Journal of Personality and Social Psychology © 2011 American Psychological Association

2011, Vol. 101, No. 6, 1262–1277 0022-3514/11/$12.00 DOI: 10.1037/a0024407

Suspicious Spirits, Flexible Minds: When Distrust Enhances Creativity

Jennifer Mayer and Thomas Mussweiler

University of Cologne

Intuitively, as well as in light of prior research, distrust and creativity appear incompatible. The social
consequences of distrust include reluctance to share information, a quality detrimental to creativity in
social settings. At the same time, the cognitive concomitants of distrust bear resemblance to creative
cognition: Distrust seems to foster thinking about nonobvious alternatives to potentially deceptive
appearances. These cognitive underpinnings of distrust hold the provocative implication that distrust may
foster creativity. Mirroring these contradictory findings, we suggest that the social versus cognitive
consequences of distrust have diverging implications for creativity. We address this question in Study 1
by introducing private/public as a moderating variable for effects of distrust on creativity. Consistent with
distrust’s social consequences, subliminal distrust (vs. trust) priming had detrimental effects on creative
generation presumed to be public. Consistent with distrust’s cognitive consequences, though, an opposite
tendency emerged in private. Study 2 confirmed a beneficial effect of distrust on private creative
generation with a different priming method and pointed to cognitive flexibility as the mediating process.
Studies 3 and 4 showed increased category inclusiveness versus increased remote semantic spread after
distrust priming, consistent with enhanced cognitive flexibility as a consequence of distrust. Taken
together, these results provide evidence for the creativity-enhancing potential of distrust and suggest
cognitive flexibility as its underlying mechanism.

Keywords: creativity, cognitive flexibility, distrust, suspicion, priming

Creativity—the generation of work that is both novel and ap-
propriate (e.g., Sternberg & Lubart, 1999)—is an important skill
for coping with many different circumstances in life. Innovations
in science or economy depend heavily on people tapping their
creative potential. More mundane problems also often demand
leaving the beaten track to find creative solutions. The significance
of creativity in various domains is reflected in a broad scientific
interest in factors that foster or hinder creative performance (e.g.,
see Hennessey & Amabile, 2010, for an overview). Intrapsychic
variations in affect, cognition, and motivation as well as important
interpersonal factors like cooperation and conflict (Carnevale &
Probst, 1998; De Dreu & Nijstad, 2008) or social power (Galinsky,
Magee, Gruenfeld, & Whitson, 2008) have been studied with
respect to their influence on creativity. Interpersonal factors or,
more generally, social context might play a particularly important
role in creativity. For one, creativity is highly desired in interactive
contexts like the classroom or work teams. Furthermore, creative
performance crucially depends on its social reception: An idea is
creative only if the social environment considers it to be so (e.g.,
Simonton, 1997). In light of the fine line between novelty and
nonsense, displaying creative behavior often means going out on a
limb and running the risk of social ridicule. In addition, people
sharing their creative ideas run the risk of others stealing their most
novel and useful ones.
The risky business of creativity (cf. George & Zhou, 2007;
Glover, 1977; Runco, 2007) is therefore likely to be particularly
sensitive to the perceived benevolence of the social surroundings.
This interplay of expected versus doubted benevolence and the
willingness to take risks is reflected in the concepts of trust and
distrust (e.g., Schoorman, Mayer, & Davis, 2007; Simpson, 2007).
When people expect others to be by and large benevolent toward
them, they trust others and become willing to take risks (cf.
Schoorman et al., 2007). The more people doubt benevolent in-
tentions, however, the more risk adverse they become. The ex-
pression of creative ideas is thus likely to differ depending on
whether spirits are trustful or suspicious.

This article was published Online First July 11, 2011.

Jennifer Mayer and Thomas Mussweiler, Department of Psychology,
University of Cologne, Cologne, Germany.
This research was supported by a Gottfried-Wilhelm-Leibniz Award by
the German Science Foundation as well as by a European Young Investi-
gator Award from the European Science Foundation to Thomas Muss-
weiler. We would like to thank the members of Social Cognition Cologne
for their input. This research was conducted in the context of Jennifer
Mayer’s doctoral dissertation at the University of Cologne, Cologne,
Germany.
Correspondence concerning this article should be addressed to Jennifer
Mayer, Department Psychologie, Universität zu Köln, Richard-Strauss-Str.
2, 50931 Köln, Germany. E-mail: jennifer.mayer@uni-koeln.de
Suspicious Spirits in Creativity Research
Research aimed directly at the link between distrust and cre-
ativity is rather scarce and predominantly originates from organi-
zational psychology. The (mostly correlational) data point to a
negative effect of distrust on creativity (e.g., Dakhli & De Clercq,
2004; Ekvall, 1996; Ekvall & Ryhammar, 1999; Klimoski &
Karol, 1976). Distrust is seen as disruptive of a supporting and
cooperative team climate and, therefore, is seen as disruptive of
creativity and innovation as well (e.g., Amabile & Gryskiewicz,
1989; Hunter, Bedell, & Mumford, 2007). Research on factors
related to the concepts of trust and distrust strongly supports this

1262
 WHEN DISTRUST ENHANCES CREATIVITY
view: For example, psychological safety (i.e., the expectation that
others will respond favorably when one expresses concerns or
proposes new ideas) is seen as conducive to creativity in the
workplace (cf. Carmeli, Reiter-Palmon, & Ziv, 2010; Hennessey,
2003). Psychological safety, in turn, is certainly associated with
trust rather than with distrust.
Nevertheless, it is especially in situations of potential danger
that a problem must be solved on the spot, often in an unusual or
novel way. Creativity as a means of problem solving might be
especially important in such situations. Researchers have begun to
take a closer look at the potentially beneficial influence of chal-
lenging or threatening states on creativity. Along the way, the
understanding of the factors that hinder and foster creativity has
become considerably more complex. On the intrapersonal level, it
has been demonstrated that not only positive but also negative
affective states can foster creativity (De Dreu, Baas, & Nijstad,
2008). With regard to interpersonal factors, effective lying, for
instance, has been shown to be positively correlated with perfor-
mance in common measures of creative ideation (Walczyk, Runco,
Tripp, & Smith, 2008). Furthermore, it has been shown that
conflict is conducive to creativity if creative performance is func-
tional in the sense that it is related to handling the conflict at hand
(De Dreu & Nijstad, 2008).
In light of the latter findings, a creativity-enhancing effect of a
negative and threat-associated state like distrust seems at least not
out of the question. The cognitive characteristics of distrust even
cast considerable doubt on a unilaterally detrimental influence of
distrust on creativity, as we delineate in the following.
The Creative Mind of Suspicious Spirits
According to the high prevalence of lying and attempts at
deception documented in research (e.g., DePaulo & Kashy, 1998;
DePaulo, Kashy, Kirkendol, Wyer, & Epstein, 1996), distrust
constitutes a ubiquitous occurrence in everyday life (cf. Kramer,
1998). It is readily evoked by subtle cues of unreliability, by a
single questionable action, or by an isolated instance of betrayal
(cf. Marchand & Vonk, 2005). The resulting mind-set is, broadly
defined, characterized by doubts about whether appearances can be
taken at face value (cf. Schul, Mayo, & Burnstein, 2008). The case
of misleading appearances usually implies a disadvantage for the
deceived individual. Walking away from every potentially suspi-
cious character or disbelieving any potentially invalid information,
however, would not be a well-adapted option. Individuals might
miss out on opportunities, if the other person were actually trust-
worthy or the information were true (Fetchenhauer & Dunning,
2009).
A better strategy would be to invest in more elaborate informa-
tion processing (Chiappe et al., 2004; Hilton, Fein, & Miller, 1993;
Schul, Burnstein, & Bardi, 1996) and to prepare for both appear-
ances being true and appearances not being true. Schul, Mayo, and
Burnstein (2004) concluded that, if the latter was true, then the
opposite of any given information should be readily available in
situations of distrust. In line with this possibility, they demon-
strated that a state of distrust indeed involves a basic tendency to
activate incongruent cognitions. The effects of distrust, however,
are unlikely to be limited to the mere tendency to consider oppo-
sites. Instead, it seems reasonable that distrust should induce
individuals to try and figure out what a potentially hidden motive
1263
or, more generally, the truth might be. To do so, not only are
people in a state of distrust likely to engage in mind reading, but
they also appear to consider possible alternative scenarios
(Chiappe et al., 2004; Fein, Hilton, & Miller, 1990; Marchand &
Vonk, 2005; Schul, 1993; Schul et al., 1996). Moreover, being
skilled deceivers themselves, people are aware that for deception
to be successful, it is crucial to make the true motive as nonobvious
as possible. In the case of impending deception, people therefore
know that they are well advised to look for uncommon, nonroutine
occurrences. Research by Schul et al. (2008) supports this hypoth-
esis by showing that participants primed with distrust were partic-
ularly sensitive to nonroutine contingencies. In summary, suspi-
cious spirits are likely to interpret information from multiple
perspectives, which tend to be remote, uncommon ones. In contrast
to the notion that distrust is necessarily detrimental to creativity,
these characteristics perfectly mirror preconditions for thinking
outside the box—a common metaphor of creative cognition.
How may these cognitive consequences of distrust influence
creative thinking? To answer this question, it is important to
identify elements of creative cognition. According to the dual
pathway to creativity model (De Dreu et al., 2008), cognitive
flexibility and persistence are the two critical processes of creative
cognition. Using the flexibility pathway, novel ideas originate
from changing perspectives (set breaking; e.g., Smith & Blanken-
ship, 1991) and flat associative hierarchies (Mednick, 1962). This
kind of creative cognition becomes manifest in the use of many
different cognitive categories as well as in the breadth and inclu-
siveness of the categories themselves. Using the persistence path-
way, original ideas are discovered by systematically and persever-
ingly exploring few cognitive categories in depth. This alternative
process of creative cognition usually results in prolonged effort
and increased time on task.
How can these pathways of creative cognition be linked to
distrust? On the one hand, preconditions of the persistence path-
way map nicely onto the affective–motivational properties of
distrust, which are comparable with those of other states involving
threatening circumstances: Compared with trust and neutral con-
ditions, distrust is associated with a more negative affective state
that also involves a higher level of arousal (cf. Schul et al., 2008).
Negative activating affective states, in turn, have been shown to
have the potential to enhance creativity through the persistence
pathway (De Dreu et al., 2008). Furthermore, distrust can be seen
as associated with an avoidance motivational orientation (Muss-
weiler & Burgmer, 2011). Avoidance orientations, in turn, have
been connected to narrow and systematic processing (cf. Friedman
& Förster, 2000), which is consistent with findings of elaborate
processing in a distrust mind-set (e.g., Chiappe et al., 2004; Hilton
et al., 1993; Schul et al., 1996) and, therefore, again, the persis-
tence pathway.
On the other hand, people in a distrust mind-set not only
elaborate more, but they do so in a specific way: They seem to
entertain multiple interpretations of potentially invalid information
rather than to elaborate intensely on that information within only
one interpretative frame (Fein et al., 1990; Schul et al., 1996). The
latter is consistent with category diversity and is thus indicative of
cognitive flexibility. A comparable basic tendency to consider
alternatives is evoked by counterfactual thinking (Galinsky &
Moskowitz, 2000), the process of mentally creating alternative
outcomes to a given event (cf. Epstude & Roese, 2008). Counter-
1264 MAYER AND MUSSWEILER
factual thinking, in turn, has been shown to foster creativity
(Markman, Lindberg, Kray, & Galinsky, 2007) and to be helpful in
overcoming functional fixedness (Galinsky & Moskowitz, 2000).
The latter is indicative of set breaking and, therefore, again, of
cognitive flexibility. Consistent with the general possibility of
increased cognitive flexibility in states involving potential threat,
De Dreu and Nijstad (2008) showed that participants in a compet-
itive conflict mind-set exhibited increased, targeted category di-
versity as well as category inclusiveness for conflict-related ma-
terial. Recently, a mind-set quite similar to the one provoked by
impending deceit (i.e., a competitive mind-set that also involves
the inclination to engage in mind reading) has been shown to
enhance executive functioning in terms of cognitive flexibility
more generally (Ybarra, Winkielman, Yeh, Burnstein, & Ka-
vanagh, 2011). Finally, a distrust mind-set has been shown to
foster sensitivity to nonroutine occurrences (Schul et al., 2008),
consistent with an increased accessibility of uncommon, remote
alternatives. The latter is consistent with flat associative hierar-
chies (Mednick, 1962) as another manifestation of the flexibility
pathway.
Valid reasons for assuming both enhanced persistence and en-
hanced cognitive flexibility in a distrust mind-set strongly impli-
cate the possibility of a creativity-enhancing effect of distrust.
Social Versus Private Creativity
The cognitive consequences of distrust strongly suggest a ben-
eficial effect of suspicious spirits on creative performance. Yet,
this seems at odds with findings from a review of the research on
the link between distrust and creativity, which indicates a detri-
mental effect. A closer look at the two contradicting lines of
research reveals a possible moderator: Most research hinting at
detrimental effects of distrust and distrust-related variables on
creativity refers to something one might term public creativity (i.e.,
creative tasks performed in group settings). For this kind of cre-
ativity, it is crucial not only to come up with creative ideas but to
share them as well. Consequently, creativity in public not only
depends on how creative the individual is but is also crucially
conditional upon the degree to which individual-level creativity
comes into effect (e.g., in a group setting; Taggar, 2002). For
individual creativity to play out in a group setting, a positive,
benevolent, and psychologically safe group climate seems to be
beneficial: Groups consisting of members with prosocial (rather
than proself) motives exhibit increased creativity (Bechtoldt, De
Dreu, Nijstad, & Choi, 2010). Similarly, individuals’ agreeable-
ness and conscientiousness contribute significantly to creative
performance on the group level (Taggar, 2002). Because group
climate also reacts upon individual level creativity, a creativity-
enhancing social environment needs to encourage constructive,
task-related criticism and must not be too soothing either. Consis-
tently, stimulation brought on by changing group composition
(Nemeth & Ormiston, 2007) or by a preceding competitive nego-
tiation (Beersma & De Dreu, 2005) has been shown to foster group
creativity. In sum, for public creativity to be successful, social
climate needs to stimulate individual creativity without preventing
idea sharing. If idea sharing is hindered (e.g., because of social
anxiousness), public creative performance decreases even though
private performance is not necessarily diminished (e.g., Camacho
& Paulus, 1995).
The latter is highly relevant with regard to the social conse-
quences of distrust: Although the cognitive characteristics of the
accompanying mind-set may foster individual creativity, distrust
will keep people from sharing ideas—and probably the more
unusual and possibly crazy ones in particular—in order to protect
themselves against exploitation and ridicule (Amabile & Gryskie-
wicz, 1989; Edmondson, 2004; Ekvall, 1996; Ekvall & Ryham-
mar, 1999; Klimoski & Karol, 1976; Petermann, 1992; Rotenberg
et al., 2010; Schoorman et al., 2007; West, 2002; see also De
Cremer, 1999). Potentially creativity-enhancing effects of distrust
and other threat-involving interpersonal factors are therefore un-
likely to surface in public settings. Consistently, findings of
heightened creativity in a conflict mind-set or related to effective
lying are characterized by the relative privacy of the creative act:
Competitive negotiation tactics are not disclosed to the other party
for scrutiny (De Dreu & Nijstad, 2008) nor will a liar reveal his or
her (creative) considerations in coming up with an effective lie.
Social climate might therefore be less important to creativity in
private, making room for the potentially beneficial cognitive con-
sequences of a distrusting state of mind.
Current Research
In accordance with the diverging social and cognitive conse-
quences of distrust, we predict a detrimental influence of distrust
on creative performance when the task involves public sharing of
creative ideas. In cases where the task does not require sharing
ideas, however, we hypothesize the opposite to be true, namely,
increased creativity in a distrusting state of mind. The following
four studies set out to investigate these influences of distrust on
creativity. Study 1 aims at disentangling the differential influence
of distrust on creativity depending on whether public information
sharing is inherent to the creativity task. Studies 2 through 4 focus
on the hypothesized beneficial influence of the cognitive conse-
quences of distrust on creative cognition in private.
For these private situations, distrust as we spontaneously picture
it—suspicion focused at a specific person in a particular situa-
tion—might not be the most relevant form of distrust. Yet, dis-
trusting states are not restricted to instances of actual interactions
with potentially untrustworthy individuals. Rather, the ubiquity of
distrust and its significance in preventing harm due to impending
deception make it likely that an integrated mental representation
has evolved, comprising the respective affective, cognitive, and
behavioral tendencies (Cosmides & Tooby, 2005; see also Schul et
al., 2008). In the case of such an integrated distrust representation,
even subtle distrust-related cues should be capable of evoking the
whole mental representation, that is, a distrust mind-set (cf. Schul
et al., 1996, 2004, 2008). Consequently, a distrust mind-set is also
likely to be triggered by cues that are typically seen as being
related to deception, for example, certain facial or vocal features
(see DePaulo et al., 2003; Schul et al., 2004; Zebrowitz, Voinescu,
& Collins, 1996) and distrust-related semantic cues (see Friesen &
Sinclair, 2011; Rotenberg et al., 2010). Therefore, the concept is
also amply relevant to a wide array of private situations, in which
we expect the creativity-enhancing effects of a distrust mind-set to
emerge.
With our operationalizations of distrust, we intended to (a)
closely model the form of distrust relevant to private creativity and
(b) instill distrust in a pure fashion (“mere state of distrust”; Schul
 WHEN DISTRUST ENHANCES CREATIVITY
et al., 2004, p. 669) that does not simultaneously involve additional
social factors, such as, for example, decreased liking of a specific
target (cf. Rule et al., 2010; Tyler, Feldman, & Reichert, 2006).
Consequently, we chose to manipulate distrust in a subtle way.
This approach has been taken not only in previous research on
distrust (Rotenberg et al., 2010; Schul et al., 2004, 2008), but also
in research on other important interpersonal variables like social
power (Galinsky et al., 2008). Specifically, Studies 1, 3, and 4 used
a subliminal priming procedure embedded in a lexical decision
task. Study 2 used a supraliminal priming procedure, namely a
scrambled sentences task. Both procedures are widely used for
concept activation in an unobtrusive manner (cf. Bargh & Char-
trand, 2000).
As our dependent measures, we used three tasks that tap differ-
ent aspects of creative thinking. In Studies 1 and 2, creativity was
assessed by one of the most widely used creativity measures,
namely, a prototypical idea-generation task. Tasks of that kind
allow not only for the investigation of originality as the core
component of creativity, but also for a preliminary exploration of
the pathways thereto (i.e., of cognitive flexibility and persistence).
In Studies 3 and 4, we took a closer look at the role of the
flexibility pathway, for which the cognitive consequences of dis-
trust suggest a particularly important role. We did so by measuring
category inclusiveness and flat associative hierarchies, respec-
tively, as two central manifestations of cognitive flexibility. All
studies were computer based, and all participants were native
German speakers.
Study 1
Study 1 was designed to examine whether distrust and trust have
differential effects on creativity depending on whether creative
ideas will be made public. To that end, we first induced a distrust
mind-set or a trust mind-set in participants by means of a sublim-
inal priming procedure. Subsequently, we asked participants to
work on a creativity task. One group of participants believed their
performance would remain private, whereas another group of
participants were told that their performance was made public to a
participant next door. In accordance with prior research, we
expected distrust to have a detrimental effect on creativity if
creative performance required public sharing. We did, however,
expect this disadvantage to vanish and the beneficial cognitive
consequences of distrust to take effect when creative performance
remained private.
Creative performance was measured by means of an idea-
generation task, which allows for the investigation of different
aspects of creativity. Idea-generation tasks are typically scored for
quantity (ideational fluency), quality (originality), and diversity of
production (flexibility; e.g., Plucker & Renzulli, 1999). Addition-
ally, from the latter two measures an indicator of persistence can
be calculated (cf. De Dreu et al., 2008; Nijstad, Stroebe, &
Lodewijkx, 2002). Out of these indicators, originality certainly
constitutes the core component of creativity. Therefore, the hy-
pothesized differential influences of distrust compared with trust
depending on public versus private performance concerns first and
foremost the originality of solutions. Quality of solutions (origi-
nality) may come along with quantity (e.g., Diehl & Stroebe, 1987;
Hocevar, 1979; Osborn, 1963), but this need not be the case. A
conceptually similar pattern for ideational fluency is therefore
1265
conceivable but not central to our hypothesis. The remaining
measures of flexibility and persistence allow for an exploratory
investigation as to which one of the two alternative pathways to
creativity proposed by De Dreu et al. (2008) may be the critical
one.
Methods
Design and participants. Study 1 is based on a 2 (type of
priming: distrust vs. trust) ⫻ 2 (context of idea generation: public
vs. private) between-subjects factorial design. We recruited 68
students as participants and randomly assigned them to one of the
four experimental conditions. Participants were approached on a
campus of the University of Cologne, Cologne, Germany, and
offered chocolate and coffee vouchers as compensation. Three
participants were excluded because of data collection irregulari-
ties, leaving a final sample of 65 participants (44 female, 21 male).
Materials and procedure. When the experimenter and the
participant arrived at the lab, they found a female confederate
waiting in front of the lab door, pretending to be another partici-
pant. Inside the lab, the experimenter ushered the participant and
the confederate into two adjacent, sound insulated cubicles and
asked them to sit down in front of a 75-Hz computer monitor.
After giving their consent, participants proceeded to the first
task, which was introduced as a lexical decision task. This task
served as a means to subliminally and repeatedly present partici-
pants with either the German verb misstrauen [to distrust] or the
verb vertrauen [to trust] (cf. Dijksterhuis, Aarts, Bargh, & van
Knippenberg, 2000; Mussweiler & Förster, 2000). Such a proce-
dure has been shown to yield reliable differences in a trust game,
indicating that compared with the trust condition, participants in
the distrust condition trusted less (Posten & Mussweiler, 2011).
Instructions to the task pointed out that participants were to
focus on the center of the screen where letter strings would appear.
Their task was to decide whether any given letter string constituted
a proper German word. Each of the 26 trials began with the
presentation of a fixation string (XWXXWWXXWX) for 3,000 ms.
Without the awareness of participants, the fixation string was
replaced by the prime word (13 ms), which was immediately
masked by the fixation string again (506 ms; cf. suggestions by
Bargh & Chartrand, 2000). Next, the target letter string overwrote
the fixation string and remained on the screen until participants
indicated their decision by pressing either a blue computer key (for
proper German words) or a yellow key (for nonwords). The first
two trials served as practice trials, and instead of the prime word,
a blank screen was presented. The subsequent 24 experimental
trials each contained the same prime word (i.e., misstrauen [to
distrust] in the distrust condition or vertrauen [to trust] in the trust
condition). The target letter strings were proper German words in
one of the practice trials and 18 of the experimental trials (e.g.,
Jacke [jacket], kämmen [to comb]) and were nonsense letter strings
in the remaining seven trials (e.g., grompem, Bealk).
After completion of the priming task, participants continued on
to a second, ostensibly unrelated study. This study consisted of an
idea-generation task and constituted the dependent measure of
creativity. In addition, instructions for the task served as a means
to introduce the second experimental factor, that is, the context of
idea generation (public vs. private). Participants were told that we
were interested in how the simultaneous processing of ideas pro-
1266 MAYER AND MUSSWEILER

vided by other people—a characteristic of brainstorming in The flexibility index was calculated by averaging the numbers of
groups—influenced generating ideas oneself. The instructions fur- different categories per participant that resulted from the ratings of
ther read that there were two different roles: creating ideas while the two independent coders. Persistence is characterized by an
processing ideas of others versus creating ideas undisturbed. Ac- in-depth exploration of categories. The more ideas participants
tually, the computer assigned all participants the latter role. De- generate within any tackled category, the higher their persistence.
pending on condition, however, participants received different Accordingly, we derived a measure of average within-category
information about what would happen to the ideas they were about fluency by dividing the number of unique ideas (ideational flu-
to generate. In the public condition, instructions explained that the ency) by the number of categories these ideas were derived from
person in the adjacent cubicle (the confederate) was assigned the (flexibility; De Dreu et al., 2008; Nijstad et al., 2002).
role of “generating ideas while processing the ideas of others.”
Participants were told that this would be achieved by transferring Results
every idea they generated to the other person’s computer screen,
implying that their ideas would be made public to the other person. In the main analysis, we submitted the originality measure to a
In the private condition, instructions did not mention the other 2 (type of priming: distrust vs. trust) ⫻ 2 (context of idea gener-
person but simply stated that the participant’s ideas would be ation: public vs. private) analysis of variance (ANOVA). Analo-
saved on a server. gous analyses were performed for the remaining three indicators.
Next, participants received detailed instructions for the idea- Main analysis: Originality. Inspection of Table 1 reveals
generation task, a so-called alternative uses task (Guilford, 1967), that the mean originality scores are in line with the predicted
in which participants were told that they were to generate as many pattern: In the public condition, participants primed with distrust
alternative uses for an everyday object as they could think of. achieved lower originality scores than participants primed with
Before beginning idea generation, participants were asked to click trust. In the private condition, participants primed with distrust
on a button to check whether the person in the adjacent cubicle was tended to exhibit higher originality scores compared with partici-
ready to begin (public condition) versus whether the connection to pants primed with trust. This pattern was borne out in a significant
the server was working properly (private condition). This was done interaction effect, F(1, 61) ⫽ 5.89, p ⫽ .02, ␩2p ⫽ .09 (all other
to lend credibility to the cover story. After a few seconds, partic- Fs ⬍ 1, ns).
ipants read that the person in the adjacent cubicle (vs. the server) Supplementary analyses: Ideational fluency, flexibility, and
had confirmed the connection. On the next screen, participants persistence. Our hypotheses for ideational fluency were con-
were presented with the object for which they should generate ceptually parallel to those for originality. The pattern of means
alternative uses (a newspaper; cf. Wallach & Kogan, 1965) as well (Table 1) is consistent with this reasoning: In the public condition,
as an input field to type in their ideas. participants primed with distrust tended to achieve lower fluency
A final questionnaire asked for demographic data and tested for scores than participants primed with trust, whereas the opposite
awareness of the true purpose of the study as well as for awareness tendency resulted from the private condition. The respective two-
of the prime word in the priming procedure. None of the partici- way interaction did not reach significance, F(1, 61) ⫽ 2.07, p ⫽
pants indicated awareness of the true purpose of the study or of the
prime word. At the end, participants were thanked and handed
their compensation. Table 1
Measures and dependent variables. Indicators of creative Means and Standard Deviations of Mean Originality of Ideas,
performance were derived from ideas generated in 2 min (cf. Number of Ideas (Fluency), Number of Categories (Flexibility),
Gilhooly, Fioratou, Anthony, & Wynn, 2007; Schoppe, 1975). In and Within-Category Fluency (Persistence) as a Function of
a first step, ideas were coded for ideational fluency, originality, Type of Priming (Distrust vs. Trust) and Context of Idea
and flexibility. In a second step, a measure of persistence was Generation Condition (Public vs. Private) in Study 1
calculated from the fluency and flexibility measures.
The number of nonredundant ideas constituted the measure of Context of idea generation
ideational fluency. For the originality measure, two independent Public Private
coders rated every idea on a 5-point scale, ranging from 1 (not at
all creative) to 5 (very creative). Interrater reliability was good Measure Distrust Trust Distrust Trust
(intraclass correlation coefficient ⫽ .78). Ratings of the two coders
Originality
were averaged per idea. The mean of these values for each par- M 1.41a 1.68b 1.58a,b 1.43a
ticipant’s set of ideas constituted the originality measure. Flexibil- SD 0.44 0.40 0.29 0.22
ity is reflected in the number of different semantic categories from Fluency
which ideas are derived. The more categories participants use, the M 6.12a 7.13a,b 8.07b 7.41a,b
higher their cognitive flexibility. To create a category system, the SD 2.23 2.03 3.06 1.91
Flexibility
first author reviewed ideas generated by a different group of 28 M 4.50a 5.63b 5.27a,b 4.94a,b
participants and extracted 14 categories (e.g., crafting, hygiene, SD 1.27 1.63 1.40 1.29
weapon) complemented by a miscellaneous category for ideas that Persistence
did not fit any of the semantic categories. Two independent coders M 1.37a,b 1.28b 1.52a 1.53a
SD 0.37 0.18 0.38 0.35
assigned a category to each idea. Interrater reliability was excellent
for single ideas (Cohen’s ␬ ⫽ .94) and for the number of different Note. Within each row, means not labeled with the same superscript
categories per participant (intraclass correlation coefficient ⫽ .98). differed according to the respective simple-contrast analysis at p ⬍ .05.
 WHEN DISTRUST ENHANCES CREATIVITY
.16. A trend toward a main effect for context of idea generation
emerged, F(1, 61) ⫽ 3.74, p ⫽ .06, ␩2p ⫽ .06. No main effect for
type of priming emerged, F ⬍ 1, ns.
For flexibility, inspection of the means in Table 1 reveals a
pattern parallel to the one for originality: In the public condition,
participants primed with distrust derived their ideas from fewer
different categories than participants primed with trust. In the
private condition, participants primed with distrust tended to ex-
hibit higher flexibility scores compared with participants primed
with trust. As for originality, this pattern resulted in a significant
interaction effect, F(1, 61) ⫽ 4.35, p ⫽ .04, ␩2p ⫽ .07 (all other
Fs ⬍ 1.33, ps ⬎ .25). For the persistence measure, inspection of
the means in Table 1 suggests a pattern different from those
observed for the other indicators. In both the public and the private
conditions, persistence was similar for participants primed with
distrust and for participants primed with trust. Independent of type
of priming condition, persistence values were higher for partici-
pants in the private condition (combined M ⫽ 1.53, SD ⫽ 0.36)
than for participants in the public condition (combined M ⫽ 1.33,
SD ⫽ 0.29), F(1, 61) ⫽ 6.09, p ⫽ .02, ␩2p ⫽ .09. No other
significant effects emerged, all Fs ⬍ 1.
Discussion
Results of Study 1 are generally in line with our hypotheses.
Effects of trust and distrust on originality crucially depended on
whether creative behavior was carried out publicly or privately. In
the public condition, we demonstrated that distrust is detrimental
to creativity: Participants who were told that their ideas would be
made public to another person produced less original ideas when in
a distrust mind-set than when in a trust mind-set. This was not the
case, however, for participants in the private condition whose ideas
were ostensibly merely saved to a special server and, thus, re-
mained private. In line with previous findings (e.g., Diehl &
Stroebe, 1987; Rietzschel, Nijstad, & Stroebe, 2007), a conceptu-
ally similar pattern for ideational fluency emerged. The same was
true for category diversity as the indicator of flexibility: In the
public condition, distrust yielded less category diversity than trust,
which was not the case in the private condition. Results from the
persistence measure indicated no differences for participants
primed with trust and participants primed with distrust regardless
of context of idea generation. The parallel between originality and
flexibility measures, and the missing parallel between originality
and persistence measures, might be taken as preliminary evidence
for cognitive flexibility as the critical factor with regard to the
influence of (dis)trust on creativity.
In sum, the results provide initial support for our hypothesis by
revealing that distrust is not necessarily detrimental to creativity.
On a descriptive level, results actually imply a reversal in the
pattern of means in the private condition: Originality, fluency, and
flexibility scores of participants primed with distrust tended to be
higher than those of participants primed with trust. In simple effect
analyses, these differences did not reach statistical significance (cf.
superscripts in Table 1). This may be due to the moderate level of
privacy evoked by our operationalization of the private context of
idea generation: For reasons of experimental control, the private
condition had to be modeled closely after the public condition. To
that end, participants in the private condition were informed that
their ideas would be saved on a server. This operationalization
1267
certainly involves a higher level of privacy in the private condition
than in the public condition where ideas ostensibly were trans-
ferred to another person. Nevertheless, participants in the private
condition may well have assumed that their ideas would be ana-
lyzed by the experimenter at some later point in time and that they
would not remain so private after all. This may have prevented a
potential creativity-enhancing effect of distrust to become readily
apparent.
Study 2
We designed Study 2 to examine the effects of distrust under
conditions of true privacy. Accordingly, participants were not
made to believe that their ideas would be transferred to a server
and were thus likely to experience a heightened sense of privacy so
that the creativity-enhancing effects of distrust may become ap-
parent.
We used the same creativity measure—an idea-generation
task—as in Study 1 but used a different priming manipulation for
generalizability purposes. To that end, we adapted another widely
used priming procedure, namely, a scrambled sentences task (Srull
& Wyer, 1979), to induce distrust versus trust. Furthermore, we
added a neutral control group. So far, trust manipulations have
repeatedly yielded results similar to those of the control conditions
in experimental studies (Klimoski & Karol, 1976; Schul et al.,
2004; cf. Schul et al., 2008), most likely because without reasons
for doubt, people typically have trust in others (e.g., Berg, Dick-
haut, & McCabe, 1995; McKnight, Cummings, & Chervany, 1998;
cf. Gilbert, 1991). In light of this default to trust, we expected
similar levels of creativity in both the trust and control conditions.
Our theoretical analysis, however, holds that— compared with the
trust and control conditions— distrust enhances creativity.
Methods
Design and participants. We recruited 60 psychology stu-
dents as participants and offered them partial course credit as
compensation. They were randomly assigned to either a distrust
condition, a trust condition, or a control condition. Three partici-
pants were excluded because of data collection irregularities, leav-
ing a final sample of 57 participants (39 female, 18 male).
Materials and procedure. On arrival in the lab, participants
were told that they would be given two short experimental tasks.
The first of these was the scrambled sentences priming task (Srull
& Wyer, 1979). Instructions to this task pointed out that, in each
trial, five words would be presented in random order. Four out of
the five words formed a meaningful sentence. Participants were
asked to quickly form the correct sentence and to type it into the
computer. Seven out of a total of 15 trials were identical for all
three conditions. These trials did not contain prime words and were
chosen to be unrelated to distrust and trust. In the control condi-
tion, the same was true for the remaining eight trials. In the distrust
and the trust conditions, these remaining trials contained prime
words closely related to distrust and trust, respectively (see Table
2; for a similar procedure, cf. Friesen & Sinclair, 2011). The
second, ostensibly unrelated, task was the alternative uses task, our
creativity measure. Instructions were mostly similar to the ones
described in Study 1 with the critical exception that this time no
mention was made that participants’ ideas were transferred to a
server or another person.
1268 MAYER AND MUSSWEILER
Table 2
German Prime Words (Uninflected Forms) and Their General-Sense English Equivalent as Used
in the Experimental Trials of the Distrust and Trust Conditions in Study 2
Type of priming
Distrust
German English
prime words equivalents
misstrauen to distrust
vorgetäuscht feigned
zweifeln to doubt
Argwohn suspicion
unsicher uncertain
fraglich questionable
verdächtig suspicious
sich täuschen to be mistaken
Note. Prime words were embedded in a scrambled sentences task.
A final questionnaire asked participants to provide some demo-
graphic data. A concluding question probed for suspicion of the
true purpose of the study. None of the participants indicated
awareness. At the end, participants were thanked and given their
course credit as compensation.
Measures and dependent variables. Measures of ideational
fluency, originality, flexibility, and persistence were derived as
described in Study 1. Interrater reliabilities were comparable with
the ones in Study 1 for the originality ratings (intraclass correlation
coefficient ⫽ .74) as well as for the flexibility coding (Cohen’s
␬ ⫽ .91 for single ideas; intraclass correlation coefficient ⫽ .98 for
the number of different categories per participant).
Results
Our hypothesis specifically predicts differing scores for the
distrust condition compared with the other two conditions. There-
fore, we used the corresponding contrast to analyze the originality
scores as well as the other indices. Means and standard deviations
for all four measures are given in Table 3.
Main analysis: Originality. As expected, participants in the
distrust condition achieved the highest originality scores out of
the three priming conditions. The planned contrast comparing the
distrust condition (M ⫽ 1.48, SD ⫽ 0.32) to the other two condi-
tions (combined M ⫽ 1.32, SD ⫽ 0.21) was statistically reliable,
t(54) ⫽ 2.21, p ⫽ .02, Cohen’s d ⫽ 0.66.1
Supplementary analyses: Ideational fluency, flexibility, and
persistence. For ideational fluency, a pattern similar to the one
observed in the originality measure emerged. Participants primed
with distrust produced more ideas (M ⫽ 8.81, SD ⫽ 2.66) com-
pared with participants in the other two conditions (combined M ⫽
6.71, SD ⫽ 1.94), t(54) ⫽ 3.28, p ⫽ .001, Cohen’s d ⫽ 0.99. In
line with the notion that cognitive flexibility may be critical for the
influence of distrust on creativity, participants primed with distrust
derived their ideas from more categories (M ⫽ 6.06, SD ⫽ 1.60)
than participants in the trust and control conditions (combined
M ⫽ 4.91, SD ⫽ 1.39), t(54) ⫽ 2.66, p ⫽ .005, Cohen’s d ⫽ 0.81.
For the persistence measure, on a descriptive level, a pattern
similar to those for the other measures emerged (i.e., highest
Trust
German English
prime words equivalents
vertraut trusting
glaubwürdig credible
authentisch authentic
Verlass dependence
sicher certain
zuverlässig reliable
ehrlich honest
solidarisch showing solidarity
persistence scores in the distrust condition). Yet, the contrast
comparing the distrust condition with the other two conditions was
not significant (|t| ⬍ 1).
Mediation analysis. The parallel pattern of results for orig-
inality and flexibility and the lack of such a parallel pattern for
originality and persistence suggest that flexibility is the more
important pathway to creativity in the present context. In a next
step, therefore, we tested whether flexibility mediated the effect of
the priming manipulation on originality (see Figure 1 for the
mediational model and the respective regression coefficients).
Following recent recommendations (Preacher & Hayes, 2004;
Zhao, Lynch, & Chen, 2010), we conducted a bootstrapping anal-
ysis (Preacher & Hayes, 2004). Results revealed that, indeed,
flexibility mediated the effect of type of priming on originality,
yielding a point estimate of .029 for the indirect effect with a 95%
confidence interval (CI) ranging from .006 to .061.
Discussion
We set up Study 2 to investigate the influence of distrust on
creative performance in a mostly nonpublic situation and with a
different priming method. As expected, distrust priming yielded
increased originality, ideational fluency, and flexibility compared
with trust and neutral control conditions. No reliable differences
emerged for the persistence measure. Thus, Study 2 provides
evidence for an actual creativity-enhancing effect of distrust under
conditions of true privacy. To determine the overall reliability of
this effect, we used a meta-analytic procedure (cf. Rosenthal,
1991) for the contrast comparing the distrust and trust conditions
in Studies 1 (private condition only) and 2. Results of these
1
All p values of reported t tests and meta-analytic procedures involving
hypothesized differences are one tailed. Although we had clear directional
predictions and many authorities recommend the use of one-tailed signif-
icance tests in such cases, this practice is controversial (e.g., Abelson,
1995). We adopted Abelson’s (1995) recommended compromise between
the two extreme positions in this debate, constructing a null-hypothesis
rejection region of 5% in the theoretically expected tail and 0.5% in the
unexpected tail.
 WHEN DISTRUST ENHANCES CREATIVITY 1269

Table 3 of priming on flexibility or, conversely, flexibility mediating the

Means and Standard Deviations of Mean Originality of Ideas, effect of priming on fluency. Additional bootstrapping analyses
Number of Ideas (Fluency), Number of Categories (Flexibility), revealed a significant indirect effect for either of the two models:
and Within-Category Fluency (Persistence) as a Function of Fluency mediated the effect of priming on flexibility (point esti-
Type of Priming in Study 2 mate ⫽ .294 and 95% CI [.108, .517]), and flexibility mediated the
effect of priming on fluency (point estimate ⫽ .352 and 95% CI
Type of priming [.087, .666]). Therefore, the data presented so far do not unequiv-
Measure Distrust Control Trust
ocally demonstrate that a distrust mind-set does indeed come along
with an increase in cognitive flexibility.
Originality In the following two studies, we aim at exploring more directly
M 1.48a,† 1.36b,† 1.29b the potential of cognitive flexibility as the underlying mechanism
SD 0.32 0.24 0.18
of the creativity-enhancing effect of distrust demonstrated earlier.
Fluency
M 8.81a 6.76b 6.65b According to the dual pathway to creativity model (De Dreu et al.,
SD 2.66 1.79 2.13 2008), manifestations of the flexibility pathway include (a) the use
Flexibility of many different categories, (b) breadth and inclusiveness of the
M 6.06a 4.79b 5.05b categories themselves, and (c) flat associative hierarchies.
SD 1.60 1.10 1.65
Persistence Whereas Studies 1 and 2 provided evidence for category diversity,
M 1.49a 1.43a 1.37a Studies 3 and 4 used measures to investigate the other two facets
SD 0.38 0.31 0.33 of the flexibility pathway—inclusiveness of categories and flat
associative hierarchies.
Note. Within each row, means not labeled with the same superscript
differed according to the respective simple-contrast analysis at p ⬍ .05.
Means that are labeled with different superscripts and also share an Study 3
additional superscript (†) differed at p ⫽ .06.
In an attempt to corroborate our preliminary findings on cogni-
tive flexibility, we chose to use the category inclusion task (Isen &
analyses confirmed that, compared with trust priming, distrust Daubman, 1984; Rosch, 1975) that has been widely used in cre-
priming reliably increased originality (Z ⫽ 2.45, p ⫽ .007), flu- ativity research (e.g., Carnevale & Probst, 1998; De Dreu &
ency (Z ⫽ 2.54, p ⫽ .005), and flexibility (Z ⫽ 1.89, p ⫽ .03). Nijstad, 2008; Friedman & Förster, 2000; Rietzschel, De Dreu, &
Furthermore, the mediation analysis suggests flexibility as the Nijstad, 2007). This measure captures cognitive flexibility through
cognitive mechanism that may be responsible for the creativity- breadth and inclusiveness of categories, which, in turn, is seen as
enhancing effect of distrust. conducive to producing original ideas (Murray, Sujan, Hirt, &
The latter finding has to be taken with caution, however: Flex- Sujan, 1990). The category inclusion task asks participants to rate
ibility indices in idea-generation tasks are disputable because of the goodness of fit of several exemplars (e.g., car, wheelchair,
their possible contamination by fluency (e.g., Hocevar, 1979; camel) to a superordinate category (e.g., vehicle). Seeing highly
Plucker & Renzulli, 1999): The more ideas individuals produce, and moderately typical exemplars (e.g., car, wheelchair) as good
not only are they more likely to run out of trivial responses representatives of their category is not presumed to require cog-
(quantity breeds quality; e.g., Diehl & Stroebe, 1987; Osborn, nitive flexibility (Isen & Daubman, 1984). These exemplars
1963), but they are also more likely to change to a different clearly belong to the category. The extent to which atypical ex-
category to find further nonredundant solutions. At the same time, emplars (e.g., camel) are included in the category, however, re-
cognitive flexibility is thought of “as a precursor of the production quires looking at these exemplars in a different way than usual and,
of many (ideational fluency) and original responses” (De Dreu et therefore, presupposes cognitive flexibility. Following this logic,
al., 2008, p. 740). Following this reasoning, enhanced fluency cognitive flexibility is captured by goodness-of-fit ratings for
could be contaminated by the increased accessibility of many atypical exemplars.
different categories rather than vice versa. These two possibilities To manipulate trust and distrust, we used the subliminal priming
imply mediational models with either fluency mediating the effect procedure from Study 1 and added a neutral control condition. We

Figure 1. Flexibility mediates the effect of type of priming on originality (Study 2). Paths are labeled with
standardized regression coefficients. Simultaneous regression coefficients are given in parentheses.
ⴱ
p ⬍ .05. ⴱⴱ p ⬍ .01. ⴱⴱⴱ p ⬍ .001.
1270 MAYER AND MUSSWEILER
expected participants primed with distrust to give higher typicality
ratings to atypical exemplars than participants in the trust and
control conditions, thus exhibiting increased cognitive flexibility.
Methods
Design and participants. Study 3 is based on a 3 (type of
priming: distrust vs. control vs. trust) ⫻ 2 (typicality: high vs. low)
mixed model design, with the last factor being manipulated within
participants. We recruited 56 students (40 female, 16 male) as
participants and randomly assigned them to one of the three
priming conditions. Participants were approached at a campus of
the University of Cologne and offered a chocolate bar as compen-
sation for participation in two short experimental tasks.
Materials and procedure. On arrival in the lab, participants
were led to separate cubicles and asked to sit down in front of a
60-Hz computer monitor. After giving their consent, participants
read that the study was about categorization and that they would
work on two different categorization tasks consecutively. The first
task was labeled as a task requiring categorization of whether or
not a letter string was a proper German word. The task consisted
of the subliminal priming procedure described in Study 1, com-
plemented by a control condition.2 In the control condition, the
fixation string (XWXXWWXXWX) was substituted for the prime
word.
After finishing the priming task, participants proceeded to the
category inclusion task (adapted from Friedman & Förster, 2000;
cf. Isen & Daubman, 1984; Rosch, 1975). In this task, four
category names (Fortbewegungsmittel [vehicle], Möbel [furni-
ture], Gemüse [vegetable], Kleidung [clothing]) were presented
followed by 12 exemplars of each. Categories were presented in
random order. The 12 exemplars within each category were pre-
sented one at a time in a fixed order and differed in typicality from
high (e.g., car for the vehicle category) through moderate (wheel-
chair) to low (camel). Participants indicated for each exemplar
how typical it was for the named category on a 10-point scale
ranging from 0 (atypical) to 9 (typical).
A final questionnaire asked for demographic data and tested for
awareness of the true purpose of the study as well as for awareness
of the prime word in the priming procedure. None of the partici-
pants indicated awareness of the true purpose of the study or of the
prime word. At the end, participants were thanked and handed
their compensation.
Measures and dependent variables. Typicality ratings in
the category inclusion task constituted the critical dependent mea-
sure. In a first step, we defined the typicality of each exemplar with
the help of the data provided by participants in the control group.
To that end, we subjected each exemplar to a t test probing for
deviation from its respective category mean. Exemplars that
either significantly exceeded their category mean (strong ex-
emplars) or did not significantly differ from their category
mean (moderate exemplars) were averaged across the four
categories and formed the high typicality index (Cronbach’s
␣ ⫽ .90). Exemplars with means significantly lower than their
respective category mean were categorized as weak exemplars
and averaged across the four categories to form the low typi-
cality index (Cronbach’s ␣ ⫽ .80).
Results and Discussion
We hypothesized that distrust induces cognitive flexibility rep-
resented by increased category inclusiveness. We therefore ex-
pected (a) participants in the distrust condition to rate the weak
exemplars as more typical than participants in the other two
conditions and (b) ratings for the remaining exemplars (high
typicality index) not to differ for the distrust condition compared
with the other two conditions. To test these predictions, we sub-
jected the typicality ratings to a 3 (type of priming: distrust vs.
control vs. trust) ⫻ 2 (typicality: high vs. low) repeated measures
ANOVA with typicality as a within-subjects factor. To test our
specific hypothesis that participants primed with distrust showed
increased category inclusiveness (i.e., higher typicality ratings for
the weak exemplars compared with the other two conditions) we
calculated the respective planned contrast, as well.
Confirming the nature of the typicality indices, a significant
main effect emerged for typicality, F(1, 53) ⫽ 1,065.67, p ⬍ .001:
Independent of type of priming, participants rated high-typicality
exemplars (combined M ⫽ 7.56, SD ⫽ 1.00) as more typical than
low-typicality exemplars (combined M ⫽ 2.83, SD ⫽ 1.27). The
main effect for type of priming was not significant, F(2, 53) ⫽
1.98, p ⫽ .15. In line with our hypotheses, we obtained a signif-
icant interaction effect, F(2, 53) ⫽ 3.36, p ⫽ .04, ␩2p ⫽ .11,
indicating that high- and low-typicality indices were differentially
affected by the priming conditions (see Figure 2).
For high-typicality exemplars, ratings in the distrust condition
(M ⫽ 7.57, SD ⫽ 1.11) did not differ significantly from those in
the control condition (M ⫽ 7.37, SD ⫽ 1.12), in the trust condition
(M ⫽ 7.71, SD ⫽ 0.79), or in the control and trust conditions
combined (M ⫽ 7.56, SD ⫽ 0.96), all |t|s ⬍ 1. As expected,
however, for exemplars low in typicality, participants primed with
distrust exhibited higher ratings (M ⫽ 3.35, SD ⫽ 1.41) than
participants in the control and trust conditions (combined M ⫽
2.56, SD ⫽ 1.13), t(53) ⫽ 2.37, p ⫽ .01, Cohen’s d ⫽ 0.66. Simple
effect analyses revealed that ratings in the distrust condition tended
to be higher than those in the trust condition (M ⫽ 2.80, SD ⫽
1.07), t(53) ⫽ 1.41, p ⫽ .08, Cohen’s d ⫽ 0.45, and were
significantly increased compared with ratings in the control con-
dition (M ⫽ 2.27, SD ⫽ 1.16), t(53) ⫽ 2.65, p ⫽ .005, Cohen’s
d ⫽ 0.86.
In accordance with prior interpretations of the category inclu-
sion task, these results support the idea of increased cognitive
flexibility under distrust. One may, however, still argue that in-
creased category inclusiveness of exemplars low in typicality is
brought on by increased persistence (De Dreu et al., 2008). By
persevering and systematically working through possibilities, par-
ticipants may end up concluding that weak exemplars do not fit the
category so badly after all. If perseverance and careful systematic
processing were responsible for increased category inclusiveness
in the distrust condition, this should be accompanied by increased
response latencies (cf. De Dreu et al., 2008). To test for this
possibility, we averaged reaction times for exemplars high in
typicality and for exemplars low in typicality, respectively. Reac-
tion times were subjected to the same ANOVA as the typicality
2
Because of the different monitor refresh rate, prime word presen-
tation was 16 ms. Presentation time was, therefore, slightly longer than
in Study 1.
 WHEN DISTRUST ENHANCES CREATIVITY
Figure 2. Mean typicality ratings as a function of type of priming
condition (between factor) and typicality of exemplars (within factor) in
Study 3. Error bars represent one standard error.
ratings. This analysis revealed only a significant main effect for
typicality, F(1, 53) ⫽ 14.09, p ⬍ .001, indicating that, independent
of type of priming, participants took longer to rate exemplars low
in typicality than they took to rate exemplars high in typicality (all
other Fs ⬍ 1.27, all other ps ⬎ .29). Inspection of the means
further revealed that, descriptively, mean response latencies in the
distrust condition were lowest among the three priming conditions
for both high-typicality exemplars and low-typicality exemplars.
Therefore, it seems unlikely that persistence was responsible for
increased category inclusion in the distrust condition.
In conclusion, we consider the results of Study 3 as evidence for
increased category inclusiveness and therefore as evidence for
increased cognitive flexibility in a distrust mind-set. A meta-
analytical procedure (cf. Rosenthal, 1991) further strengthens this
contention by attesting a reliable increase in measures of cognitive
flexibility in the distrust condition compared with the control
condition across Studies 2 and 3 (Z ⫽ 3.62, p ⬍ .001). Study 4 was
designed to test whether these results can be conceptually repli-
cated when looking at the third indicator of cognitive flexibility,
namely, flat associative hierarchies.
Study 4
Flat associative hierarchies are considered another manifestation
of the flexibility pathway to creativity (De Dreu et al., 2008).
Associative hierarchies differ with regard to their patterns of the
associative strength of close and remote associations (Mednick,
1962). In the case of steep associative hierarchies, close associates
(e.g., chair) to a given stimulus (e.g., table) come to mind much
more easily than relatively remote ones (e.g., leg or food; example
from Mednick, 1962, p. 223). Put differently, there is a steep
decline in associative strength from close to remote associates. In
the case of flat associative hierarchies, close and remote associates
1271
are markedly more similar in their associative strength to the
stimulus. If associative hierarchies are flat rather than steep, re-
mote associations should come to mind more easily.
To measure associative hierarchies, we used materials from a
direct/indirect priming procedure developed by Kiefer, Weisbrod,
Kern, Maier, and Spitzer (1998). In this task, participants react to
target words that are preceded by a prime word. The semantic
association of the prime–target pairs varies from proximal to
remote to unrelated. Typically, that is, under neutral conditions,
individuals show faster reactions in proximal trials than in remote
trials (e.g., Kiefer et al., 1998), a pattern consistent with steep
associative hierarchies. Increased remote semantic spread is re-
flected in relatively faster reactions to remotely associated prime–
target pairs. If reactions to proximally related trials are not af-
fected, this would further imply a diminished discrepancy in
response latencies to proximally and remotely related trials. The
latter is conceptually consistent with flat associative hierarchies
and, in turn, with increased cognitive flexibility. Building on the
data presented so far, we expected participants primed with distrust
to exhibit a pattern consistent with flat associative hierarchies, that
is, a pattern of increased remote semantic spread. As the results of
both Studies 2 and 3 were largely in line with the notion that trust
and control conditions produce similar results, we chose to contrast
the distrust condition with a trust condition only. Consistent with
the conceptualization of trust as the default (cf. Berg et al., 1995;
McKnight et al., 1998), we expected to find the typical pattern of
prolonged response latencies in remote trials in the trust condition.
For reasons of compatibility with the direct/indirect priming pro-
cedure, we used a slightly modified version of our subliminal
trust/distrust priming (cf. Studies 1 and 3).
Methods
Design and participants. Study 4 is based on a 2 (type of
priming: distrust vs. trust) ⫻ 2 (semantic distance: proximal vs.
remote) mixed design, with the last factor manipulated within
participants. We recruited 48 individuals on a University of Co-
logne campus as participants and randomly assigned them to one
of the two priming conditions. Participants were offered 5€ (ap-
proximately $6.50 at the time) as compensation. One participant
was excluded because of data collection irregularities, leaving a
final sample of 47 participants (36 female, 11 male).
Materials and procedure. On arrival in the lab, participants
were led to separate cubicles and asked to sit down in front of an
85-Hz computer monitor. After giving their consent, participants
worked on a lexical decision task. This task combined our sub-
liminal priming procedure with a direct/indirect priming paradigm.
The material for the direct/indirect priming paradigm (Kiefer et al.,
1998) consisted of a total of 270 word pairs. Each pair consisted of
a prime word and a target letter string. Primes were all proper
German words and so were half of the target letter strings. The
remaining target letter strings constituted meaningless yet pro-
nounceable nonwords. Pairs with proper German words as targets
differed with respect to the semantic relatedness of prime and
target: proximal (e.g., König–Krone [king– crown]), remote (e.g.,
Bier–Traube [beer– grape]), and unrelated (e.g., Bleistift–Uhr
[pencil– clock]), with each category being represented by 45 pairs.
Participants worked through 132 trials drawn randomly from the
pool of 270 prime–target pairs. This procedure yields, on average,
1272 MAYER AND MUSSWEILER

22 trials per category of semantic distance (proximal, remote,

unrelated). Each trial began with the presentation of a fixation
string (XQFBZRMQWGBX). After 494 ms, the fixation string was
briefly (12 ms, subliminal) replaced, by either the word misstrauen
(in the distrust condition) or the word vertrauen (in the trust
condition). The same fixation string was then presented again for
another 494 ms. Next, the prime word was presented for 153 ms
(supraliminal). This prime word was replaced by a blank screen
(94 ms) that was, in turn, followed by the target letter string. The
target remained on the screen until participants indicated whether
or not it constituted a proper German word by pressing the appro-
priate key (for procedures, see Kiefer et al., 1998; Topolinski &
Deutsch, 2009).
A final questionnaire asked for demographic data and tested for
awareness of the true purpose of the study as well as for awareness
of the prime word in the subliminal priming procedure. None of
the participants indicated awareness of the true purpose of the
study or of the prime word. At the end, participants were thanked
and handed their compensation.
Measures and dependent variables. For our purposes, only
trials with real German words as targets were of interest (cf. Kiefer
Figure 3. Mean reaction times as a function of type of priming and
et al., 1998; Topolinski & Deutsch, 2009). From these trials, in a semantic distance in Study 4. Error bars represent one standard error.
first step, incorrect responses (1.82% of the data) and response
latencies exceeding 2,000 ms (0.46% of the data) were dropped.
Additionally, response latencies longer than twice the individual Subjecting the corresponding error rates to the same ANOVA did
mean per trial category (proximal, remote, unrelated) were not not yield any significant effects (all Fs ⬍ 2.3, all ps ⬎ .13).
considered (1.01% of the data; cf. Kiefer et al., 1998). The re- These results shed additional light on the cognitive mechanisms
maining response latencies were averaged per trial category. that underlie the creativity-enhancing effect of distrust: Distrust-
primed individuals exhibited increased remote semantic spread
Results and Discussion and thereby an activation pattern in accordance with flat associa-
tive hierarchies. Flat associative hierarchies, in turn, are considered
Responses to proximally and remotely associated prime–target a manifestation of the flexibility pathway to creativity (De Dreu et
pairs constitute our main variable of interest; responses to unre- al., 2008). We, therefore, interpret this finding as yet another
lated prime–target pairs served as a baseline. For baseline trials, indicator of increased cognitive flexibility in a distrust mind-set.
mean response latencies in the distrust condition (M ⫽ 638 ms, A meta-analytic procedure (Rosenthal, 1991) corroborates the
SD ⫽ 141 ms) and in the trust condition (M ⫽ 657 ms, SD ⫽ 153 increase in cognitive flexibility in a distrust mind-set: Across
ms) did not differ significantly and neither did the error rates, all Studies 1 (private condition only) through 4, distrust priming
Fs ⬍ 1. yielded reliably increased flexibility indices compared with trust
In the main analysis, we subjected mean response latencies to priming (Z ⫽ 2.63, p ⫽ .004). A parallel analysis comparing the
proximally and remotely related prime–target pairs to a 2 (type of distrust condition to both control conditions (i.e., to the trust
priming: distrust vs. trust) ⫻ 2 (semantic distance: proximal vs. conditions in Studies 1 and 4; to the combined trust and control
remote) repeated measures ANOVA with semantic distance as a conditions in Studies 2 and 3) also reveals significantly increased
within-subjects factor. Inspection of the means in Figure 3 reveals cognitive flexibility after distrust priming (Z ⫽ 2.89, p ⫽ .002).
a pattern of means consistent with our predictions. In the trust
condition, the typical pattern indicating steep associative hierar-
General Discussion
chies emerged: Response latencies for remotely related trials (M ⫽
610 ms, SD ⫽ 131 ms) were prolonged compared with proximally At first sight, it might seem plausible to assume that distrust
related trials (M ⫽ 574 ms, SD ⫽ 118 ms), F(1, 45) ⫽ 8.93, p ⫽ thwarts creativity—an intuition that is also consistent with empir-
.005, ␩2p ⫽ .17, for the simple effect. In line with our predictions, ical evidence so far. Contained, however, in recent findings on the
participants primed with distrust, however, did not show the typ- cognitive concomitants of distrust (Schul et al., 1996, 2004, 2008)
ical pattern of prolonged response latencies in remotely related is the provocative implication that distrust may foster rather than
trials. Rather, they were as fast in reacting to remotely related hinder creativity. To solve the thereby emerging contradiction, we
prime–target pairs (M ⫽ 570 ms, SD ⫽ 92 ms) as to proximally proposed that whether creative thoughts remain private or are
related ones (M ⫽ 574 ms, SD ⫽ 126 ms), F ⬍ 1, ns, for the simple made public constitutes a critical moderating variable: In public,
effect, indicating flat associative hierarchies. This pattern of means the social consequences of distrust are likely to prove detrimental
was borne out in a statistically reliable two-way interaction, F(1, to creativity. In private, however, these social consequences are
45) ⫽ 5.71, p ⫽ .02, ␩2p ⫽ .11. In addition, a trend for a main effect incapacitated, so that the beneficial cognitive consequences of
for semantic distance emerged, F(1, 45) ⫽ 3.53, p ⫽ .07, ␩2p ⫽ .07. distrust may develop their creativity-enhancing potential. In line
The main effect for type of priming was not significant, F ⬍ 1. with this reasoning, our results replicated earlier findings of a
 WHEN DISTRUST ENHANCES CREATIVITY
detrimental influence of distrust on creativity when ideas gener-
ated were allegedly made public to another participant. When ideas
remained private, however, inducing a distrust mind-set proved
beneficial to creative performance. The present findings also shed
light on the cognitive mechanism underlying this creativity-
enhancing effect of distrust. Our studies provided sound evidence
that a distrust mind-set is accompanied by increased cognitive
flexibility in terms of category diversity, category inclusiveness,
and flat associative hierarchies. Cognitive flexibility also mediated
the increase in originality after distrust priming under conditions of
true privacy. The counterintuitive finding of enhanced creativity in
a distrust mind-set, its specific association with cognitive flexibil-
ity, and the moderating influence of public versus private contexts
have important implications for research on both creativity and
distrust. Before discussing these implications in detail, together
with avenues for future research, we first address the question of
the broader applicability of our findings as well as potential
alternative explanations.
Applicability of the Presented Findings and
Alternative Explanations
The methods used to elicit distrust in our studies may seem far
from how one usually pictures the natural emergence of distrust.
Yet, we propose that our priming manipulations closely mirror
how distrust emerges in a wide array of real world settings.
Specifically, whenever there is no prior information available
about the trustworthiness of a person or a source of information,
subtle cues may easily trigger a distrust mind-set. This follows
from the nature of deceptive attempts, in which deceivers often go
to great lengths to hide their true motives. Therefore, we can
expect deceptive intentions to show up primarily in unintentionally
disclosed cues that are easily overlooked and often ambiguous
with respect to their informational value (Anderson, DePaulo,
Ansfield, Tickle, & Green, 1999; DePaulo et al., 2003). The
activation of a distrust mind-set triggered by such subtle cues—
and not necessarily accompanied by conscious awareness—applies
to many real world settings.
As has been shown for distrust (Friesen & Sinclair, 2011;
Rotenberg et al., 2010; Schul et al., 1996, 2004, 2008) but also for
many other variables (e.g., Corcoran, Hundhammer, & Muss-
weiler, 2009; De Dreu & Nijstad, 2008; Markman et al., 2007),
once elicited, the respective mind-set can linger and influence
behavior unrelated to the eliciting circumstances. It has further
been shown that mind-sets can reliably be instantiated by primes
semantically related to the targeted concept (e.g., Damisch, Sto-
berock, & Mussweiler, 2010; Galinsky et al., 2008; Mussweiler &
Förster, 2000). Taking these considerations into account, we feel
confident that the effects observed in our experimental setting are
applicable to a wide array of naturally occurring instances.
Apart from the question of the ecological validity of our meth-
ods to elicit a distrust mind-set, one might consider qualities of the
mind-set other than the targeted cognitive ones as influential to our
results. As mentioned earlier, a distrust mind-set is being associ-
ated with negative, activating, avoidance-oriented qualities (cf.
Mussweiler & Burgmer, 2011; Schul et al., 2008). Hedonic tone
and activation (cf. Baas, De Dreu, & Nijstad, 2008), as well as
action orientation in terms of approach versus avoidance (e.g.,
Friedman & Förster, 2000), in turn, have been of prominent
1273
interest in creativity research. Respective differences are therefore
often discussed as alternative explanations in creativity research.
With regard to hedonic tone and arousal, it is important to keep
in mind that our manipulations were subtle and specifically de-
signed not to produce substantial differences in hedonic tone and
arousal. Consistently, additional tests of our priming procedures
revealed no significant differences between the distrust condition,
the trust condition, and the neutral condition in explicit measures
of mood and arousal.3 Second, even if our priming procedures had
changed affective states to a certain degree not captured by explicit
measures, the data are still unlikely to be the result of these
variations alone. If anything, affective qualities relevant to distrust
yield creativity by increased persistence, not by increased cogni-
tive flexibility (Baas et al., 2008; De Dreu et al., 2008). However,
our data reveal increased cognitive flexibility in a distrust mind-
set, whereas measures of persistence remained unaffected.
With regard to the qualities of distrust in terms of action orien-
tations, avoidance orientations have mostly been shown to be
negatively related to creativity (e.g., Friedman & Förster, 2000).
Notably, however, we demonstrate an increase in creativity after
distrust priming. Recent research further suggests that the specific
combination of hedonic tone, arousal, and action orientation of a
particular state plays a critical role for creativity (Baas et al., 2008;
Baas, De Dreu, & Nijstad, 2011). For other states that share the
negative, activating, avoidance-related affective properties of dis-
trust (e.g., anxiety), evidence is somewhat mixed with regard to
their overall influence on creativity and yet quite consistent with
regard to their negative influence on cognitive flexibility (Baas et
al., 2008, 2011). However, as mentioned before, our data show
enhanced cognitive flexibility in a distrust mind-set.
In sum, it seems unlikely that our findings are the result of
differences in affective states, approach versus avoidance orienta-
tions, or a combination of both.
Theoretical Implications and Avenues for Future
Research
Our findings extend previous research on distrust and creativity
in several important ways. First, our work contributes to the
research on factors that foster or hinder creativity. Creativity is
predominantly perceived as a positive and desirable quality that
produces positive and socially desirable outcomes (Mumford &
Gustafson, 1988). By showing that a negatively valued state like
distrust has the potential to foster creativity, our research contrib-
utes to a more complex understanding of creativity. Our findings
therefore add to the growing body of research investigating the
3
We recruited 47 people at a University of Cologne campus. They
worked on the subliminal priming procedure (cf. Study 3) and afterward
answered three questions assessing hedonic tone and three questions as-
sessing arousal. Separate three-level ANOVAs for the hedonic tone and
arousal scales revealed no significant differences between the three con-
ditions (distrust, control, trust), all Fs(2, 46) ⬍ 1.5, all ps ⬎ .23. A different
(online) sample of 43 participants first worked on the scrambled sentences
task (cf. Study 2) and afterward filled in a German questionnaire assessing
mood, alertness, and calmness (Steyer, Schwenkmezger, Notz, & Eid,
1997). Separate three-level ANOVAs for all three scales again revealed no
significant differences between the three conditions (distrust, control,
trust), all Fs ⬍ 1 (cf. Mussweiler & Burgmer, 2011, for the latter data set).
1274 MAYER AND MUSSWEILER
creativity-enhancing potential of unpleasant or threatening states
(e.g., Baas et al., 2008; De Dreu et al., 2008; De Dreu & Nijstad,
2008).
More specifically, we identified distrust as a literally two-faced
predictor with the potential of both fostering and hindering cre-
ative performance. Thereby, our research makes a significant con-
tribution to creativity research beyond earlier research on, for
example, counterfactual thinking (Galinsky & Moskowitz, 2000;
Kray, Galinsky, & Wong, 2006; Markman et al., 2007): Although
both the distrust and counterfactual mind-sets entail a basic ten-
dency to flexibly consider alternatives, they do not share the
differential effects of fostering creative performance in private
while diminishing it in public. These conditional effects are char-
acteristic of a distrust mind-set with its diverging cognitive versus
social consequences. Other than in a distrust mind-set, a counter-
factual mind-set has been shown to foster rather than hinder the
sharing of unique information (Galinsky & Kray, 2004).
Future research will have to clarify whether the diverging ef-
fects on creativity that we observed for distrust may also apply to
other trust/distrust-related variables. Extrapolating from our re-
search, it may well be the case that some of the positive effects
observed for trust-related variables (e.g., Bechtoldt et al., 2010;
Carmeli et al., 2010) are restricted to group settings, that is, to
creative performance in public. It may well be that the presence of
others provides the activating component (e.g., Mullen, Bryant, &
Driskell, 1997) that has been shown to be critical for fueling
cognitive flexibility and creativity in positive affective states (De
Dreu et al., 2008). Our data would be largely consistent with this
contention: Within the trust condition of Study 1, we observed
increased creativity (i.e., originality) in the public context com-
pared with the private context (see Table 1).
Although the creativity-enhancing effects of trust-related vari-
ables may at times be restricted to public settings, the complete
opposite seems likely for variables that share distrust’s context
sensitivity. Creative performance in situations associated with
threat coming from the social environment, more generally, may
be observed exclusively in private. Future research may look into
whether, for example, findings of targeted creativity in a conflict
mind-set (De Dreu & Nijstad, 2008) disappear if the creativity task
involves public sharing of ideas. Considering a potentially mod-
erating influence of whether creative performance is carried out
publicly or privately may also be helpful in investigating so-called
malevolent creativity (Cropley, Kaufman, & Cropley, 2008), that
is, creativity that yields creative but socially undesirable end
products. Even though we usually do not think of lying (Walczyk
et al., 2008) or terrorist attacks (cf. Cropley et al., 2008) in terms
of creativity, they often meet the criteria by being novel and
appropriate means to a specific end. Because instances of malev-
olent creativity are subject to social disapproval, creativity in this
realm is likely to show up only in private.
Second, our research adds to the understanding of the cognitive
consequences of distrust. We show that distrust is accompanied not
only by increased creativity but also by cognitive flexibility man-
ifested in category diversity, category inclusiveness, and remote
semantic spread. This is of considerable significance, because the
affective–motivational qualities of distrust have quite consistently
been associated with a decrease rather than an increase in broad
and flexible processing (Baas et al., 2008, 2011; De Dreu et al.,
2008; Friedman & Förster, 2000; Isen & Daubman, 1984). With
regard to the specific requirements in situations of impending
deceit, on the other hand, cognitive flexibility seems highly func-
tional and, therefore, quite likely to have evolved as an integral
part of the distrust mind-set: Because the deceiver will carefully
hide his or her true motives, the targeted victim is unlikely to
quickly and unambiguously figure out whether ulterior motives
exist and, if so, what they might be. Consequently, it should be
more adaptive not to commit to a particular alternative and explore
it in depth early on, because this particular alternative might be
wrong. Considering many and quite different perspectives, in
contrast, increases the likelihood of eventually figuring out the
carefully hidden truth. Only when clear-cut evidence for a partic-
ular alternative is presented or, alternatively, when several ambig-
uous pieces of evidence add up to implicate a particular alternative
does it seem safe to pursue this idea in more depth. The emergence
of increased cognitive flexibility in a distrust mind-set— despite
the contradicting effects of the accompanying affective–
motivational qualities—thus seems easily attributable to the func-
tional aspect of protecting against impending deceit. The idea that
functional aspects of a particular state may predominate other
tendencies associated with a particular state has, for example, been
shown in the research on social conflict and creativity (De Dreu &
Nijstad, 2008): Even though conflict has traditionally been related
to threat rigidity and low levels of cognitive flexibility (Carnevale
& Probst, 1998), cognitive flexibility can still be fueled if it proves
functional for handling the conflict at hand (De Dreu & Nijstad,
2008).
In much the same way, distrust may—through the mechanism of
cognitive flexibility— have beneficial effects. Suspicious spirits
have flexible and—as a consequence— creative minds.
References
Abelson, R. P. (1995). Statistics as principled argument. Hillsdale, NJ:
Erlbaum.
Amabile, T. M., & Gryskiewicz, N. D. (1989). The Creative Environment
Scales: Work Environment Inventory. Creativity Research Journal, 2,
231–253. doi:10.1080/10400418909534321
Anderson, D. E., DePaulo, B. M., Ansfield, M. E., Tickle, J. J., & Green,
E. (1999). Beliefs about cues to deception: Mindless stereotypes or
untapped wisdom? Journal of Nonverbal Behavior, 23, 67– 89. doi:
10.1023/A:1021387326192
Baas, M., De Dreu, C. K. W., & Nijstad, B. A. (2008). A meta-analysis of
25 years of mood– creativity research: Hedonic tone, activation, or
regulatory focus? Psychological Bulletin, 134, 779 – 806. doi:10.1037/
a0012815
Baas, M., De Dreu, C. K. W., & Nijstad, B. A. (2011). When prevention
promotes creativity: The role of mood, regulatory focus, and regulatory
closure. Journal of Personality and Social Psychology, 100, 794 – 809.
doi:10.1037/a0022981
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.
Bechtoldt, M. N., De Dreu, C. K. W., Nijstad, B. A., & Choi, H.-S. (2010).
Motivated information processing, social tuning, and group creativity.
Journal of Personality and Social Psychology, 99, 622– 637. doi:
10.1037/a0019386
Beersma, B., & De Dreu, C. K. W. (2005). Conflict’s consequences:
Effects of social motives on postnegotiation creative and convergent
 WHEN DISTRUST ENHANCES CREATIVITY
group functioning and performance. Journal of Personality and Social
Psychology, 89, 358 –374. doi:10.1037/0022-3514.89.3.358
Berg, J., Dickhaut, J., & McCabe, K. (1995). Trust, reciprocity, and social
history. Games and Economic Behavior, 10, 122–142. doi:10.1006/
game.1995.1027
Camacho, L. M., & Paulus, P. B. (1995). The role of social anxiousness in
group brainstorming. Journal of Personality and Social Psychology, 68,
1071–1080. doi:10.1037/0022-3514.68.6.1071
Carmeli, A., Reiter-Palmon, R., & Ziv, E. (2010). Inclusive leadership and
employee involvement in creative tasks in the workplace: The mediating
role of psychological safety. Creativity Research Journal, 22, 250 –260.
doi:10.1080/10400419.2010.504654
Carnevale, P. J., & Probst, T. M. (1998). Social values and social conflict
in creative problem solving and categorization. Journal of Personality
and Social Psychology, 74, 1300 –1309. doi:10.1037/0022-
3514.74.5.1300
Chiappe, D., Brown, A., Dow, B., Koontz, J., Rodriguez, M., & Mc-
Culloch, K. (2004). Cheaters are looked at longer and remembered better
than cooperators in social exchange situations. Evolutionary Psychology,
2, 108 –120. Retrieved from http://www.epjournal.net/filestore/
ep021081202.pdf
Corcoran, K., Hundhammer, T., & Mussweiler, T. (2009). A tool for
thought! When comparative thinking reduces stereotyping effects. Jour-
nal of Experimental Social Psychology, 45, 1008 –1011. doi:10.1016/
j.jesp.2009.04.015
Cosmides, L., & Tooby, J. (2005). Neurocognitive adaptations designed for
social exchange. In D. M. Buss (Ed.), The handbook of evolutionary
psychology (pp. 584 – 627). Hoboken, NJ: Wiley.
Cropley, D. H., Kaufman, J. C., & Cropley, A. J. (2008). Malevolent
creativity: A functional model of creativity in terrorism and crime.
Creativity Research Journal, 20, 105–115. doi:10.1080/
10400410802059424
Dakhli, M., & De Clercq, D. (2004). Human capital, social capital, and
innovation: A multi-country study. Entrepreneurship & Regional De-
velopment, 16, 107–128. doi:10.1080/08985620410001677835
Damisch, L., Stoberock, B., & Mussweiler, T. (2010). Keep your fingers
crossed! How superstition improves performance. Psychological Sci-
ence, 21, 1014 –1020. doi:10.1177/0956797610372631
De Cremer, D. (1999). Trust and fear of exploitation in a public goods
dilemma. Current Psychology: A Journal for Diverse Perspectives on
Diverse Psychological Issues, 18, 153–163. doi:10.1007/s12144-999-
1024-0
De Dreu, C. K. W., Baas, M., & Nijstad, B. A. (2008). Hedonic tone and
activation level in the mood– creativity link: Toward a dual pathway to
creativity model. Journal of Personality and Social Psychology, 94,
739 –756. doi:10.1037/0022-3514.94.5.739
De Dreu, C. K. W., & Nijstad, B. A. (2008). Mental set and creative
thought in social conflict: Threat rigidity versus motivated focus. Jour-
nal of Personality and Social Psychology, 95, 648 – 661. doi:10.1037/
0022-3514.95.3.648
DePaulo, B. M., & Kashy, D. A. (1998). Everyday lies in close and casual
relationships. Journal of Personality and Social Psychology, 74, 63–79.
doi:10.1037/0022-3514.74.1.63
DePaulo, B. M., Kashy, D. A., Kirkendol, S. E., Wyer, M. M., & Epstein,
J. A. (1996). Lying in everyday life. Journal of Personality and Social
Psychology, 70, 979 –995. doi:10.1037/0022-3514.70.5.979
DePaulo, B. M., Lindsay, J. J., Malone, B. E., Muhlenbruck, L., Charlton,
K., & Cooper, H. (2003). Cues to deception. Psychological Bulletin,
129, 74 –118. doi:10.1037/0033-2909.129.1.74
Diehl, M., & Stroebe, W. (1987). Productivity loss in brainstorming
groups: Toward the solution of a riddle. Journal of Personality and
Social Psychology, 53, 497–509. doi:10.1037/0022-3514.53.3.497
Dijksterhuis, A., Aarts, H., Bargh, J. A., & van Knippenberg, A. (2000). On
the relation between associative strength and automatic behavior. Jour-
1275
nal of Experimental Social Psychology, 36, 531–544. doi:10.1006/
jesp.2000.1427
Edmondson, A. C. (2004). Psychological safety, trust, and learning in
organizations: A group-level lens. In R. M. Kramer & K. S. Cook (Eds.),
Trust and distrust in organizations: Dilemmas and approaches (pp.
239 –272). New York, NY: Russell Sage Foundation.
Ekvall, G. (1996). Organizational climate for creativity and innovation.
European Journal of Work and Organizational Psychology, 5, 105–123.
doi:10.1080/13594329608414845
Ekvall, G., & Ryhammar, L. (1999). The creative climate: Its determinants
and effects at a Swedish university. Creativity Research Journal, 12,
303–310. doi:10.1207/s15326934crj1204_8
Epstude, K., & Roese, N. J. (2008). The functional theory of counterfactual
thinking. Personality and Social Psychology Review, 12, 168 –192.
doi:10.1177/1088868308316091
Fein, S., Hilton, J. L., & Miller, D. T. (1990). Suspicion of ulterior
motivation and the correspondence bias. Journal of Personality and
Social Psychology, 58, 753–764. doi:10.1037/0022-3514.58.5.753
Fetchenhauer, D., & Dunning, D. (2009). Do people trust too much or too
little? Journal of Economic Psychology, 30, 263–276. doi:10.1016/
j.joep.2008.04.006
Friedman, R. S., & Förster, J. (2000). The effects of approach and avoid-
ance motor actions on the elements of creative insight. Journal of
Personality and Social Psychology, 79, 477– 492. doi:10.1037/0022-
3514.79.4.477
Friesen, J., & Sinclair, L. (2011). Distrust and simultaneous activation of
multiple categories. Social Psychological and Personality Science, 2,
112–118. doi:10.1177/1948550610382666
Galinsky, A. D., & Kray, L. J. (2004). From thinking about what might
have been to sharing what we know: The effects of counterfactual
mind-sets on information sharing in groups. Journal of Experimental
Social Psychology, 40, 606 – 618. doi:10.1016/j.jesp.2003.11.005
Galinsky, A. D., Magee, J. C., Gruenfeld, D. H., & Whitson, J. A. (2008).
Power reduces the press of the situation: Implications for creativity,
conformity, and dissonance. Journal of Personality and Social Psychol-
ogy, 95, 1450 –1466. doi:10.1037/a0012633
Galinsky, A. D., & Moskowitz, G. B. (2000). Counterfactuals as behavioral
primes: Priming the simulation heuristic and consideration of alterna-
tives. Journal of Experimental Social Psychology, 36, 384 – 409. doi:
10.1006/jesp.1999.1409
George, J. M., & Zhou, J. (2007). Dual tuning in a supportive context: Joint
contributions of positive mood, negative mood, and supervisory behav-
iors to employee creativity. Academy of Management Journal, 50, 605–
622. doi:10.5465/AMJ.2007.25525934
Gilbert, D. T. (1991). How mental systems believe. American Psycholo-
gist, 46, 107–119. doi:10.1037/0003-066X.46.2.107
Gilhooly, K. J., Fioratou, E., Anthony, S. H., & Wynn, V. (2007). Diver-
gent thinking: Strategies and executive involvement in generating novel
uses for familiar objects. British Journal of Psychology, 98, 611– 625.
doi:10.1111/j.2044-8295.2007.tb00467.x
Glover, J. A. (1977). Risky shift and creativity. Social Behavior and
Personality, 5, 317–320. doi:10.2224/sbp.1977.5.2.317
Guilford, J. P. (1967). The nature of human intelligence. New York, NY:
McGraw-Hill.
Hennessey, B. A. (2003). The social psychology of creativity. Scandina-
vian Journal of Educational Research, 47, 253–271. doi:10.1080/
00313830308601
Hennessey, B. A., & Amabile, T. M. (2010). Creativity. Annual Review of
Psychology, 61, 569 –598. doi:10.1146/annurev.psych.093008.100416
Hilton, J. L., Fein, S., & Miller, D. T. (1993). Suspicion and dispositional
inference. Personality and Social Psychology Bulletin, 19, 501–512.
doi:10.1177/0146167293195003
Hocevar, D. (1979). Ideational fluency as a confounding factor in the
1276 MAYER AND MUSSWEILER
measurement of originality. Journal of Educational Psychology, 71,
191–196. doi:10.1037/0022-0663.71.2.191
Hunter, S. T., Bedell, K. E., & Mumford, M. D. (2007). Climate for
creativity: A quantitative review. Creativity Research Journal, 19, 69 –
90. doi:10.1080/10400410709336883
Isen, A. M., & Daubman, K. A. (1984). The influence of affect on
categorization. Journal of Personality and Social Psychology, 47, 1206 –
1217. doi:10.1037/0022-3514.47.6.1206
Kiefer, M., Weisbrod, M., Kern, I., Maier, S., & Spitzer, M. (1998). Right
hemisphere activation during indirect semantic priming: Evidence from
event-related potentials. Brain and Language, 64, 377– 408. doi:
10.1006/brln.1998.1979
Klimoski, R. J., & Karol, B. L. (1976). The impact of trust on creative
problem solving groups. Journal of Applied Psychology, 61, 630 – 633.
doi:10.1037/0021-9010.61.5.630
Kramer, R. M. (1998). Paranoid cognition in social systems: Thinking and
acting in the shadow of doubt. Personality and Social Psychology
Review, 2, 251–275. doi:10.1207/s15327957pspr0204_3
Kray, L. J., Galinsky, A. D., & Wong, E. M. (2006). Thinking within the
box: The relational processing style elicited by counterfactual mind-sets.
Journal of Personality and Social Psychology, 91, 33– 48. doi:10.1037/
0022-3514.91.1.33
Marchand, M. A. G., & Vonk, R. (2005). The process of becoming
suspicious of ulterior motives. Social Cognition, 23, 242–256. doi:
10.1521/soco.2005.23.3.242
Markman, K. D., Lindberg, M. J., Kray, L. J., & Galinsky, A. D. (2007).
Implications of counterfactual structure for creative generation and
analytical problem solving. Personality and Social Psychology Bulletin,
33, 312–324. doi:10.1177/0146167206296106
McKnight, D. H., Cummings, L. L., & Chervany, N. L. (1998). Initial trust
formation in new organizational relationships. Academy of Management
Review, 23, 473– 490. doi:10.2307/259290
Mednick, S. A. (1962). The associative basis of the creative process.
Psychological Review, 69, 220 –232. doi:10.1037/h0048850
Mullen, B., Bryant, B., & Driskell, J. E. (1997). Presence of others and
arousal: An integration. Group Dynamics: Theory, Research, and Prac-
tice, 1, 52– 64. doi:10.1037/1089-2699.1.1.52
Mumford, M. D., & Gustafson, S. B. (1988). Creativity syndrome: Inte-
gration, application, and innovation. Psychological Bulletin, 103, 27– 43.
doi:10.1037/0033-2909.103.1.27
Murray, N., Sujan, H., Hirt, E. R., & Sujan, M. (1990). The influence of
mood on categorization: A cognitive flexibility interpretation. Journal of
Personality and Social Psychology, 59, 411– 425. doi:10.1037/0022-
3514.59.3.411
Mussweiler, T., & Burgmer, P. (2011). Effects of trust and distrust on
approach and avoidance tendencies. Manuscript in preparation.
Mussweiler, T., & Förster, J. (2000). The sex 3 aggression link: A
perception– behavior dissociation. Journal of Personality and Social
Psychology, 79, 507–520. doi:10.1037/0022-3514.79.4.507
Nemeth, C. J., & Ormiston, M. (2007). Creative idea generation: Harmony
versus stimulation. European Journal of Social Psychology, 37, 524 –
535. doi:10.1002/ejsp.373
Nijstad, B. A., Stroebe, W., & Lodewijkx, H. F. M. (2002). Cognitive
stimulation and interference in groups: Exposure effects in an idea
generation task. Journal of Experimental Social Psychology, 38, 535–
544. doi:10.1016/S0022-1031(02)00500-0
Osborn, A. F. (1963). Applied imagination: Principles and procedures of
creative thinking. New York, NY: Scribner.
Petermann, F. (1992). Psychologie des Vertrauens [The psychology of
trust] (2nd ed.). München, Germany: Quintessenz.
Plucker, J. A., & Renzulli, J. S. (1999). Psychometric approaches to the
study of human creativity. In R. J. Sternberg (Ed.), Handbook of cre-
ativity (pp. 35– 61). New York, NY: Cambridge University Press.
Posten, A.-C., & Mussweiler, T. (2011). Priming trust. Manuscript in
preparation.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for
estimating indirect effects in simple mediation models. Behavior Re-
search Methods, Instruments, & Computers, 36, 717–731. doi:10.3758/
BF03206553
Rietzschel, E. F., De Dreu, C. K. W., & Nijstad, B. A. (2007). Personal
need for structure and creative performance: The moderating influence
of fear of invalidity. Personality and Social Psychology Bulletin, 33,
855– 866. doi:10.1177/0146167207301017
Rietzschel, E. F., Nijstad, B. A., & Stroebe, W. (2007). Relative accessi-
bility of domain knowledge and creativity: The effects of knowledge
activation on the quantity and originality of generated ideas. Journal of
Experimental Social Psychology, 43, 933–946. doi:10.1016/j
.jesp.2006.10.014
Rosch, E. (1975). Cognitive representations of semantic categories. Jour-
nal of Experimental Psychology: General, 104, 192–233. doi:10.1037/
0096-3445.104.3.192
Rosenthal, R. (1991). Meta-analytic procedures for social research. New-
bury Park, CA: Sage.
Rotenberg, K. J., Addis, N., Betts, L. R., Corrigan, A., Fox, C., Hobson, Z.,
& Boulton, M. J. (2010). The relation between trust beliefs and loneli-
ness during early childhood, middle childhood, and adulthood. Person-
ality and Social Psychology Bulletin, 36, 1086 –1100. doi:10.1177/
0146167210374957
Rule, N. O., Ambady, N., Adams, R. B., Ozono, H., Nakashima, S.,
Yoshikawa, S., & Watabe, M. (2010). Polling the face: Prediction and
consensus across cultures. Journal of Personality and Social Psychol-
ogy, 98, 1–15. doi:10.1037/a0017673
Runco, M. A. (2007). Creativity: Theories and themes: Research, devel-
opment, and practice. San Diego, CA: Academic Press.
Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative
model of organizational trust: Past, present, and future. Academy of
Management Review, 32, 344 –354. doi:10.5465/AMR.2007.24348410
Schoppe, K.-J. (1975). Verbaler Kreativitäts-Test (V-K-T). Ein Verfahren
zur Erfassung verbal-produktiver Kreativitätsmerkmale [The verbal cre-
ativity test. A procedure for measuring verbal-productive aspects of
creativity]. Göttingen, Germany: Verlag für Psychologie.
Schul, Y. (1993). When warning succeeds: The effect of warning on
success in ignoring invalid information. Journal of Experimental Social
Psychology, 29, 42– 62. doi:10.1006/jesp.1993.1003
Schul, Y., Burnstein, E., & Bardi, A. (1996). Dealing with deceptions that
are difficult to detect: Encoding and judgment as a function of preparing
to receive invalid information. Journal of Experimental Social Psychol-
ogy, 32, 228 –253. doi:10.1006/jesp.1996.0011
Schul, Y., Mayo, R., & Burnstein, E. (2004). Encoding under trust and
distrust: The spontaneous activation of incongruent cognitions. Journal
of Personality and Social Psychology, 86, 668 – 679. doi:10.1037/0022-
3514.86.5.668
Schul, Y., Mayo, R., & Burnstein, E. (2008). The value of distrust. Journal
of Experimental Social Psychology, 44, 1293–1302. doi:10.1016/
j.jesp.2008.05.003
Simonton, D. K. (1997). Creative productivity: A predictive and explana-
tory model of career trajectories and landmarks. Psychological Review,
104, 66 – 89. doi:10.1037/0033-295X.104.1.66
Simpson, J. A. (2007). Foundations of interpersonal trust. In A. W. Krug-
lanski & E. T. Higgins (Eds.), Social psychology: Handbook of basic
principles (2nd ed., pp. 587– 607). New York, NY: Guilford.
Smith, S. M., & Blankenship, S. E. (1991). Incubation and the persistence
of fixation in problem solving. American Journal of Psychology, 104,
61– 87. doi:10.2307/1422851
Srull, T. K., & Wyer, R. S. (1979). The role of category accessibility in the
interpretation of information about persons: Some determinants and
 WHEN DISTRUST ENHANCES CREATIVITY
implications. Journal of Personality and Social Psychology, 37, 1660 –
1672. doi:10.1037/0022-3514.37.10.1660
Sternberg, R. J., & Lubart, T. I. (1999). The concept of creativity: Pros-
pects and paradigms. In R. J. Sternberg (Ed.), Handbook of creativity
(pp. 3–15). New York, NY: Cambridge University Press.
Steyer, R., Schwenkmezger, P., Notz, P., & Eid, M. (1997). Der Mehrdi-
mensionale Befindlichkeitsfragebogen (MDBF) [Multidimensional
Comfort Questionnaire]. Göttingen, Germany: Hogrefe.
Taggar, S. (2002). Individual creativity and group ability to utilize indi-
vidual creative resources: A multilevel model. Academy of Management
Journal, 45, 315–330. doi:10.2307/3069349
Topolinski, S., & Deutsch, R. (2009). Phasic affective modulation of
semantic spread. Manuscript submitted for publication.
Tyler, J. M., Feldman, R. S., & Reichert, A. (2006). The price of deceptive
behavior: Disliking and lying to people who lie to us. Journal of
Experimental Social Psychology, 42, 69 –77. doi:10.1016/j
.jesp.2005.02.003
Walczyk, J. J., Runco, M. A., Tripp, S. M., & Smith, C. E. (2008). The
creativity of lying: Divergent thinking and ideational correlates of the
resolution of social dilemmas. Creativity Research Journal, 20, 328 –
342. doi:10.1080/10400410802355152
Wallach, M. A., & Kogan, N. (1965). Modes of thinking in young children:
1277
A study of the creativity–intelligence distinction. Oxford, United King-
dom: Holt, Rinehart, & Winston.
West, M. A. (2002). Sparkling fountains or stagnant ponds: An integrative
model of creativity and innovation implementation in work groups.
Applied Psychology: An International Review, 51, 355–387. doi:
10.1111/1464-0597.00951
Ybarra, O., Winkielman, P., Yeh, I., Burnstein, E., & Kavanagh, L. (2011).
Friends (and sometimes enemies) with cognitive benefits: What types of
social interactions boost executive functioning? Social Psychological
and Personality Science, 2, 253–261. doi:10.1177/1948550610386808
Zebrowitz, L. A., Voinescu, L., & Collins, M. A. (1996). “Wide-eyed” and
“crooked-faced”: Determinants of perceived and real honesty across the
life span. Personality and Social Psychology Bulletin, 22, 1258 –1269.
doi:10.1177/01461672962212006
Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and
Kenny: Myths and truths about mediation analysis. Journal of Consumer
Research, 37, 197–206. doi:10.1086/651257
Received November 25, 2010
Revision received May 9, 2011
Accepted May 11, 2011 䡲