately high rates of NSSI. However, it is unclear whether these traits play a similar role in adolescents. We expected that higher levels of sensation seeking would positively relate to any NSSI history, lifetime frequency of NSSI, and earlier age at onset of NSSI among a sample of 200 adolescents in a psychiatric hospital. Consistent with previous research, results indicated that females were more likely to engage in NSSI than males. Additionally, increased sensation seeking was associated with greater likelihood of ever engaging in NSSI and a greater number of different NSSI methods tried. Though we expected sensation seeking would be significantly related to lifetime NSSI frequency and earlier onset of NSSI, it was not. Findings suggest that individual differences may alter relations between the benefits and barriers of NSSI and that measuring sensation seeking in adolescents, especially females, and especially those experiencing psychological distress, may identify those at highest risk for engaging in NSSI and may allow for targeted intervention with these individuals.
Article
This scholarship utilized self-study methodology to examine the first author’s experience as a non-formal educator and non-formal teacher educator within the context of making and tinkering with elementary-aged students and university students. Through analysis of multiple forms of data, tensions arose and took shape through my interactions with both sets of students as we navigated and negotiated our roles and actions within two activity systems—a formal learning environment (i.e. university course) and a non-formal learning environment (i.e. the makerspace). These tensions included (a) the framing and implementation of the making task, (b) my authority and power as the making “expert” and teacher educator, and (c) undergraduate students’ lack of competence and knowledge in making and science, technology, engineering, and mathematics (STEM). The push and pull or tug-of-war of my personal existence within and between the 2 activity systems are highlighted within these tensions as my ways of being were bounded and constrained by formal schooling practices and structures. The tensions should challenge teacher educators to consider the role of non-formal educators in shaping how we work with students of all ages. This involves developing and maintaining shared practices, knowledges, and discourses (i.e. third space), and an understanding of how postmodern perspectives may uncover our notions of “education.”
Chapter
Full-text available
The development and validation of a new measure, the Eating Disorder Inventory (EDI) is described. The EDI is a 64 item, self-report, multiscale measure designed for the assessment of psychological and behavioral traits common in anorexia nervosa (AN) and bulimia. The EDI consists of eight sub-scales measuring: 1) Drive for Thinness, 2) Bulimia, 3) Body Dissatisfaction, 4) Ineffectiveness, 5) Perfectionism, 6) Interpersonal Distrust, 7) Interoceptive Awareness and 8) Maturity Fears. Reliability (internal consistency) is established for all subscales and several indices of validity are presented. First, AN patients (N = 113) are differentiated from female comparison (FC) subjects (N = 577) using a cross-validation procedure. Secondly, patient self-report subscale scores agree with clinician ratings of subscale traits. Thirdly, clinically recovered AN patients score similarly to FCs on all subscales. Finally, convergent and discriminate validity are established for subscales. The EDI was also administered to groups of normal weight bulimic women, obese, and normal weight but formerly obese women, as well as a male comparison group. Group differences are reported and the potential utility of the EDI is discussed.
Article
Full-text available
Objective self-awareness theory has undergone fundamental changes in the 3 decades since Duval and Wicklund's (1972) original formulation. We review new evidence that bears on the basic tenets of the theory. Many of the assumptions of self-awareness theory require revision, particularly how expectancies influence approach and avoid - ance of self-standard discrepancies; the nature of standards, especially when they are changed; and the role of causal attribution in directing discrepancy reduction. How - ever, several unresolved conceptual issues remain; future theoretical and empirical directions are discussed. The human dilemma is that which arises out of a man's capacity to experience himself as both subject and ob- ject at the same time. Both are necessary—for the sci- ence of psychology, for therapy, and for gratifying liv- ing. (May, 1967, p. 8)
Article
The development and validation of a new measure, the Eating Disorder Inventory (EDI) is described. The EDI is a 64 item, self-report, multiscale measure designed for the assessment of psychological and behavioral traits common in anorexia nervosa (AN) and bulimia. The EDI consists of eight subscales measuring: Drive for Thinness, Bilimia, Body Dissatisfaction, Ineffectiveness, Perfectionism, Interpersonal Distrust, Interoceptive Awareness and Maturity Fears. Reliability (internal consistency) is established for all subscales and several indices of validity are presented. First, AN patients (N=113) are differentiated from femal comparison (FC) subjects (N=577) using a cross-validation procedure. Secondly, patient self-report subscale scores agree with clinician ratings of subscale traits. Thirdly, clinically recovered AN patients score similarly to FCs on all subscales. Finally, convergent and discriminant validity are established for subscales. The EDI was also administered to groups of normal weight bulimic women, obese, and normal weight but formerly obese women, as well as a male comparison group. Group differences are reported and the potential utility of the EDI is discussed.
Article
This article proposes that binge eating is motivated by a desire to escape from self-awareness. Binge eaters suffer from high standards and expectations, especially an acute sensitivity to the difficult (perceived) demands of others. When they fall short of these standards, they develop an aversive pattern of high self-awareness, characterized by unflattering views of self and concern over how they are perceived by others. These aversive self-perceptions are accompanied by emotional distress, which often includes anxiety and depression. To escape from this unpleasant state, binge eaters attempt the cognitive response of narrowing attention to the immediate stimulus environment and avoiding broadly meaningful thought. This narrowing of attention disengages normal inhibitions against eating and fosters an uncritical acceptance of irrational beliefs and thoughts. The escape model is capable of integrating much of the available evidence about binge eating.
Article
A review of past research reveals apparent gaps in many current measures of positive affect (PA) and negative affect (NA) including the Bradburn (1969) Affect Balance subscales popular among gerontologists and other social scientists. In many cases, the measures produce PA/NA correlations so high that some researchers claim subjective well-being is unidimensional. A viable alternative measure of positive and negative affect is the PANAS (Watson, Clark, and Tellegen 1988), which appears to distinguish clearly between these two emotional dimensions of subjective well-being. Unfortunately, however, only younger samples have served as subjects in research on the PANAS. Accordingly, the current study assessed the psychometric qualities of the PANAS applied to a sample of the old-old. Consistent with prior studies of younger samples, the results here clearly support the viability of the measure among the elderly. Of particular significance, the PA and NA dimensions appear completely independent of each other. More broadly, the results also support using the circumplex model of emotions when selecting mood adjectives to represent the affective dimensions of subjective well-being.