An Emphasis on Brilliance Fosters Masculinity Contest Cultures

Abstract

Women are underrepresented in fields where success is believed to require brilliance, but the reasons for this pattern are poorly understood. We investigate perceptions of a “masculinity contest culture,” an organizational environment of ruthless competition, as a key mechanism whereby a perceived emphasis on brilliance discourages female participation. Across three pre-registered correlational and experimental studies involving lay participants online (N = 870) and academics from 30+ disciplines (N = 1,347), we find a positive association between the perception that a field or organization values brilliance and the perception that this field or organization is characterized by a masculinity contest culture. This association was particularly strong among women. In turn, perceiving a masculinity contest culture predicted lower interest and sense of belonging, and stronger impostor feelings. Experimentally reducing the perception of a masculinity contest culture eliminated gender gaps in interest and belonging in a brilliance-oriented organization, suggesting possible avenues for intervention.

Full text

https://doi.org/10.1177/09567976211044133
Psychological Science
1 –18
© The Author(s) 2022
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/09567976211044133
www.psychologicalscience.org/PS
ASSOCIATION FOR
PSYCHOLOGICAL SCIENCE
Research Article
Women are underrepresented in fields in which success
is believed to require exceptional intelligence (or “bril-
liance”), such as philosophy, economics, and mathemat-
ics (Leslie etal., 2015; Meyer etal., 2015; Storage etal.,
2016). Although this relationship between a field’s per-
ceived emphasis on brilliance and women’s representa-
tion is well documented, the underlying causal
mechanisms are less understood. Here, we proposed
and tested the hypothesis that an emphasis on brilliance
has a negative effect on gender diversity because it
fosters a workplace climate in which a masculine-
coded, competition-based style of interpersonal interac-
tion is perceived to be dominant. Perceptions of such
a “masculinity-contest culture” (MCC; Berdahl etal.,
2018) may then lower women’s interest and well-being
in organizations or fields in which brilliance is strongly
emphasized.
MCC refers to organizational environments in which
individuals feel the “need to aggressively compete and
dominate other [people]” (Kupers, 2005, p. 713; see also
Berdahl etal., 2018). In such contexts, individuals may
be encouraged to display stereotypically masculine
behaviors and attitudes, such as aggressiveness, inde-
pendence, ambition, and competitiveness (Prentice
& Carranza, 2002; Spence etal., 1979), that may be
enacted in routine “mine’s bigger than yours” contests
(Berdahl etal., 2018). MCCs are experienced negatively
by men as well as women (Glick etal., 2018; Reid etal.,
2018), but they might be particularly difficult to navi-
gate for women, who are traditionally socialized to be
modest about their achievements and to avoid being
dominant or competitive (Heatherington etal., 1993;
Williams & Tiedens, 2016).
1044133PSSXXX10.1177/09567976211044133Vial et al.Psychological Science, XX(X)
research-article2022
Corresponding Author:
Andrea C. Vial, Department of Psychology, New York University Abu
Dhabi
Email: andrea.vial@nyu.edu
An Emphasis on Brilliance Fosters
Masculinity-Contest Cultures
Andrea C. Vial1, Melis Muradoglu2, George E. Newman3,
and Andrei Cimpian2
1Department of Psychology, New York University Abu Dhabi; 2Department of Psychology,
New York University; and 3School of Management, Yale University
Abstract
Women are underrepresented in fields in which success is believed to require brilliance, but the reasons for this
pattern are poorly understood. We investigated perceptions of a “masculinity-contest culture,” an organizational
environment of ruthless competition, as a key mechanism whereby a perceived emphasis on brilliance discourages
female participation. Across three preregistered correlational and experimental studies involving adult lay participants
online (N = 870) and academics from more than 30 disciplines (N = 1,347), we found a positive association between
the perception that a field or an organization values brilliance and the perception that this field or organization
is characterized by a masculinity-contest culture. This association was particularly strong among women. In turn,
perceiving a masculinity-contest culture predicted lower interest and sense of belonging as well as stronger impostor
feelings. Experimentally reducing the perception of a masculinity-contest culture eliminated gender gaps in interest
and belonging in a brilliance-oriented organization, suggesting possible avenues for intervention.
Keywords
brilliance, impostor feelings, gender stereotypes, masculinity-contest culture, sense of belonging, open data, open
materials, preregistered
Received 4/6/21; Revision accepted 8/4/21
TC

2 Vial et al.
An emphasis on brilliance may promote elements of
an MCC (Berdahl etal., 2018) in a few ways. Because
brilliance is associated with men (Bian etal., 2017;
Storage etal., 2020), privileging stereotypically mascu-
line traits could favor the proliferation of male-typed
behavior as the default (Cheryan etal., 2017; Cheryan
& Markus, 2020). Moreover, because brilliance is com-
monly viewed as a fixed attribute rather than something
that can be cultivated (Rattan etal., 2012), an emphasis
on brilliance may promote performance goals (Dweck,
2008): Individuals may feel pressured to show off their
intellectual talent and demonstrate that their intelli-
gence is superior to that of others. In turn, this felt
pressure may incite competition to attain “star status,”
encourage intellectually oriented dominance behaviors
(e.g., harsh criticism, dismissing opposing views), and
discourage collaboration. Indeed, past work suggests
that organizations in which talent is viewed as fixed
tend to be characterized by less collaborative norms
and more unethical behavior (Canning etal., 2020).
Thus, MCCs—or the perception of such cultures—
may be part of the reason that the belief that brilliance
is required for success, a belief that seems unprejudiced
on its face, creates gender inequality. This mechanism
fills an important gap in the literature: Although women
in past studies reported lower interest in brilliance-
oriented jobs, as well as a lower sense of belonging
(Bian etal., 2018; Deiglmayr etal., 2019) and stronger
feelings of being “impostors” (Muradoglu etal., 2021)
in such positions, we still know little about the proximal
variables underlying these effects: Why is it, exactly,
that messages that emphasize brilliance undermine
women’s interest and well-being? Here, we propose that
perceptions of MCCs are a key missing piece of the
puzzle.
Across three preregistered studies (and a preregis-
tered pilot study), we tested whether (a) fields or orga-
nizations that are perceived to value brilliance are also
perceived to have an MCC (Studies 1 and 2) and
whether (b) perceiving such a culture is in turn associ-
ated with lower interest and well-being (Studies 1–3).
Importantly, we also tested whether gender moderates
these hypothesized relationships: If the perception of
an MCC is part of the mechanism through which
brilliance-oriented domains undermine women’s par-
ticipation, as we hypothesized, then one or both of
these relationships should be stronger for women than
for men. That is, (a) women may be more likely than
men to perceive undesirable, masculinity-contest-type
behaviors as a result of an emphasis on brilliance, and/
or (b) women may be more adversely affected by such
perceptions than men.
To test the generalizability of our conclusions, we
examined these hypothesized links across a wide range
of academic fields (pilot study and Study 1) as well as
in business settings (Studies 2 and 3). Similarly, we
sampled both the general population (pilot study and
Studies 2 and 3) and academics at various career stages
from multiple universities (Study 1). To be able to speak
to the causal mechanisms involved, we gathered not
just correlational (pilot study and Study 1) but also
experimental (Studies 2 and 3) data. Our findings across
these studies highlight perceptions of an MCC as a key
mechanism by which an emphasis on brilliance under-
mines women’s success in academia and industry.
This investigation was approved by New York Uni-
versity’s institutional review board. For all studies,
including the pilot study, results were not examined
until data collection was complete. Throughout, we
report all measures, manipulations, and exclusions.
Materials, data, analysis scripts, and copies of the pre-
registrations for all studies can be found on OSF at
https://osf.io/fsven/.
Pilot Study
Method
Participants. A convenience sample of 302 individuals
(age: M = 33.57 years, SD = 10.05; 54.4% female; 72.6%
White) was recruited via Amazon’s Mechanical Turk
(MTurk), an online crowd-sourcing platform (Buhrmester
etal., 2011). Although the pilot participants were not in
academia, their perceptions of academic fields served as
an initial test of our hypotheses. Moreover, laypeople’s
beliefs on these topics could be consequential because
Statement of Relevance
Psychological scientists have repeatedly docu-
mented that women are underrepresented in
domains that prize raw intellectual talent (“bril-
liance”), including many fields in academia (e.g.,
philosophy, mathematics). However, the reasons
for this pattern of underrepresentation have
remained unclear. This research identifies a pos-
sible reason: An emphasis on brilliance in the
workplace gives rise to a “dog eat dog” atmo-
sphere of ruthless competition, which discourages
women’s participation and undermines their well-
being. This work suggests that we can make work-
places more inclusive by promoting a culture of
free exchange and openness and avoiding the
zero-sum logic of “separating the wheat from the
chaff”—the brilliant superstars from the plodding
masses.

Psychological Science, XX(X) 3
they may be transmitted to and shape the attitudes of oth-
ers (e.g., their children; Gunderson etal., 2012; Simpkins,
2015). Additional recruitment details, including how sam-
ple size was determined, are reported in the supplemen-
tal material available at https://osf.io/fsven/.
Procedures and measures. A detailed description of
the procedure and measures used in the pilot study is
available in the supplemental material. Each participant
answered 10 questions for each of nine randomly
assigned fields, including social sciences; humanities;
and science, technology, engineering, and mathematics
(STEM) disciplines (Leslie etal., 2015; Meyer etal., 2015).
Each of 27 fields was rated by approximately 100 partici-
pants. Participants rated their perceptions of each field’s
emphasis on brilliance (two items from Meyer et al.,
2015; r = .66, p < .001) and MCC (six items from Glick
etal., 2018; α = .86). Two additional items allowed us to
test two possible alternative explanations for the hypoth-
esized relationship between a field’s perceived emphasis
on brilliance and its perceived MCC. In response to the
first item, participants indicated how much they believed
that each field required systemizing (thinking systemati-
cally and abstractly) over empathizing (understanding
thoughts and emotions; Baron-Cohen, 2002; Billington
etal., 2007). Systemizing is a stereotypically male attri-
bute that has been argued to influence career choices
and success (e.g., Baron-Cohen, 2002), so fields that are
perceived to emphasize it may also be perceived to dis-
play stronger masculinity-contest norms. In response to
the second item, participants estimated the percentage of
all doctoral degrees in each field that were granted to
women in 2018 in the United States. If a participant
assumed, say, that there are few women in a field, they
may use this as a basis for inferring both a strong empha-
sis on brilliance (Leslie etal., 2015; Meyer etal., 2015)
and a strong MCC (Glick etal., 2018). Thus, adjusting for
participants’ estimates of the percentage of women with
PhDs in a field should account for a potential confound.
For a similar reason, we also recorded the actual percent-
age of women with PhDs in a field from authoritative
sources (Association of American Medical Colleges, 2018;
National Science Foundation, 2019) and used it as a
covariate in our analyses. We observed a positive correla-
tion between estimated and actual percentages of PhDs
granted to women (r = .41, p < .001), suggesting modest
levels of accuracy.
Analytic strategy. We analyzed the pilot data both at
the field level (aggregating responses across participants)
and at the participant level (for details, see the supple-
mental material at https://osf.io/fsven/).
1 For this pilot
study, we report standardized coefficients (βs), which
indicate the fraction of a standard deviation by which the
dependent variable changes in response to a 1-standard-
deviation increase in a continuous predictor or a shift from
one category to another in a categorical predictor. Predic-
tors were mean-centered in models that included interac-
tion terms, which facilitates interpretation of the lower
order coefficients. We report ωp2 as a measure of effect
size, calculated either with the estate esize command in
Stata Version 16 (for linear regressions) or with the effect-
size package Version 0.3.3 in R (for the mixed-effects
models; Ben-Shachar etal., 2020). We chose ωp2 over the
more common ηp2 because it is less biased (Albers &
Lakens, 2018). We follow Field’s (2013) guidelines for
interpreting the magnitude of ωp2 (very small effect: ωp2 <
.01; small effect: .01 ≤ ωp2 < .06; medium effect: .06 ≤ ωp2 <
.14; large effect: ωp2 ≥ .14). Because ωp2 corrects for bias, it
can be negative; we report negative ωp2 values as .00.
Results
For the field-level data (N = 27 fields), a linear regres-
sion indicated that a stronger perceived emphasis on
brilliance was associated with a stronger perceived MCC
(see Fig. 1; β = 0.69, SE = 0.14, 95% confidence interval
[CI] = [0.39, 0.99], p < .001, ωp2 = .44, a large effect). For
the participant-level data (N = 302), a linear mixed-
effects model with crossed random intercepts for par-
ticipant and field indicated that a stronger perceived
emphasis on brilliance was again associated with a
stronger perceived MCC (β = 0.25, SE = 0.02, 95% CI =
[0.22, 0.29], p < .001, ωp2 = .08, a medium effect). Par-
ticipant gender (0 = man, 1 = woman) did not moderate
this relationship (β = −0.01, SE = 0.03, 95% CI = [−0.07,
0.05], p = .73, ωp2 = .00). In addition, the relationship
between perceived emphasis on brilliance and per-
ceived MCC remained significant when we included
systemizing versus empathizing scores and the esti-
mated percentage of women with PhDs as covariates
or, in separate models, the actual percentage of women
with PhDs (see Tables S3 and S4 in the supplemental
material at https://osf.io/fsven/).
2
Study 1
The preregistered pilot study found support for the first
link in the proposed mechanism in the context of aca-
demia with a lay sample. Specifically, we found that
perceptions of an academic field’s emphasis on brilliance
were positively associated with perceptions of an MCC
in that field. This relationship was robust beyond the
(actual or perceived) representation of women and the
perceived importance of systemizing (over empathizing),
and it emerged among both women and men. In Study
1, we recruited a large sample of academics from a wide
range of fields, which provided a more ecologically valid

4 Vial et al.
test of the same relationship. This sample also allowed
us to investigate the link between academics’ perceptions
of an MCC in their field and their well-being (the second
link in the hypothesized pathway), focusing particularly
on belonging and impostor feelings.
Method
Participants. We contacted 43,607 academics using a
previously assembled database of email addresses for fac-
ulty, postdoctoral fellows, and graduate students from
nine public and private research-intensive (i.e., R1) uni-
versities across the United States (Muradoglu etal., 2021;
for details, see the supplemental material at https://osf.io/
fsven/). We decided a priori to include any academic who
accepted our invitation to participate and met the prereg-
istered criteria below. We obtained consent from 1,769
(4.06%) academics. Of those who consented, 146 indi-
viduals did not complete the study (provided no data). Of
participants who completed at least some portion of the
study, we excluded those who (a) did not select a field
from our predefined set or typed in a field that we could
not place in this set (n = 120), (b) indicated that they were
staff or an undergraduate student or did not indicate a
position (n = 53), (c) indicated that some of their
answers were jokes or random (n = 17), or (d) did not
fill out our two key measures, perceived emphasis on
brilliance and perceived MCC (n = 86). Although
we did not preregister this last exclusion criterion,
these participants were automatically dropped from
analyses involving the two key variables, so we decided
to exclude them to maintain a consistent sample size
across analyses.
The final sample comprised 1,347 participants (age:
M = 43.90 years, SD = 15.66; 45.5% female; 78.4%
White), including graduate students (29.2%), postdoc-
toral fellows (8.3%), medical residents (0.9%), non-
tenure-track faculty (25.7%), untenured tenure-track
faculty (7.3%), tenured faculty (26.7%), and retired and
emeriti faculty (1.9%). There were 30 nonmedical aca-
demic disciplines represented (29.3% of participants
were in STEM, 35.3% in the social sciences, and 12.3%
in the humanities) and 33 medical fields (23.1% of
participants). The number of participants per discipline
is reported in Table S6 in the supplemental material at
https://osf.io/fsven/.
Art History
Classics Comparative Lit
English Lit
History Sociology
Music
Philosophy
Spanish
Film
Education
Anthropology
Archaeology
Communication
Economics
Linguistics
Geography
Psychology
Astronomy
Chemistry
Computer Science
Earth Science
Engineering
Mathematics
Physics
Neuroscience
Biology
2.5
3.0
3.5
3.0 3.5 4.0 4.5 5.0 5.5
Perceived Emphasis on Brilliance
Perceived Masculinity-Contest Culture
Fig. 1. Relationship between perceived emphasis on brilliance and perceptions of a masculinity-contest culture at
the field level in the pilot study. The line indicates the best-fitting regression, and the error band represents ±1 SE.

Psychological Science, XX(X) 5
Procedures and measures. Participants selected their
academic discipline from a predetermined list and com-
pleted four measures, in random order: (a) perceived
emphasis on brilliance in their field, (b) perceived MCC in
their field, (c) their own impostor feelings, and (d) their
own sense of belonging. Items within each measure were
presented randomly. Participants were then asked to esti-
mate the percentage of doctoral degrees in their field
granted to (a) women and (b) underrepresented minori-
ties, in that order.
Perceived emphasis on brilliance. Participants indi-
cated their agreement with the same two statements as
in the pilot study but with reference to their own disci-
pline (Leslie etal., 2015) from 1 (strongly disagree) to 7
(strongly agree). The two statements were, “Being a top
performer in my field requires a special aptitude that just
can’t be taught,” and “If you want to succeed in my field,
hard work alone just won’t cut it; you need to have an
innate gift or talent.” Participants rated each statement
twice (i.e., four items total): once in regard to their own
beliefs about their current field (self-ratings) and once
in regard to the perceived beliefs of other academics in
that field (other-ratings). Self- and other-items were pre-
sented randomly within two separate blocks in random-
ized order. We first averaged the two self-ratings and the
two other-ratings separately; the resulting scores were
moderately correlated (r = .56, p < .001). Similar patterns
of results were observed in the models below when the
self- and other-ratings were examined separately; thus,
we combined the four items into a single measure of
perceived emphasis on brilliance (α = .85).
Perceived MCC. We used the six statements from the
pilot study to assess participants’ perceptions that their
current discipline is characterized by an MCC (Glick etal.,
2018). The items were modified to include the stem, “In my
field,” and were rated from 1 (not at all true of field) to 5
(entirely true of field). We included two items each from the
“show no weakness” subscale (“Admitting you don’t know
the answer looks weak”; “Seeking others’ advice is seen as
weak”); the “put work first” subscale (“Taking days off is
frowned upon”; “People with significant demands outside
of work don’t make it very far”); and the “dog eat dog”
subscale (“One person’s loss is another person’s gain”; “If
you don’t stand up for yourself people will step on you”).
We did not include any items from the “strength and stam-
ina” subscale because beliefs about physical strength and
stamina are less relevant in academia. The two items within
each subscale were significantly correlated (rs = .46–.55).
We averaged the six items together (α = .81).
Impostor feelings. Participants reported their impostor
feelings in their current field with five items (α = .93)
from the study by Muradoglu et al. (2021). These items
were based on the Clance Impostor Phenomenon Scale
(Clance & Imes, 1978; Simon & Choi, 2018) and were
rated from 1 (strongly disagree) to 7 (strongly agree). A
sample item is, “Sometimes I’m afraid others will discover
how much knowledge or ability I really lack.”
Sense of belonging. We measured participants’ sense
of belonging in their current field with eight items (α =
.91) adapted from the study by Good et al. (2012), which
were rated from 1 (strongly disagree) to 7 (strongly agree).
A sample item is, “I feel accepted by other members in
my field.”
Alternative explanation: estimated percentage of PhDs
granted to women and underrepresented minorities. As
in the pilot study, we examined whether the relation-
ship between perceived emphasis on brilliance and per-
ceived MCC might be explained by a third variable: a
field’s assumed gender balance. We asked participants to
estimate the percentage of all doctoral degrees in their
current field that were granted to women in 2018 in the
United States, from 0 (0% of PhD degrees to women) to
100 (100% of PhD degrees to women). As in the pilot
study, we also examined the actual percentage of women
with PhDs as a covariate. Actual and estimated percent-
ages were strongly correlated (r = .70, p < .001). Finally,
we asked participants to estimate the percentage of all
doctoral degrees in their current field that were granted
to “members of racial/ethnic minority groups traditionally
underrepresented in academia (e.g., Black, Latinx, Native
American)” in 2018 in the United States, from 0 (0% of
PhD degrees to racial/ethnic minorities) to 100 (100% of
PhD degrees to racial/ethnic minorities).
Demographic information and debriefing. At the end
of the study, participants indicated their current position
(e.g., graduate student) and provided basic demographic
information (e.g., gender, race). Finally, participants were
asked to type any thoughts they had about the study (open
ended) and to indicate whether any of their answers were
random or meant as jokes (yes/no).
Results
The analytic strategy was the same as for the pilot study.
Means, standard deviations, and bivariate correlations
between all study variables are reported in Table S7 in
the supplemental material at https://osf.io/fsven/. We
report standardized coefficients for regression and
mediation models.
Link between perceived emphasis on brilliance and
perceived MCC. First, we conducted a linear mixed-
effects model in which we regressed perceived MCC on
perceived emphasis on brilliance. The model included a

6 Vial et al.
random intercept for field to account for the nesting of
participants in fields. Perceived emphasis on brilliance
was positively associated with perceived MCC (β = 0.17,
SE = 0.03, 95% CI = [0.12, 0.23], p < .001, ωp2 = .03, a small
effect). Next, we added participant gender to the mixed-
effects model to test for moderation by this variable.
Female academics were overall more likely to perceive
an MCC than male academics were (β = 0.39, SE = 0.05,
95% CI = [0.29, 0.50], p < .001, ωp2 = .04). Critically, as
seen in Figure 2, the model also revealed a significant
two-way interaction between perceived emphasis on bril-
liance and participant gender (β = 0.19, SE = 0.05, 95% CI =
[0.09, 0.30], p < .001, ωp2 = .01): The association between
perceived emphasis on brilliance and perceived MCC
was significantly stronger among female academics (β =
0.29, SE = 0.04, 95% CI = [0.22, 0.37], p < .001, ωp2 = .08,
a medium effect) than among male academics (β = 0.10,
SE = 0.04, 95% CI = [0.03, 0.17], p = .006, ωp2 = .01, a
small effect). These results did not change appreciably
when we adjusted for participants’ estimates of the per-
centage of PhDs granted to women and underrepre-
sented minorities or the actual percentages of women
with PhDs in a discipline (see Table S8 in the supplemen-
tal material at https://osf.io/fsven/).
Perceived MCC as a mediator between perceived
emphasis on brilliance and well-being. We hypoth-
esized that (a) perceiving a stronger emphasis on bril-
liance in one’s field would be associated with stronger
perceptions of an MCC and that (b) in turn, the latter
perceptions would predict lower well-being. In addition,
we hypothesized that one or both of these relationships
would be stronger for women than for men. In the con-
text of a mediation model, this set of hypotheses led to
the prediction of stronger indirect effects (via perceived
MCC) for women than for men.
Consistent with this prediction, a moderated media-
tion analysis using the PROCESS module for SPSS
(Hayes, 2015; Model 8; see Fig. 3) revealed that partici-
pant gender significantly moderated the indirect effect
of a field’s perceived emphasis on brilliance (X) through
its perceived MCC (M) on both impostor feelings (Y;
index of moderated mediation: 0.06, SE = 0.02, 95%
CI = [0.03, 0.10]) and sense of belonging (Y; index of
moderated mediation: −0.11, SE = 0.03, 95% CI = [−0.17,
−0.05]). (The use of Model 8 was a deviation from our
preregistered analytic plans, as detailed in the supple-
mental material at https://osf.io/fsven/, including Table
S13.) The indirect effects of a field’s perceived emphasis
on brilliance were significantly stronger for women
(impostor feelings—ab = 0.09, SE = 0.01, 95% CI = [0.07,
0.13]; belonging—ab = −0.16, SE = 0.02, 95% CI = [−0.21,
−0.12]) compared with men (impostor feelings—ab =
0.03, SE = 0.01, 95% CI = [0.01, 0.05]; belonging—ab =
−0.05, SE = 0.02, 95% CI = [−0.09, −0.02]). Note, how-
ever, that participant gender did not moderate the rela-
tionship between perceived MCC and the two outcomes
(i.e., the b paths; impostor feelings: β = −0.07, SE = 0.05,
1
2
3
4
5
1234567
Perceived Emphasis on Brilliance
Perceived Masculinity-Contest Culture
Women
Men
Fig. 2. Relationship between perceived emphasis on brilliance and perceived masculinity-
contest culture among academics in Study 1, separately for each participant gender. The
lines indicate the best-fitting regressions, and the error bands represent ±1 SE.

Psychological Science, XX(X) 7
95% CI = [−0.17, 0.03], p = .170, ωp2 = .001; sense of
belonging: β = −0.07, SE = 0.04, 95% CI = [−0.16, 0.01],
p = .102, ωp2 = .001). These results did not change
appreciably when we adjusted for the estimated per-
centage of women and underrepresented minorities
with PhDs or for the actual percentages of PhDs granted
to women (see Tables S9 and S10 in the supplemental
material at https://osf.io/fsven/).
Ancillary analysis: low self-confidence as alterna-
tive explanation? Finally, we investigated the possibil-
ity that the negative relationship between perceived MCC
and sense of belonging is explained by an internalized
lack of confidence, particularly among female academics.
We used impostor feelings as a proxy for low confidence
(e.g., Muradoglu etal., 2021) and examined whether the
relationship between perceived MCC and sense of belong-
ing emerged above and beyond any variance explained
by impostor feelings. The results, reported in full in the
supplemental material (see Table S11 at https://osf.io/
fsven/), suggested that the relationship between perceived
MCC and sense of belonging emerges independently of
the variance explained by impostor feelings, among both
female and male academics.
Discussion
Academics who thought that their field valued brilliance
also perceived their work environments to be character-
ized by an MCC. This relationship was stronger among
a
b
Perceived
Emphasis on
Brilliance
Impostor
Feelings
Perceived
Masculinity-
Contest Culture
Participant Gender
(0 = Man,
1 = Woman)
Men: 0.07 (0.04)
Women: 0.18∗∗∗ (0.09∗)
Perceived
Emphasis on
Brilliance
Sense of
Belonging
Men: −0.50∗∗∗
Women: −0.57∗∗∗
Men: 0.36∗∗∗
Women: 0.29∗∗∗
Men: 0.10∗∗
Women: 0.29∗∗∗
Men: 0.10∗∗
Women: 0.29∗∗∗
Perceived
Masculinity-
Contest Culture
Men: −0.02 (0.03)
Women: −0.13∗∗ (0.02)
Participant Gender
(0 = Man,
1 = Woman)
Fig. 3. Effects of perceived emphasis on brilliance through perception of a masculinity-contest culture on (a)
impostor feelings and (b) sense of belonging in Study 1, as moderated by participant gender. Standardized coef-
ficients are reported. On the path from the independent to the dependent variable, values outside parentheses
show the total effect, and values inside parentheses show the direct effect after controlling for the mediator
and moderators. All path coefficients were calculated using linear mixed-effects models in Stata Version 16,
whereas the indirect effects and indices of moderated mediation were calculated via simple regressions using
the PROCESS module for SPSS. Asterisks indicate significant paths (*p < .05, **p < .01, ***p <.001).

8 Vial et al.
female academics than male academics, unlike in the
pilot study with laypeople, where we found no modera-
tion by gender: Perhaps more first-hand exposure to
academia (Study 1) prompts women’s and men’s
responses to diverge. For example, women may be
more sensitive than men to the effects that an emphasis
on brilliance (a stereotypically masculine trait) has on
other academics’ behaviors because these behaviors
often place women at a disadvantage (Cheryan &
Markus, 2020). The results of Study 1 also revealed, as
hypothesized, that the indirect effects of a field’s per-
ceived emphasis on brilliance on academics’ well-being
via their perception of an MCC were stronger for female
than male academics. We tested these relationships
experimentally in Studies 2 and 3.
Study 2
In Study 2, we experimentally tested the first link in the
proposed causal mechanism—namely, that a work envi-
ronment that emphasizes brilliance licenses perceptions
of an MCC more than an environment that does not
emphasize brilliance. We also measured participants’
interest in working in this environment and their antici-
pated well-being, which afforded a (nonexperimental)
test of the second link—namely, that perceptions of an
MCC are associated with lower interest and well-being.
As before, we expected one or both of these links to
be moderated by gender, resulting in stronger indirect
effects of the manipulation for women than men.
Method
Participants. An a priori power analysis (G*Power Ver-
sion 3.1; Faul etal., 2009) for a regression model with up
to four predictors indicated that a sample size of 273 par-
ticipants would be required to detect a small to medium
effect size (f2 = .029), assuming power of .80 and an α of
.05. This effect size was the average reported by Bian et al.
(2018). We increased the target sample size by 15% to
account for exclusions and recruited a convenience sam-
ple of 316 individuals via MTurk (Buhrmester etal., 2011).
The study was available to workers in the United States
with prior approval rates of 95% or higher, and partici-
pants received $0.55 for their time. Following our prereg-
istered criteria, we excluded 24 participants who (a)
indicated that some of their answers were jokes or ran-
dom, (b) provided nonsense responses in an open-ended
question described in the procedure, or (c) had duplicate
Internet protocol (IP) addresses (final N = 292; age: M =
34.25 years, SD = 11.04; 55.3% female; 69.9% White).
Procedures and measures. In this study, we extended
our investigation from academia to nonacademic profes-
sional opportunities, where an emphasis on brilliance
has similarly been found to discourage women’s partici-
pation (Bian etal., 2018). We employed an experimental
manipulation adapted from the work by Bian et al. (2018)
to convey a focus on brilliance. Participants read informa-
tion ostensibly from a company’s website advertising
new openings to join the company’s workforce, which
included a description of the types of attributes the com-
pany values in its employees (for the full script, see at
https://osf.io/fsven/). Half of the participants were ran-
domly assigned to a brilliance condition in which the
company advertisement emphasized candidates’ excep-
tional intellectual ability (e.g., “a high IQ,” “superior rea-
soning skills,” “natural intelligence”). The other half of
participants were assigned to a control condition in
which the advertisement emphasized candidates’ skills
without reference to brilliance (e.g., “broad range of
skills”; “comfortable with a modern, dynamic workplace”;
“positive thinking and productivity”). Bian et al.’s results
suggested that women’s and men’s attitudes toward the
company described in the control condition were similar
(i.e., this condition was gender neutral).
After the manipulation, participants completed four
measures in random order: (a) perceptions of an MCC
(Glick etal., 2018), (b) interest in working in the com-
pany, (c) anticipated impostor feelings (Clance & Imes,
1978), and (d) anticipated sense of belonging (Good
etal., 2012). Item order was random within measures,
and a manipulation check followed.
Perceived MCC. We assessed participants’ perceptions
that the company was characterized by an MCC with the
six statements from Study 1, adapted to the hypotheti-
cal scenario (i.e., “In this company . . .”). Statements were
rated from 1 (not at all true) to 5 (entirely true). The
two items within each of the three subscales were signifi-
cantly correlated (rs = .50–.58). As before, we averaged
all items to compute an MCC score (α = .88).
Interest in the company. Participants answered three
questions (α = .95) borrowed from the study by Bian
et al. (2018) to gauge their interest in the company (e.g.,
“Assuming you were looking for a job, how likely would
you be to apply for a position at this particular com-
pany?”). Statements were rated from 1 (not at all inter-
ested) to 9 (extremely interested).
Anticipated impostor feelings. To measure partici-
pants’ anticipated impostor feelings if they were to work
at the company, we used the five items (α = .94) from
Study 1 (Clance & Imes, 1978; Simon & Choi, 2018),
adapted to the hypothetical scenario (e.g., “If I worked at
this company, I would be afraid that people in the com-
pany may find out that I’m not as capable as they think
I am”). Items were rated from 1 (strongly disagree) to 7
(strongly agree).

Psychological Science, XX(X) 9
Anticipated sense of belonging. We measured partici-
pants’ anticipated sense of belonging if they were to work
at the company using the eight items (α = .92) from Study
1 (Good et al., 2012), adapted to the hypothetical sce-
nario, as in the study by Bian et al. (2018; e.g., “I would
feel valued by other company employees”). Items were
rated from 1 (strongly disagree) to 7 (strongly agree).
Alternative explanation: estimated percentage of female
employees. Following a similar procedure as in the pilot
study and Study 1, we investigated whether the effects
of the brilliance manipulations boiled down to an effect
on participants’ inferences about the company’s gender
composition, which by itself could influence women’s
and men’s interest and sense of well-being (e.g., Heilman,
1979). We might expect, for example, that women would
be more interested and anticipate higher well-being in
contexts with a higher percentage of women through basic
homophily effects (Holman & Morandin, 2019; McPherson
etal., 2001). To measure this potential confound, we asked
participants to estimate the percentage of all employees at
the company who were women from 0 to 100.
Manipulation check. Participants indicated their agree-
ment with a single item: “The company emphasizes employ-
ees’ natural intelligence and inherent aptitude,” rated from
1 (strongly disagree) to 5 (strongly agree).
Demographic information and debriefing. The study
ended by asking participants to type any thoughts they
had about the study (open ended), provide demographic
information (e.g., gender, race), and indicate whether
any of their answers were random or jokes (yes/no).
Results
Means, standard deviations, and bivariate correlations
between all study variables are reported in Table S14
in the supplemental material at https://osf.io/fsven/.
We report unstandardized coefficients for regression
and mediation models.
Manipulation check. As intended, participants in the
brilliance condition were significantly more likely to
agree that the company emphasized employees’ natural
intelligence and inherent aptitude (M = 4.16, SD = 0.96),
compared with participants in the control condition (M =
3.76, SD = 0.93), F(1, 287) = 12.79, p < .001, ωp2 = .04.
There was no significant main effect of participant gen-
der, F(1, 287) = 0.02, p = .894, ωp2 = .00, and no signifi-
cant Condition × Participant Gender two-way interaction,
F(1, 287) = 0.02, p = .882, ωp2 = .00.
Link between emphasis on brilliance and percep-
tions of an MCC. We regressed the perception of an
MCC on experimental condition (control = 0, brilliance =
1), participant gender (man = 0, woman = 1), and their
interaction. As expected, the perception of an MCC was
significantly stronger in the brilliance condition (M =
3.31, SD = 0.89) compared with the control condition (M =
2.91, SD = 0.86; b = 0.37, SE = 0.10, 95% CI = [0.17, 0.58],
p < .001, ωp2 = .04, a small to medium effect). There was
no main effect of gender (b = 0.06, SE = 0.10, 95% CI =
[−0.14, 0.27], p = .53, ωp2 = .00), but the Condition × Par-
ticipant Gender two-way interaction was significant (b =
0.42, SE = 0.21, 95% CI = [0.17, 0.83], p = .041, ωp2 = .01;
see Fig. 4). The effect of condition on perception of
masculinity-contest norms was not significant for male
participants (b = 0.16, SE = 0.15, 95% CI = [−0.14, 0.47],
p = .29, ωp2 = .00) but was significant for female partici-
pants (b = 0.59, SE = 0.14, 95% CI = [0.32, 0.86], p < .001,
ωp2 = .11, a medium effect). These results did not change
appreciably when we adjusted for the estimated percent-
age of female employees in the company (see Table S15
in the supplemental material at https://osf.io/fsven/).
Perceptions of an MCC as a mediator between
emphasis on brilliance and downstream outcomes.
We expected to find stronger indirect relationships for
female than for male participants between experimen-
tal condition (X) and the three downstream outcomes
Women Men
Control Brilliance ControlBrilliance
1
2
3
4
5
Perceived Masculinity-Contest Culture
Fig. 4. Perception of masculinity-contest cultures as a function of
condition, separately for each participant gender in Study 2. Each dot
represents an individual participant’s response. Within each box plot,
the solid line in the middle represents the median, and the diamond
represents the mean. The bottom and top edges of the box indicate
the 25th and 75th percentiles, respectively, and whiskers extend 1.5
times the interquartile range.

10 Vial et al.
(Y; interest in joining the company, anticipated impostor
feelings, and anticipated sense of belonging) via percep-
tions of an MCC (M). We again used PROCESS Model 8,
which was a deviation from our preregistered analytic
plans, as detailed in the supplemental material at https://
osf.io/fsven/, including Table S20. Indeed, participant
gender significantly moderated the indirect effects for all
three downstream outcomes (see Fig. 5 and Tables 1 and
2)—interest in the company: index of moderated media-
tion: −0.44, SE = 0.22, 95% CI = [−0.89, −0.01]; anticipated
impostor feelings: index of moderated mediation: 0.39,
SE = 0.19, 95% CI = [0.02, 0.77]; and anticipated sense of
belonging: index of moderated mediation: −0.23, SE =
0.12, 95% CI = [−0.47, −0.004]. Specifically, these indirect
relationships were significant for female participants
(interest—ab = −0.61, SE = 0.15, 95% CI = [−0.93, −0.33];
impostor feelings—ab = 0.54, SE = 0.13, 95% CI = [0.29,
0.81]; and sense of belonging—ab = −0.32, SE = 0.08, 95%
CI = [−0.50, −0.17]). However, for male participants, none
of the three indirect pathways were significant (interest—
ab = −0.17, SE = 0.17, 95% CI = [−0.53, 0.16]; impostor
feelings—ab = 0.15, SE = 0.15, 95% CI = [−0.13, 0.45]; and
sense of belonging—ab = −0.09, SE = 0.09, 95% CI = [−0.28,
0.08]). These differences were largely due to the a paths
(see Fig. 4); the b paths (from perceptions of an MCC
to outcomes) were not significantly different for women
and men (see Table 2).
None of these results changed appreciably when we
adjusted for the estimated percentage of female employ-
ees in the company (see Tables S15–S17 in the supple-
mental material at https://osf.io/fsven/). Additionally, as
in Study 1, the relationships between perceived MCC and
interest and anticipated sense of belonging remained sig-
nificant when we adjusted for impostor feelings (see Table
S18 in the supplemental material at https://osf.io/fsven/).
Discussion
When a company emphasized brilliance, participants
expected it to have a stronger MCC. This causal link
emerged only among women, which is consistent with
the findings of Study 1, where the correlation between
perceptions of a field’s brilliance orientation and its
expected masculinity-contest norms was stronger for
female than for male academics. Also consistent with
Study 1, results of Study 2 showed that stronger percep-
tions of an MCC were associated with more negative
outcomes for both women and men.
Finally, the brilliance-emphasis manipulation had an
indirect effect—via perceptions of an MCC—on interest
and well-being only for women. This finding provides
support for the proposal that the perception of an MCC
functions as a mechanism by which an emphasis on
brilliance discourages women’s participation.
Table 1. Results of Regression Models Predicting Downstream Outcomes (Interest in
Company, Anticipated Impostor Feelings, and Anticipated Sense of Belonging) as a Function
of Participant Gender and Experimental Condition in Study 2
Dependent variable and predictor b SE p ωp2
Interest in company
Participant gendera−0.42 [−0.91, 0.07] 0.25 .097 .01
Experimental conditionb−1.03 [−1.52, −0.54] 0.25 < .001 .05
Participant Gender × Experimental Condition −1.13 [−2.12, −0.15] 0.50 .024 .01
Anticipated impostor feelings
Participant gendera0.14 [−0.22, 0.50] 0.18 .443 .00
Experimental conditionb0.61 [0.24, 0.97] 0.18 .001 .03
Participant Gender × Experimental Condition 0.85 [0.12, 1.58] 0.37 .022 .01
Anticipated sense of belonging
Participant gendera−0.08 [−0.33, 0.18] 0.13 .548 .00
Experimental conditionb−0.59 [−0.85, −0.34] 0.13 < .001 .06
Participant Gender × Experimental Condition −0.44 [−0.95, 0.07] 0.26 .089 .01
Note: Predictors were mean-centered to facilitate interpretation of the coefficients. The coefficients of participant
gender and experimental condition in this table can be interpreted as one would interpret the main effects in
an analysis-of-variance table. These results did not change appreciably when we adjusted for the estimated
percentage of female employees in the company (see Table S15 in the supplemental material at https://osf.io/
fsven/). Values in brackets are 95% confidence intervals.
aParticipant gender was coded 0 for man and 1 for woman. bExperimental condition was coded 0 for control
and 1 for brilliance.

Psychological Science, XX(X) 11
Men: −1.16∗∗∗
Women: −1.14∗∗∗
Men: −0.56∗∗∗
Women: −0.65∗∗∗
Men: 0.97∗∗∗
Women: 0.96∗∗∗
Men: 0.16
Women: 0.59∗∗∗
Men: 0.16
Women: 0.59∗∗∗
Men: 0.16
Women: 0.59∗∗∗
Condition
(0 = Control,
1 = Brilliance)
Interest
Perceived
Masculinity-
Contest Culture
Men: −0.47 (−0.30)
Women: −1.60∗∗∗ (−0.99∗∗)
Men: 0.18 (0.03)
Women: 1.03∗∗∗ (0.50∗)
Men: −0.37 (−0.28)
Women: −0.81∗∗∗ (−0.49∗∗)
Participant Gender
(0 = Man,
1 = Woman)
Condition
(0 = Control,
1 = Brilliance)
Anticipated
Impostor Feelings
Perceived
Masculinity-
Contest Culture
Participant Gender
(0 = Man,
1 = Woman)
Condition
(0 = Control,
1 = Brilliance)
Anticipated Sense
of Belonging
Perceived
Masculinity-
Contest Culture
Participant Gender
(0 = Man,
1 = Woman)
a
b
c
Fig. 5. Effects of experimental condition through perception of a masculinity contest culture on (a) interest, (b)
anticipated impostor feelings, and (c) anticipated sense of belonging in Study 2, as moderated by participant gen-
der. Unstandardized coefficients are reported. On the path from the independent to the dependent variable, values
outside parentheses show the total effect, and values inside parentheses show the direct effect after controlling for
the mediator and moderators. All path coefficients were calculated via simple regression models in Stata Version
16, whereas the indirect effects and indices of moderated mediation were calculated using the PROCESS module
for SPSS. Asterisks indicate significant paths (*p < .05, **p < .01, ***p <.001).

12 Vial et al.
Study 3
In Study 3, we experimentally tested the second link in
the proposed causal mechanism—namely, that work
environments perceived to be characterized by an MCC
undermine interest and well-being more than work
environments not perceived to be characterized by an
MCC. In this study, we kept the emphasis on brilliance
constant—and high—across conditions and manipu-
lated only the perceived workplace norms, measuring
their effects on participants’ interest and anticipated
well-being, as well as whether these effects differ for
women and men. Although gender did not moderate
the negative relationship between perceived MCC and
these outcomes in Studies 1 and 2, Study 3 provided
the first opportunity to investigate these potential mod-
eration effects experimentally. Notably, this study also
speaks to potential interventions: If an emphasis on
brilliance is discouraging to women because it licenses
the expectation of an MCC, then it is important to know
whether countering these expectations makes a brilliance-
oriented workplace equally motivating and psychologi-
cally safe for women and men.
Method
Participants. As in Study 2, we based our target sam-
ple size on effect sizes from the study by Bian et al. (2018)
and an a priori power analysis (G*Power Version 3.1;
Faul etal., 2009) for a regression model with up to four
predictors. The power analysis indicated that to detect
the small to medium effect size (f
2 = .029) found by Bian
etal., assuming power of .80 and an α of .05, we would
need to recruit 273 participants. We increased the target
sample size by 15% to account for exclusions. We
recruited a convenience sample of 357 individuals via
MTurk (Buhrmester etal., 2011). The study was available
to workers in the United States with prior approval rates
of 95% or higher, and participants received $0.55 for
completing the study. Following our preregistered crite-
ria, we excluded 81 participants who (a) indicated that
some of their answers were jokes or random, (b) pro-
vided nonsense responses in an open-ended question
described in the procedure, or (c) had duplicate IP
addresses (final N = 276; age: M = 33.22 years, SD =
11.51; 55.8% female; 65.6% White). Gender information
was missing for three participants.
Procedures and measures. As in Study 2, we extended
our investigation from academia to nonacademic profes-
sional opportunities (Bian etal., 2018). Participants read
information ostensibly from a company’s website adver-
tising new openings to join the company’s workforce. In
Study 3, for all participants, the advertisement empha-
sized employees’ exceptional intellectual ability (i.e.,
identical to the brilliance condition in Study 2). To manip-
ulate perceptions of an MCC at the hypothetical com-
pany, we asked participants to imagine that they had an
Table 2. Results of Regression Models Predicting Downstream Outcomes
(Interest in Company, Anticipated Impostor Feelings, and Anticipated Sense of
Belonging) as a Function of Participant Gender and Masculinity-Contest Culture
(MCC) Ratings in Study 2
Dependent variable and
predictor b SE p ωp2
Interest in company
Participant gendera−0.33 [−0.78, 0.12] 0.23 .151 .003
MCC −1.15 [−1.40, −0.90] 0.13 < .001 .22
Participant Gender × MCC 0.03 [−0.47, 0.53] 0.25 .915 .00
Anticipated impostor feelings
Participant gendera0.07 [−0.24, 0.39] 0.16 .642 .00
MCC 0.96 [0.79, 1.14] 0.09 < .001 .29
Participant Gender × MCC −0.01 [−0.36, 0.34] 0.18 .948 .00
Anticipated sense of belonging
Participant gendera−0.03 [−0.26, 0.20] 0.12 .785 .00
MCC −0.60 [−0.73, −0.47] 0.06 < .001 .22
Participant Gender × MCC −0.09 [−0.35, 0.17] 0.13 .500 .00
Note: Predictors were mean-centered to facilitate interpretation of the coefficients. The
coefficients of participant gender and MCC in this table can be interpreted as one would
interpret the main effects in an analysis-of-variance table. These results did not change
appreciably when we adjusted for the estimated percentage of female employees in the
company (see Table S16 in the supplemental material at https://osf.io/fsven/). Values in
brackets are 95% confidence intervals.
aParticipant gender was coded 0 for man and 1 for woman.

Psychological Science, XX(X) 13
acquaintance currently employed at the company and
that they had sent this person an email asking what it
was like to work there. Each participant was randomly
assigned to either a high-MCC condition, in which the
acquaintance’s response suggested that the company was
strongly characterized by an MCC (e.g., “There’s some-
times a sense here that admitting you don’t know the
answer or seeking others’ advice looks weak”), or a low-
MCC condition, in which the acquaintance’s response
indicated that the company was not characterized by an
MCC (e.g., “There’s usually a sense here that admitting
you don’t know the answer or seeking others’ advice is
okay”). The full manipulation is reported in the supple-
mental material at https://osf.io/fsven/.
After the manipulation, participants completed the
same three measures as in Study 2, in random order:
(a) interest in working in the company (α = .93), (b)
anticipated impostor feelings (α = .92), and (c) antici-
pated sense of belonging (α = .90), followed by a ques-
tion asking participants to estimate the percentage of
all employees at the company who were women, as in
Study 2. Item order was random within measures, and
two manipulation checks followed.
Manipulation checks. To confirm that our manipu-
lation shaped perceptions of an MCC as intended, we
asked participants to indicate their agreement with the
item, “The company has a work environment of ruthless
competition,” on a scale from 1 (strongly disagree) to 5
(strongly agree). In addition, we considered the possibility
that the relationship between emphasis on brilliance and
the perception of an MCC might be bidirectional. Thus, to
investigate whether the low-MCC condition inadvertently
lowered the perception of an emphasis on brilliance rela-
tive to the high-MCC condition (which would compromise
our conclusions), we asked participants to indicate their
agreement with a second manipulation-check item, “The
company emphasizes employees’ natural intelligence and
inherent aptitude” (1 = strongly disagree, 5 = strongly agree).
Demographic information and debriefing. Study 3
ended by asking participants to type any thoughts they
had about the study (open ended), provide demographic
information (e.g., gender, race), and indicate whether
any of their answers were random or jokes (yes/no).
Results
Means, standard deviations, and bivariate correlations
between all study variables are reported in Table S21
in the supplemental material at https://osf.io/fsven/.
We report unstandardized coefficients for the regression
and mediation analyses.
Manipulation checks. As expected, participants in the
high-MCC condition were significantly more likely to
perceive a work environment of ruthless competition
(M = 4.24, SD = 0.88), compared with participants in the
low-MCC condition (M = 2.05, SD = 1.07), F(1, 269) =
331.03, p < .001, ωp2 = .56, a large effect. The effect of
participant gender and its interaction with condition were
not significant (ps > .336).
Contrary to our concerns, results showed that par-
ticipants rated the company’s emphasis on brilliance
as being higher in the low-MCC condition (M = 4.38,
SD = 0.68) than in the high-MCC condition (M = 4.13,
SD = 1.00), F(1, 269) = 5.20, p = .023, ωp2 = .01, a small
effect. It is possible that this is a positivity spillover
effect: If participants perceived the low-MCC company
more positively as a result of its weak MCC, they may
have also inferred that this company is able to attract
the most competent people. Regardless of the reason
for this difference, the important point is that it works
against our ability to find the predicted effects. The
effect of participant gender and its interaction with
condition were not significant (ps > .55).
Link between perception of an MCC and downstream
outcomes. First, we examined whether the MCC manip-
ulation influenced the three downstream outcomes: (a)
interest in the company, (b) anticipated impostor feel-
ings, and (c) anticipated sense of belonging. To do this,
we regressed each of the three outcome variables on
MCC condition (low = 0, high = 1). As expected, partici-
pants in the high-MCC condition reported lower interest
in employment opportunities at the hypothetical com-
pany compared with those in the low-MCC condition
(b = −2.67, SE = 0.22, 95% CI = [−3.10, −2.24], p < .001,
ωp2 = .35, a large effect). High-MCC participants also
anticipated stronger impostor feelings (b = 1.47, SE =
0.16, 95% CI = [1.15, 1.79], p < .001, ωp2 = .23, a large
effect) and a lower sense of belonging (b = −1.74, SE =
0.13, 95% CI = [−1.99, −1.49], p < .001, ωp2 = .40, a large
effect).
Given the unexpected effect of the MCC manipula-
tion on participants’ perceptions of the company’s
emphasis on brilliance, we also tested whether the
effects of MCC condition on interest and well-being
emerged above and beyond these perceptions. (This
analysis was not preregistered.) Indeed, the effects of
MCC condition remained significant for all outcomes
(see Table S22 in the supplemental material at https://
osf.io/fsven/).
Moderation by gender. Next, we investigated whether
the consequences of perceiving an MCC in a brilliance-
oriented organization varied on the basis of participant
gender. For the purposes of this analysis, we regressed
each of the three outcome variables on MCC condition
(low = 0, high = 1), participant gender (man = 0, woman =
1), and their interaction. The results of these models,

14 Vial et al.
reported in Table 3, revealed a significant Condition ×
Gender interaction on interest in the company (b = −0.95,
SE = 0.44, 95% CI = [−1.82, −0.09], p = .031, ωp2 = .01). For
the two other outcomes, this interaction was not signifi-
cant (impostor feelings: b = −0.03, SE = 0.33, 95% CI =
[−0.68, 0.61], p = .92, ωp2 = .00; belonging: b = −0.22, SE =
0.26, 95% CI = [−0.72, 0.28], p = .39, ωp2 = .00). Neverthe-
less, given our preregistered analytic strategy, we exam-
ined the effect of condition separately for female and
male participants for the three outcome variables. The
condition differences were statistically significant for all
three outcomes among both women and men (ps < .001).
Numerically, the high- versus low-MCC differences were
greater among women than among men for interest
(women: b = −3.12, SE = 0.29, 95% CI = [−3.69, −2.55],
ωp2 = .29; men: b = −2.16, SE = 0.33, 95% CI = [−2.82,
−1.51], ωp2 = .13) and belonging (women: b = −1.84, SE =
0.17, 95% CI = [−2.17, −1.51], ωp2 = .30; men: b = −1.62,
SE = 0.19, 95% CI = [−2.00, −1.24], ωp2 = .21) but were
virtually identical for impostor feelings (women: b = 1.46,
SE = 0.22, 95% CI = [1.03, 1.89], ωp2 = .14; men: b = 1.50,
SE = 0.25, 95% CI = [1.01, 1.98], ωp2 = .12). The results
reported here did not change appreciably when we
adjusted for the estimated percentage of female employ-
ees in the company and participants’ perceptions of the
company’s emphasis on brilliance (see Table S23 in the
supplemental material at https://osf.io/fsven/).
From a practical, intervention-focused perspective,
we might also ask whether the outcomes for women
and men were more similar when the brilliance-
oriented company was portrayed as having low levels
of an MCC. With respect to interest, female participants
reported lower scores than male participants in the
high-MCC condition (b = −0.72, SE = 0.31, 95% CI =
[−1.32, −0.12], p = .020, ωp2 = .02) but not in the low-
MCC condition (b = 0.23, SE = 0.32, 95% CI = [−0.40,
0.86], p = .46, ωp2 = .00; see Fig. 6a). With respect to
anticipated impostor feelings, gender differences were
not significant in either condition (high-MCC condition:
b = 0.38, SE = 0.23, 95% CI = [−0.07, 0.83], p = .10, ωp2 =
.01; low-MCC condition: b = 0.41, SE = 0.24, 95% CI =
[−0.05, 0.88], p = .083, ωp2 = .01; see Fig. 6b). In both
cases, women anticipated numerically stronger impos-
tor feelings than men. With respect to anticipated sense
of belonging, female participants reported lower scores
than male participants in the high-MCC condition (b =
−0.38, SE = 0.18, 95% CI = [−0.73, −0.03], p = .033,
ωp2 = .01) but not in the low-MCC condition (b = −0.16,
SE = 0.18, 95% CI = [−0.52, 0.20], p = .38, ωp2 = .00; see
Fig. 6c). Similar to the results found in Study 2, the
effects of MCC condition on interest and sense of
belonging remained significant when we adjusted for
impostor feelings (see Table S24 in the supplemental
material at https://osf.io/fsven/).
Table 3. Results of Regression Models Predicting Downstream Outcomes (Interest in
Company, Anticipated Impostor Feelings, and Anticipated Sense of Belonging) as a
Function of Participant Gender and Masculinity-Contest Culture (MCC) Condition in
Study 3
Dependent variable and predictor b SE p ωp2
Interest in company
Participant gendera−0.26 [−0.69, 0.17] 0.22 .241 .001
MCC conditionb−2.70 [−3.13, −2.27] 0.22 < .001 .36
Participant Gender × MCC Condition −0.95 [−1.82, −0.09] 0.44 .031 .01
Anticipated impostor feelings
Participant gendera0.39 [0.07, 0.72] 0.16 .018 .02
MCC conditionb1.48 [1.15, 1.80] 0.16 < .001 .23
Participant Gender × MCC Condition −0.03 [−0.68, 0.61] 0.33 .916 .00
Anticipated sense of belonging
Participant gendera−0.27 [−0.53, −0.02] 0.13 .032 .01
MCC conditionb−1.75 [−1.99, −1.50] 0.13 < .001 .41
Participant Gender × MCC Condition −0.22 [−0.72, 0.28] 0.25 .391 .00
Note: Predictors were mean-centered to facilitate interpretation of the coefficients. The coefficients of
participant gender and MCC condition in this table can be interpreted as one would interpret the main
effects in an analysis-of-variance table. These results did not change appreciably when we adjusted
for the estimated percentage of female employees in the company and participants’ perceptions of the
company’s emphasis on brilliance (see Table S23 in the supplemental material at https://osf.io/fsven/).
Values in brackets are 95% confidence intervals.
aParticipant gender was coded 0 for man and 1 for woman. bMCC condition was coded 0 for low and 1
for high.

Psychological Science, XX(X) 15
Discussion
When a brilliance-oriented company was said to display
high (vs. low) levels of an MCC, participants showed
lower interest in joining the company and were more
likely to anticipate feeling like impostors who would
not belong. These findings provide evidence for the
second link in our hypothesized causal pathway—
namely, that MCCs undermine interest and expected
well-being. These links were only partially moderated
by gender, consistent with what we observed in Studies
1 and 2, where the relationship between perceptions
of workplace norms and psychological outcomes was
similar in magnitude for women and men. Of relevance
to future interventions, results showed that when par-
ticipants were led to believe that a brilliance-oriented
company was not characterized by an MCC, women
were just as interested in this company as men were
and anticipated similar levels of belonging.
General Discussion
We proposed that contexts in which brilliance is prized
can be unwelcoming for women because the emphasis
on brilliance—a stereotypically male trait that is viewed
as relatively fixed—fosters the perception of an MCC.
The results of three preregistered studies (and a prereg-
istered pilot study) provided support for this proposal
in the context of academia (pilot study and Study 1) and
in a hypothetical industry context (Studies 2 and 3).
These findings contribute to the growing literature
on field-specific ability beliefs by identifying a mecha-
nism through which an emphasis on brilliance under-
mines gender diversity—namely, the perception of a
negative work environment. Although the effect sizes
in our studies were somewhat variable and may not
generalize to academic or professional contexts outside
of the United States, the results as a whole indicate that
perceptions of an MCC may play a key role in the
maintenance of gender disparities in brilliance-focused
domains. Indeed, countering the perception of a mas-
culine cultural ethos might be an effective way to
increase the participation of women in domains in
which brilliance is prized (Study 3).
Our results also suggest that the gender composition
of brilliance-focused contexts might not, by itself, be
what makes these environments unwelcoming to
women. In our studies, beliefs about what is valued in
a context were consequential beyond the estimated
gender ratios. Thus, it may be possible to foster a more
inclusive environment in brilliance-oriented contexts
by changing the work culture even if current gender
ratios are still imbalanced. Similarly, it does not seem
that an internalized lack of self-confidence was driving
a
Low MCC High MCC
Women Men Women Men
1
2
3
4
5
6
7
Impostor Feelings
b
Women Men Women Men
1
2
3
4
5
6
9
7
8
Interest
c
Low MCC High MCC
Low MCC High MCC
Women Men Women Men
1
2
3
4
5
6
7
Sense of Belonging
Fig. 6. Effects of a masculinity-contest culture (MCC; high vs. low)
and participant gender on (a) interest, (b) anticipated impostor feel-
ings, and (c) anticipated sense of belonging in Study 3. Each dot
represents an individual participant’s response. Within each box plot,
the solid line in the middle represents the median, and the diamond
represents the mean. The bottom and top edges of the box indicate
the 25th and 75th percentiles, respectively, and whiskers extend 1.5
times the interquartile range.

16 Vial et al.
the negative effects of brilliance-focused contexts, for
either women or men: We found the expected down-
stream effects on sense of belonging (Studies 1–3) and
interest (Studies 2 and 3) through perceptions of an
MCC even when we adjusted for participants’ impostor
feelings, which indicates that perceptions of the cul-
ture—rather than low confidence—drove the negative
effects of an emphasis on brilliance.
In future investigations, it will be worthwhile to
examine objective markers of an MCC (rather than
individual perceptions) and to investigate precisely
why emphasizing brilliance leads the negative ele-
ments of stereotypic masculinity (e.g., dominance,
competition) to become the norm. In particular, it
would be informative to disentangle the role of stereo-
typically masculine standards (i.e., the association of
brilliance with men; Cheryan & Markus, 2020) from the
role of fixed mindsets (i.e., the tendency to view bril-
liance as innate and unchangeable; Rattan etal., 2012).
Future research should also explore why women have
a lower threshold for anticipating an undesirable work-
place culture from an emphasis on brilliance. Although
we found that both men and women perceive MCCs
in contexts that emphasize brilliance, women also
appeared to be more sensitive to this connection
(Study 2). To some extent, this lower threshold seems
to be a function of women’s prior experiences (e.g.,
compare nonacademic and academic women’s percep-
tions of academia in the pilot study and Study 1,
respectively), but which aspects of experience are rel-
evant remains to be determined.
In summary, an emphasis on brilliance leads indi-
viduals to perceive an environment characterized by a
competitive struggle for intellectual dominance. Women
seem particularly attuned to this link, and because per-
ceiving such an environment is generally demotivating,
professions in which brilliance is prized continue to
confront gender gaps.
Transparency
Action Editor: Kate Ratliff
Editor: Patricia J. Bauer
Author Contributions
A. C. Vial, A. Cimpian, and G. E. Newman developed the
research concept. A. C. Vial and A. Cimpian designed the
studies. A. C. Vial collected the data for all studies, with
the collaboration of M. Muradoglu for Study 2. A. C. Vial
and A. Cimpian analyzed the data, and M. Muradoglu
created the figures. A. C. Vial drafted the manuscript, and
M. Muradoglu, G. E. Newman, and A. Cimpian provided
critical revisions. All the authors approved the final manu-
script for submission.
Declaration of Conflicting Interests
The author(s) declared that there were no conflicts of
interest with respect to the authorship or the publication
of this article.
Funding
This work was funded in part by U.S. National Science
Foundation Grant BCS-1733897 awarded to A. Cimpian.
Open Practices
All data, analysis code, and materials have been made
publicly available via OSF and can be accessed at https://
osf.io/92vn6/. The design and analysis plans for all studies
were preregistered on AsPredicted (copies can be found
at https://osf.io/92vn6/). This article has received the
badges for Open Data, Open Materials, and Preregistration.
More information about the Open Practices badges can be
found at http://www.psychologicalscience.org/publica
tions/badges.
TC
ORCID iDs
Andrea C. Vial https://orcid.org/0000-0001-9367-2081
Melis Muradoglu https://orcid.org/0000-0002-5162-8741
Andrei Cimpian https://orcid.org/0000-0002-3553-6097
Acknowledgments
We thank the members of the Cognitive Development Lab at
New York University for helpful comments on previous drafts
of this article.
Notes
1. Unless otherwise noted, all analyses were conducted in Stata
Version 16 (StataCorp, 2019).
2. In all studies, we also examined the potential moderating
roles of participant race and (when relevant) the role of partici-
pant exposure to the different academic fields. There were no
interactions with these variables, and the relationship between
perceived emphasis on brilliance and perceived MCC remained
significant in all models (see Tables S5, S12, S19, and S25 in the
supplemental material at https://osf.io/fsven/).
References
Albers, C., & Lakens, D. (2018). When power analyses based
on pilot data are biased: Inaccurate effect size estima-
tors and follow-up bias. Journal of Experimental Social
Psychology, 74, 187–195. https://doi.org/10.1016/j.jesp
.2017.09.004
Association of American Medical Colleges. (2018). Faculty
roster: U.S. medical school faculty. https://www.aamc
.org/data-reports/faculty-institutions/interactive-data/
data-reports/faculty-institutions/interactive-data/2018-
us-medical-school-faculty
Baron-Cohen, S. (2002). The extreme male brain theory
of autism. Trends in Cognitive Sciences, 6(6), 248–254.
https://doi.org/10.1016/s1364-6613(02)01904-6
Ben-Shachar, M., Lüdecke, D., & Makowski, D. (2020). effect-
size: Estimation of effect size indices and standardized
parameters. Journal of Open Source Software, 5(56),
Article 2815. https://doi.org/10.21105/joss.02815
Berdahl, J. L., Cooper, M., Glick, P., Livingston, R. W., &
Williams, J. C. (2018). Work as a masculinity contest.

Psychological Science, XX(X) 17
Journal of Social Issues, 74(3), 422–448. https://doi.org/10
.1111/josi.12289
Bian, L., Leslie, S. J., & Cimpian, A. (2017). Gender stereo-
types about intellectual ability emerge early and influence
children’s interests. Science, 355(6323), 389–391. https://
doi.org/10.1126/science.aah6524
Bian, L., Leslie, S. J., Murphy, M. C., & Cimpian, A. (2018).
Messages about brilliance undermine women’s interest
in educational and professional opportunities. Journal of
Experimental Social Psychology, 76, 404–420. https://doi
.org/10.1016/j.jesp.2017.11.006
Billington, J., Baron-Cohen, S., & Wheelwright, S. (2007).
Cognitive style predicts entry into physical sciences
and humanities: Questionnaire and performance tests
of empathy and systemizing. Learning and Individual
Differences, 17(3), 260–268. https://doi.org/10.1016/
j.lindif.2007.02.004
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s
Mechanical Turk: A new source of inexpensive, yet high-
quality, data? Perspectives on Psychological Science, 6(1),
3–5. https://doi.org/10.1177/1745691610393980
Canning, E. A., Murphy, M. C., Emerson, K. T. U., Chatman,
J. A., Dweck, C. S., & Kray, L. J. (2020). Cultures of
genius at work: Organizational mindsets predict cultural
norms, trust, and commitment. Personality and Social
Psychology Bulletin, 46(4), 626–642. https://doi.org/10
.1177/0146167219872473
Cheryan, S., & Markus, H. R. (2020). Masculine defaults:
Identifying and mitigating hidden cultural biases.
Psychological Review, 127(6), 1022–1052. https://doi.org/
10.1037/rev0000209
Cheryan, S., Ziegler, S. A., Montoya, A. K., & Jiang, L. (2017).
Why are some STEM fields more gender balanced than
others? Psychological Bulletin, 143(1), 1–35. http://doi
.org/10.1037/bul0000052
Clance, P. R., & Imes, S. A. (1978). The imposter phenom-
enon in high achieving women: Dynamics and thera-
peutic intervention. Psychotherapy: Theory, Research &
Practice, 15(3), 241–247. https://doi.org/10.1037/h00
86006
Deiglmayr, A., Stern, E., & Schubert, R. (2019). Beliefs in
“brilliance” and belonging uncertainty in male and female
STEM students. Frontiers in Psychology, 10, Article 1114.
https://doi.org/10.3389/fpsyg.2019.01114
Dweck, C. S. (2008). Mindset: The new psychology of success.
Random House.
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009).
Statistical power analyses using G*Power 3.1: Tests for
correlation and regression analyses. Behavior Research
Methods, 41, 1149–1160. https://doi.org/10.3758/BRM.41
.4.1149
Field, A. (2013). Discovering statistics using IBM SPSS statis-
tics. SAGE.
Glick, P., Berdahl, J. L., & Alonso, N. M. (2018). Development
and validation of the Masculinity Contest Culture scale.
Journal of Social Issues, 74(3), 449–476. https://doi.org/
10.1111/josi.12280
Good, C., Rattan, A., & Dweck, C. S. (2012). Why do women
opt out? Sense of belonging and women’s representa-
tion in mathematics. Journal of Personality and Social
Psychology, 102(4), 700–717. https://doi.org/10.1037/
a0026659
Gunderson, E. A., Ramirez, G., Levine, S. C., & Beilock, S. L.
(2012). The role of parents and teachers in the develop-
ment of gender-related math attitudes. Sex Roles, 66(3),
153–166. https://doi.org/10.1007/s11199-011-9996-2
Hayes, A. F. (2015). An index and test of linear moderated
mediation. Multivariate Behavioral Research, 50(1), 1–22.
https://doi.org/10.1080/00273171.2014.962683
Heatherington, L., Daubman, K. A., Bates, C., Ahn, A., Brown,
H., & Preston, C. (1993). Two investigations of “female
modesty” in achievement situations. Sex Roles, 29(11),
739–754. https://doi.org/10.1007/BF00289215
Heilman, M. E. (1979). High school students’ occupational
interest as a function of projected sex ratios in male-dom-
inated occupations. Journal of Applied Psychology, 64(3),
275–279. https://doi.org/10.1037/0021-9010.64.3.275
Holman, L., & Morandin, C. (2019). Researchers collabo-
rate with same-gendered colleagues more often than
expected across the life sciences. PLOS ONE, 14(4), Article
e0216128. https://doi.org/10.1371/journal.pone.0216128
Kupers, K. A. (2005). Toxic masculinity as a barrier to mental
health treatment in prison. Journal of Clinical Psychology,
61(6), 713–724. https://doi.org/10.1002/jclp.20105
Leslie, S. J., Cimpian, A., Meyer, M., & Freeland, E. (2015).
Expectations of brilliance underlie gender distributions
across academic disciplines. Science, 347(6219), 262–265.
https://doi.org/10.1126/science.1261375
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of
a feather: Homophily in social networks. Annual Review
of Sociology, 27(1), 415–444. https://doi.org/10.1146/
annurev.soc.27.1.415
Meyer, M., Cimpian, A., & Leslie, S. J. (2015). Women are
underrepresented in fields where success is believed to
require brilliance. Frontiers in Psychology, 6, Article 235.
https://doi.org/10.3389/fpsyg.2015.00235
Muradoglu, M., Horne, Z., Hammond, M. D., Leslie, S.-J.,
& Cimpian, A. (2021). Women—particularly underrep-
resented minority women—and early-career academics
feel like impostors in fields that value brilliance. Journal
of Educational Psychology. Advance online publication.
https://doi.org/10.1037/edu0000669
National Science Foundation. (2019). Doctorate recipients
from U.S. universities: 2018 (Special report NSF 20-301).
https://ncses.nsf.gov/pubs/nsf20301/
Prentice, D. A., & Carranza, E. (2002). What women and men
should be, shouldn’t be, are allowed to be, and don’t have
to be: The contents of prescriptive gender stereotypes.
Psychology of Women Quarterly, 26(4), 269–281. https://
doi.org/10.1111/1471-6402.t01-1-00066
Rattan, A., Savani, K., Naidu, N. V. R., & Dweck, C. S. (2012).
Can everyone become highly intelligent? Cultural differ-
ences in and societal consequences of beliefs about the
universal potential for intelligence. Journal of Personality
and Social Psychology, 103(5), 787–803. https://doi.org/
10.1037/a0029263
Reid, E. M., O’Neill, O. A., & Blair-Loy, M. (2018). Masculinity
in male-dominated occupations: How teams, time, and
tasks shape masculinity contests. Journal of Social Issues,
74(3), 579–606. https://doi.org/10.1111/josi.12285

18 Vial et al.
Simon, M., & Choi, Y. J. (2018). Using factor analysis to vali-
date the Clance Impostor Phenomenon Scale in sample
of science, technology, engineering and mathematics
doctoral students. Personality and Individual Differences,
121, 173–175. https://doi.org/10.1016/j.paid.2017.09.039
Simpkins, S. D. (2015). The role of parents in the ontog-
eny of achievement-related motivation and behavioral
choices. Monographs of the Society for Research in Child
Development, 80(2), 1–151. https://doi.org/10.1111/mono
.12156
Spence, J. T., Helmreich, R. L., & Holahan, C. K. (1979).
Negative and positive components of psychological mas-
culinity and femininity and their relationships to self-
reports of neurotic and acting out behaviors. Journal
of Personality and Social Psychology, 37(10), 1673–1682.
https://doi.org/10.1037/0022-3514.37.10.1673
StataCorp. (2019). Stata statistical software (Version 16)
[Computer software]. https://www.stata.com/
Storage, D., Charlesworth, T. E. S., Banaji, M. R., & Cimpian, A.
(2020). Adults and children implicitly associate brilliance
with men more than women. Journal of Experimental
Social Psychology, 90, Article 104020. https://doi.org/
10.1016/j.jesp.2020.104020
Storage, D., Horne, Z., Cimpian, A., & Leslie, S. J. (2016).
The frequency of “brilliant” and “genius” in teaching
evaluations predicts the representation of women and
African Americans across fields. PLOS ONE, 11(3), Article
e0150194. https://doi.org/10.1371/journal.pone.0150194
Williams, M. J., & Tiedens, L. Z. (2016). The subtle suspension of
backlash: A meta-analysis of penalties for women’s implicit
and explicit dominance behavior. Psychological Bulletin,
142(2), 165–197. https://doi.org/10.1037/bul0000039