Single-dose testosterone administration increases men’s preference for status goods

Abstract

In modern human cultures where social hierarchies are ubiquitous, people typically signal their hierarchical position through consumption of positional goods—goods that convey one’s social position, such as luxury products. Building on animal research and early correlational human studies linking the sex steroid hormone testosterone with hierarchical social interactions, we investigate the influence of testosterone on men’s preferences for positional goods. Using a placebo-controlled experiment (N = 243) to measure individuals’ desire for status brands and products, we find that administering testosterone increases men’s preference for status brands, compared to brands of similar perceived quality but lower perceived status. Furthermore, testosterone increases positive attitudes toward positional goods when they are described as status-enhancing, but not when they are described as power-enhancing or high in quality. Our results provide novel causal evidence for the biological roots of men’s preferences for status, bridging decades of animal behavioral studies with contemporary consumer research.

Full text

ARTICLE
Single-dose testosterone administration increases
men’s preference for status goods
G. Nave1, A. Nadler 2, D. Dubois3, D. Zava4, C. Camerer5& H. Plassmann 3,6
In modern human cultures where social hierarchies are ubiquitous, people typically signal
their hierarchical position through consumption of positional goods—goods that convey one’s
social position, such as luxury products. Building on animal research and early correlational
human studies linking the sex steroid hormone testosterone with hierarchical social inter-
actions, we investigate the influence of testosterone on men’s preferences for positional
goods. Using a placebo-controlled experiment (N=243) to measure individuals’desire for
status brands and products, we find that administering testosterone increases men’s pre-
ference for status brands, compared to brands of similar perceived quality but lower per-
ceived status. Furthermore, testosterone increases positive attitudes toward positional goods
when they are described as status-enhancing, but not when they are described as power-
enhancing or high in quality. Our results provide novel causal evidence for the biological roots
of men’s preferences for status, bridging decades of animal behavioral studies with con-
temporary consumer research.
DOI: 10.1038/s41467-018-04923-0 OPEN
1Marketing Department, The Wharton School of the University of Pennsylvania, 3730 Walnut St., JMHH #700, Philadelphia, PA 19104, USA. 2Finance
Department, Ivey Business School, Western University, 1255 Western Rd., London, ON N6G 0N1, Canada. 3Marketing Area, INSEAD, Boulevard de
Constance, 77300 Fontainebleau, France. 4ZRT Laboratory, 8605 SW Creekside Pl., Beaverton, OR 97008, USA. 5Humanities and Social Sciences Division,
California Institute of Technology, 1200 E California Blvd MC 228-77, Pasadena, CA 91125, USA. 6Social and Affective Neuroscience (SAN) Team, Institut du
Cerveau et de la Moelle épinière (ICM), Sorbonne Université, INSERM UMR 1127, CNRS UMR 7225, 75013 Paris, France. These authors contributed equally:
G. Nave, A. Nadler. Correspondence and requests for materials should be addressed to G.N. (email: gnave@wharton.upenn.edu)
or to H.P. (email: hilke.plassmann@insead.edu)
NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications 1
1234567890():,;
Content courtesy of Springer Nature, terms of use apply. Rights reserved

From schools of fish to modern human communities, social
hierarchies are ubiquitous across species1,2. Hierarchies give
rise to advantages at the group level, such as facilitating
leader–follower coordination and reducing resource conflict3.At
the individual level, higher social rank improves mating oppor-
tunities, promotes access to resources, reduces stress, and
increases social influence4–7. Therefore, individuals exert con-
siderable effort to enhance their social rank by gaining status (i.e.,
respect and admiration from others, sometimes also referred to as
prestige) and power (i.e., control over valuable resources, some-
times also referred to as dominance)8,9.
How do people achieve higher status? In early human societies,
displays of hunting skills and physical aggression were primary in
promoting one’s standing in society. In contemporary settings,
however, hunting and aggression have been replaced by different
strategies, such as displays of culturally valued skills and beha-
viors (e.g., obtaining academic degrees). Another prevalent route
to higher status rests on the display of wealth through positional
consumption10,11. This idea was introduced by Thorstein
Veblen’s seminal work, The Theory of the Leisure Class12, which
describes how wasteful expenditures on positional goods, which
display one’s apparent resources to others, shape the social strata
over time8. Such goods are particularly effective signals of status
because they separate the “haves”from the “have nots”through
economic (e.g., high price) or physical (e.g., restricted access for
private club members) barriers. Although Veblen’s insights were
overlooked by classical market theories, modern economic the-
ories began to incorporate this view by showing that a balance of
prices and goods sustains the market for costly signals13,14.
Indeed, goods that wealthier individuals gravitate toward (here-
after, “positional goods”) also tend to be more visible to others
than other goods that are more affordable and thus accessible to
everyone15,16.
Understanding the drivers of costly signaling through posi-
tional consumption is important because this behavior is, by
definition, wasteful—in the sense that less expensive goods could
have the same functional value as their high-status counterparts
(e.g., cars and houses). Status consumption therefore creates
inefficiencies. Spending resources to elevate perceived status
might, for instance, perpetuate poverty by reducing self-
investment in health and education among the poor, who
spend disproportionately more on status signals and thus sub-
stitute status signaled through consumption for long-run wealth
accumulation17–19. While recent work has explored the socio-
psychological antecedents of status-driven consumption20–22,
little is known about its biological basis, via genes, hormones, or
brain activity.
An analogy to human conspicuous consumption in animal
behavior lies in the “handicap principle”of the evolutionary
theory of sexual selection23. Many species undergo adaptations
that wastefully consume physiological resources without yielding
immediate survival benefits, such as the stag’s heavy antlers and
the peacock’s vivid train. The handicap principle explains these
adaptations as costly signals of male fitness: because only the
fittest can afford to waste resources on traits that do not directly
increase survival probability, these adaptations become reliable
indicators of fitness. Moreover, given that the proximal purpose
of such adaptations is to promote the spread of genes by
increasing attractiveness to mates, these traits must be displayed
conspicuously—hence the length of the stag’s antlers and irides-
cence of the peacock’s tail.
The male sex steroid hormone testosterone (T) is associated
with a range of male reproductive and social behaviors in non-
human and human species. In non-humans, individual differ-
ences in T levels have been linked to social rank24,25, and a
context-sensitive rise of T during the breeding season is
associated with conspicuous displays of costly signals, such as
complex courtship singing in male birds and the growth of bulky
antlers in stags26–28.
In humans, too, T levels can situationally increase in contexts
related to social rank and male reproductive behavior29, e.g., during
competitions and after winning them, in the presence of an
attractive mate, and even following acts of conspicuous consump-
tion, such as driving a sports car (vs. a family sedan)30. While early
human studies (conducted mainly among prisoners) reported cor-
relations between T and aggression31, subsequent research has
proposed that T does not increase aggression per se32,butratherthe
motivation to promote one’sstatus
24,33,34. These studies (conducted
in both males and females) showed that pharmacologically elevated
T increased generosity35, cooperation36, and honesty37,allofwhich
are pro-social non-aggressive behaviors that may promote one’s
status24. Other studies further reported intriguing correlations
between the 2D:4D digit ratio, a candidate proxy of prenatal tes-
tosterone exposure (though see ref. 38), and behavioral measures of
courtship-related consumption39,40 (a relation that was not evident
in our own data).
Building on Veblen’s theory of positional consumption, as well
as the evidence that a situational increase in T leads to rank-
promoting behaviors in animals and humans, we hypothesized
that elevated T levels would cause men to exhibit stronger pre-
ference toward goods that promote their social rank. To test this
hypothesis, we randomly administered either T (N=125) or
placebo (N=118) topical gel to 243 males, following a double-
blind administration protocol41,42. The sample size was max-
imized according to the study’s budget constraints.
Participants completed two tasks. In the first task, we showed
participants pairs of apparel brands that differed in their asso-
ciations with social rank and asked them to indicate their pre-
ferences for one or the other. The second task investigated
whether T influenced attitudes toward the same goods when they
were positioned differently. Specifically, we measured partici-
pants’attitudes toward products that were positioned either as
status-enhancing, power-enhancing, or high in quality.
The results confirmed our hypothesis: we found that partici-
pants who received T showed greater preference for brands that
were associated with higher social rank (task 1), and that T
increased positive attitudes toward goods that were positioned as
status-enhancing but not those positioned as power-enhancing or
high in quality (task 2). We thus conclude that T elevates men’s
desire to promote their social status through economic
consumption.
Results
Manipulation check. To monitor the levels of T and other hor-
mones that might influence decision making during the experi-
ment (e.g., cortisol)43, participants provided one pre-treatment
saliva sample (that also allowed us to investigate the correlation
between behavior and basal T) and three post-treatment saliva
samples, that were assayed by liquid chromatography tandem
mass spectrometry (LC-MS/MS). We observed elevated post-
treatment saliva T measurements in the T group compared to the
placebo group, providing a robust manipulation check (see
Fig. 1). There were no treatment effects on the levels of hormones
that were not expected to be influenced by it (Supplementary
Table 3) or mood (Supplementary Table 4)
Testosterone’s effect on brand preference. Participants viewed
five pairs of pretested apparel brands in a randomized, counter-
balanced order. One brand of each pair was associated with
higher social rank than the other (e.g., Calvin Klein, high vs.
Levi’s, low). Importantly, perceived social rank difference between
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0
2NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

the brands in each pair was substantially greater than perceived
difference in quality, mitigating the possibility that the latter
influenced participants’preferences in our task (see Fig. 2b and
Supplementary Table 5). Participants indicated the extent to
which they preferred one brand relative to the other using 10-
point rating scales (Fig. 2a).
A series of mixed-effects linear models that included random
intercepts for participant and brand pair examined the effect of T
–0.2
–0.1
0
0.1
0.2
Preference for high-social
rank brand (z-scored)
cb
0
10
20
30
40
50
60
70
80
Social rank Quality
Association rating
High-vs. low-rank brands:
manipulation check
High-rank brands Low-rank brands
a
Brand A
Strongly brand A Strongly brand BIndifferent
Brand B
Testosterone
Placebo
Which brand do you prefer, and to what extent?
12345678910
6
Fig. 2 Task to assess preference for brands high versus low in social rank and results. aPreference task showing the setup and main dependent variable. b
Mean social rank and quality association ratings of brands pre-classified based on a pretest as high vs. low rank, by main study participants (N=243);
perceived differences in social rank associations between the brands were substantially greater than the difference in perceived quality. cMean preference
toward the high (versus low) social rank brands for the two treatment groups (z-scored at the pair level). Error bars denote s.e.m. For the corresponding
dot plot, see Supplementary Fig. 2
Survey
& hand
scan
Sample
A
4.5
Placebo
Testosterone
3.5
2.5
Testosterone pg/mL (logged values)
1.5
1
0.5
0
Time (not to scale)
2
3
4
Sample
B
Sample
C
Sample
D
9:00 9:30 9:35 14:05 14:10 15:15 15:20 16:00 16:15
10:00-
14:00
Gel T
loading
period
Behavioral
tasks
Payout
Behavioral
tasks
Fig. 1 Experimental setup and salivary testosterone levels. Participants (N=243) arrived at the lab at 9:00 a.m., had their hands scanned to take 2D:4D
measures, completed an intake survey, and gave a baseline saliva sample (“A”) before application of either T or placebo topical gel. After a 4-h loading
period, participants came back to the lab and took part in a battery of behavioral tasks. Three additional saliva samples (“B,”“C,”and “D”) were collected
during the experiment, all of which indicated elevated T levels in the treatment group compared to the placebo group. The behavioral tasks reported were
the main focus of the study, and took place immediately following the participants’return the lab in the afternoon (after saliva sample B). Error bars denote
s.e.m.
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved

on brand preferences (with higher ratings indicating greater
preference for the high over the low social-rank brands). We
found that the T group preferred the brands with higher social
rank compared to the placebo group (standardized β=0.18, 95%
CI =[0.045 0.359], z=2.01, p=0.04; Fig. 2c and Supplementary
Table 7). The effect was robust when controlling for age, mood,
treatment expectancy, the 2D:4D digit finger ratio (a proxy of
prenatal T exposure), and post-treatment levels of all measured
hormones that were unaffected by T treatment.
Additional analyses revealed that participants’baseline (pre-
treatment) T levels were also associated with greater preference
for brands with higher social rank (β=0.13, 95% CI =[0.042
0.220], z=2.867, p=0.004; see Supplementary Table 12; the
treatment effects were robust to controlling for baseline T),
suggesting that the effect of T is both “activational”(i.e., behavior
is influenced by transient changes in T levels) and “dispositional”
(i.e., behavior is related to baseline T levels, which are relatively
stable when measured within individuals at the same time of day).
We found a similar effect for baseline levels of androstenedione
(z=2.53, p< 0.01), a downstream metabolite of testosterone.
Levels of androstenedione correlate with testosterone (in our
sample r=0.44, p< 0.01), and it has been suggested (by animal
studies) to have potent behavioral effects that are similar to those
of T (see Supplementary Table 12).
Finally, we investigated whether the shift toward brands high
in social rank was driven by a shift in their rank-related
associations, rather than a preference shift per se. We found that
the social rank associations of the brands used in our task were
not affected by T treatment (see Supplementary Table 6).
Testosterone’s effect on product liking. The second task inves-
tigated the effect of T on two distinct paths individuals can use
consumption to climb the social ladder: increasing status (defined
as the prestige, respect, and admiration an individual has in the
eyes of others) or power (defined as feelings of control over
valued resources)8,9. Although power and status are inextricably
intertwined in most animal social groups, the two can be
decoupled within human social contexts. For example, a political
adviser unknown to the public can have significant control over
important decisions without receiving social recognition (high
power, low status); conversely, a well-known academic may enjoy
high status and be respected by the public but have little power
over policy decisions regarding her research findings8. This
experiment provides a unique opportunity to disentangle the
extent to which T affects power- versus status-enhancement
motives. (Note that apart from previous finding suggesting a
relation between T and status seeking, there is also evidence that
circulating T levels are associated with implicit power motiva-
tion.)44,45
Participants viewed brief text descriptions of six different goods
in a randomized, counterbalanced order and then indicated their
attitudes toward each using three 10-point scales (favorable/
unfavorable, good/bad, positive/negative). For each good, we
composed and pretested three different text descriptions, identical
except for specific phrases emphasizing associations with status,
power, or high quality (Fig. 3a). Descriptions were randomized so
that each participant saw two goods in each condition (status,
power, quality) in a counterbalanced fashion. This yielded a 2
(treatment: testosterone, placebo) × 3 (association: status, power,
quality) between-participants factorial design, repeated within-
participants over six different goods (see Fig. 3and Supplemen-
tary Methods). Participants also indicated their hypothetical
purchase intentions (10-point scale) and willingness to pay
(WTP, open text entry) for the goods; these measures were
standardized and averaged to create an index for hypothetical
purchasing behavior.
A series of mixed-effects linear models that included product
and participant random intercepts and participant random slopes
for positioning conditions estimated the effect of T on product
attitudes. We found that the T group had more positive attitudes
toward goods described as status-enhancing compared to the
placebo group (reflected by a significant treatment × status
interaction, standardized β=0.275, 95% CI =[0.042 0.508], z
=2.31, p=0.02; Fig. 3c and Supplementary Tables 8–9).
However, there was no difference between the two treatment
–0.2
–0.1
0
0.1
0.2
Mean product liking (z-scored)
Testosterone Placebo
dQuality Power Status
Extreme robustness, high
precision, technology, and comfort
are part of this watch’s DNA.
Our supreme quality watches have
been a symbol of reliability in the
most demanding situations, and a
benchmark for innovation, combining
state-of-the-art swiss machinery and
strict quality control tests for centuries.
a
c
b
Watch
brand
Watch
brand
Watch
brand
Indestructibility, sport, power,
and confidence are part of this
watch’s DNA.
Our cutting-edge watches
have been a symbol of power
and athletic excellence, and a
benchmark for performance,
combining daring german design
and strive for perfection for centuries.
Prestige, artisanal spirit, luxury,
and attention to detail are part of
this watch’s DNA.
Our world-famous watches have
been a symbol of a way of life,
and a benchmark for fashionable
style, combining sophisticated
Italian design and timeless
reputation for centuries
Fig. 3 Task measuring attitudes towards identical goods associated with different rank-enhancing strategies and results. For each of six goods we created
three different text ads emphasizing either its associations with (a) high quality, (b) power, or (c) status. The ads were identical otherwise. dMean
attitudes of the goods (N=243 participants) for each of the three conditions (z-scored at the ad level). Error bars denote s.e.m. For the corresponding dot
plot, see Supplementary Fig. 3. The watch clipart was created by Professor Amos Nadler
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0
4NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

groups when the same goods were described as power-enhancing
(β=–0.016, 95% CI =–0.242 0.209], z=–0.14, p> 0.89) or high
in quality (β=–0.044, 95% CI =[–0.243 0.155], z=–0.44, p>
0.66). Results were robust to controlling for age, mood, treatment
expectancy, 2D:4D, and hormones that were unaffected by T
treatment. The results hold when using either quality or power as
the baseline category, indicating that T’s positive effect on
preferences for status was significantly greater than its effects on
preferences for either quality or power (see Supplementary
Table 11).
An analogous analysis for the secondary measure of hypothe-
tical purchasing behavior revealed a similar yet statistically
weaker pattern (see Supplementary Table 10). We found a
marginally significant treatment × status interaction (β=0.160,
95% CI =[–0.023 0.342], z=17, p<0.10). There were neither
treatment × power (β=0.02, 95% CI =[–0174 0.215], z=0.36, p
>0.80) nor treatment × quality (β=–0.042, 95% CI =[–0.206
0.123], z=–0.495, p>0.62) interactions for the intentions
measures.
As in task 1, we performed additional analyses exploring the
association between-participants’baseline (pre-treatment) T and
their attitudes toward the goods. However, we did not detect a
reliable baseline T × status interaction, though the coefficient was
positive in sign (see Supplementary Table 13; the treatment
effects were robust to controlling for baseline T).
Discussion
Taken together, these findings suggest that the consumption of
positional goods may stem, at least partly, from biological
motives. By adopting an evolutionary perspective, we contribute
to a growing body of work in economics uncovering the adaptive
function of consumption and complement the increasingly rich
nomological network around how status processes govern indi-
viduals’thoughts, feelings, and behaviors46.
Our results demonstrate for the first time that T causally
influences rank-related consumer preferences (task 1), and that
the effect is driven by status enhancement and not power motives
or inclination for high quality (task 2). These findings contribute
to a burgeoning literature shedding light on the association
between T and humans’desire for status. Of note, the bulk of
efforts to date mostly relied on behavior in economic games, in
which participants make trade-offs between monetary and social
rewards. As both reward types can promote one’s status, it is not
always clear, a priori, what behavior should increase status in
such games. For example, one study found that T administration
increased cooperation;36 a different study47 found that T reduced
cooperation. In both cases, the authors postulated that T-induced
status-seeking behavior caused the effect, by proposing opposite
influences of status-seeking behavior on cooperation. The current
study overcomes this limitation by directly measuring and
manipulating status preferences.
Our findings may be useful for generating new hypotheses
regarding contexts in which positional consumption occurs17–19.
Men experience situational elevation in T during and following
sporting events, in the presence of attractive mates, and following
meaningful life events such as graduation and divorce29,48,49. Our
results suggest that in such contexts, male consumers might be
more likely to engage in positional consumption, and might find
status-related brand communications more appealing. We hope
that our findings will guide further research exploring how con-
textual variations in real-world settings that change T levels (e.g.,
sporting events, changes in marital status) affect status-related
economic preferences.
The current work has several limitations, which represent
opportunities for future research. First, our behavioral measures
did not involve actual purchases. Although the pattern of results
for our secondary, non-incentivized index of hypothetical pur-
chase behavior in task 2 was consistent with the primary results of
T’s effect on attitudes toward positional goods, the effects were
smaller in magnitude (statistically significant only at the p< 0.10
level, and even weaker when taking into account various controls;
see Supplementary Table 10). However, attitudes toward the
target goods predicted both hypothetical purchase intentions (r
(1,446) =0.329, 95% CI =[0.282 0.374], p<0.001) and WTP (r
(1,441) =0.249, 95% CI =[0.200 0.297], p<0.001), suggesting a
link between these measures. Of note, research suggests that
hypothetical purchase intentions and WTP do not always reflect
precise value signals on a reported50 or neural51 level. Further-
more, unlike attitudes, purchase intentions and WTP might be
more subject to factors tied to unmeasured individual differences
(e.g., budget constraints). Thus, these measures might have been
too noisy to detect a statistically significant effect in our study.
Future research should further investigate T’sinfluence on
incentive-compatible economic behaviors promoting social rank,
controlling for individual differences as best as possible.
Second, to increase the ecological validity of our pharmacolo-
gical treatment, we administered T using a widely prescribed gel
at a typical daily dose (100 mg) that leads to elevated serum T
levels within the normal male physiological range41. This design
choice might have led to effects that are moderate in size (task 1:
d=0.18; task 2: d=0.28), though sufficiently large for detection
in our study. Future research should explore whether dose, timing
of exposure (i.e., the lag between T administration and the
behavioral task), and delivery type (e.g., topical, oral) might lead
to different effects on behavior.
Third, the causal connection between T and status appears to
be bi-directional. We find that changes in T causally promote
attitudes toward status-enhancing goods, and other studies have
shown that consumption of such goods changes T levels30. This
relationship suggests a T-mediated process in which status con-
sumption and T level might reinforce each other8, and calls for
further investigations of mutual influence of T and status con-
sumption over time. Moreover, given that a reliable association
between baseline T levels and consumer preferences emerged only
in the first task, and that this is the first study testing this rela-
tionship, more research is needed to establish robustness and
variation of a dispositional effect of T on consumer preferences.
Furthermore, while the effects reported here were robust when
controlling for mood, beliefs about treatment, and levels of other
hormones that might have influenced behavior, we cannot
entirely rule out that the effects of T on economic behavior were
generated by other indirect mechanisms. For example, we could
not fully control participants’behaviors during the drug loading
period (i.e., the time interval between when the T gel was applied
and when they returned to the lab).
Fourth, because the T system is sexually dimorphic, and given
that most of the behavioral literature in animals is concerned with
males, we relied on an all-male sample (the use of same-sex
participants is a common practice in the literature). It is impor-
tant to note, however, that women also engage in conspicuous
consumption, and preliminary evidence suggests that biological
factors (including hormones that relate to the menstrual cycle)
are involved52–54. As there is evidence that T promotes status-
related behaviors in females24,35,36, further research should
explore whether the effects of T on consumer preferences are
generalizable to females, while taking into account that which
brands and goods are status-enhancing is likely to differ across
sexes55.
Finally, it is important to keep in mind that cultural differences
might play a role in the biological underpinnings of status
behavior and that status signals are not universal (e.g., status of
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications 5
Content courtesy of Springer Nature, terms of use apply. Rights reserved

different occupations varies substantially across countries56).
Preliminary evidence suggests that individual cultural differences
such as self-construal may moderate the behavioral effects of T57.
Moreover, some cultures frown upon overt expression of material
status. Replicating our results in different cultures would provide
an opportunity to test whether the expression of status is uniquely
culturally concordant or catholically materialistic. Overall, future
research should aim to generalize our findings to other popula-
tions to create robust empirical foundations for the biological
basis of consumer preferences.
Methods
General procedure. Two hundred forty-three males, aged 18–55 (M =23.63, SD
=7.22), mostly (217, 89%) students at a private Southern California consortium,
participated in the study. Non-student participants were community members
from surrounding cities. For pre-screening criteria see Supplementary Methods;
detailed demographic characteristics of the two treatment groups are available on
Supplementary Table 1. The institutional review boards of Caltech and Claremont
Graduate University approved the study, all participants gave informed consent,
and no adverse events occurred.
Similar to a previously described study (42), participants arrived at the lab at 9
a.m., signed informed consent forms, and proceeded to a designated room for hand
scanning (see Supplementary Methods). Participants were then randomly assigned
to private cubicles, where they completed demographic and mood questionnaires
(see Supplementary Methods) and provided an initial saliva sample by passive
drool. Next, participants proceeded to gel application (further details below), after
which they were instructed to refrain from bathing, any activity that might cause
excessive perspiration, and direct physical contact with females before the
afternoon session; finish eating no later than 1 p.m.; and return to the lab promptly
at 1:55 p.m. well hydrated. Participants were given printed material containing
these precautions and instructions prior to dismissal.
Participants returned to the lab at 2 p.m. (no participant was late), provided a
second saliva sample, and began the behavioral experiment in the same cubicle they
had occupied in the morning. The experiment lasted approximately 2 h and
consisted of a battery of behavioral tasks, none of which included feedback about
monetary payoffs or performance. Only the final task included feedback regarding
the participants’performance relative to others. The rationale for conducting a
battery of tasks is maximizing the knowledge gained from each human participant
undergoing a pharmacological manipulation, a practice that is standard58,59. The
two tasks reported in the paper were focal and therefore were conducted at the
outset, immediately after participants’arrival at the lab in the afternoon and the
first post-treatment saliva sample. On average, participants earned $68.12 USD
(SD =$17.36) for participating in the experiment. Payout varied as a function of
their performance in some of the other tasks.
To standardize hormonal measurements among participants, we did not
randomize the order of the behavioral tasks, in similar fashion to previous
studies58,59. Following the experiment, participants completed an exit survey,
where they indicated their beliefs about the treatment they had received, using a
five-point scale, and were privately paid in cash.
Testosterone administration. Participants were escorted in groups of two to six to
a semiprivate room where a research assistant provided a small plastic cup con-
taining 10 g of clear gel and stated that it was equally likely to contain T or placebo.
The cups were filled in advance by the lab manager, who did not interact with
participants and did not reveal the contents of the cup to the assistant. The gel
contained either topical T 1% (2 × 50 mg packets Vogelxo®by Upsher-Smith) or
the volume equivalent of an inert placebo (80% alcogel, 20% Versagel®). Partici-
pants were instructed to remove the clothing from their upper bodies and apply the
entire contents of the gel container to their shoulders, upper arms, and chest, as
demonstrated by the research assistant, and were told to wait until the gel fully
dried before putting their clothes back on.
We chose to administer T using topical gel as this was the only T administration
method for which the pharmacokinetics of a single-dose administration had been
investigated at the time41. That study41 demonstrated that plasma T levels peaked
3 h after single-dose exogenous topical administration, and that T measurements
stabilized at high levels during the time window between 4 and 7 h following
administration. Therefore, we had all participants return to the lab 4.5 h after
receiving gel, when androgen levels were elevated and stable.
Saliva sampling. Each participant provided four saliva samples by passively
drooling into a plastic tube, at predetermined sampling times throughout the study:
(1) before treatment administration; (2) upon return to the lab, just prior to
starting the behavioral tasks; (3) in the middle of the behavioral tasks battery; and
(4) following the one and only task involving performance feedback, at the end of
the experiment. We used saliva samples to avoid potential stress that might be
induced by high-resolution blood drawing throughout the experimental session.
Each saliva sample was time stamped. No food or drinks were allowed into the
laboratory, and the only water given to the participants was after their third saliva
draw (an hour before the fourth and final saliva draw).
To allow robust manipulation checks and obtain control for the participant’s
biological state, we used LC-MS/MS (detection levels and precision are available in
Supplementary Table 2) to measure the following salivary steroids: estrone,
estradiol, estriol, testosterone, androstenedione, DHEA, 5-alpha DHT,
progesterone, 17OH-progesterone, 11-deoxycortisol, cortisol, cortisone, and
corticosterone. A series of one-sample Kolmogorov–Smirnov tests for conformity
to a Gaussian distribution (Supplementary Table 3) indicated that all hormonal
measurement distributions were best approximated by a Gaussian distribution
following a log-transformation, as indicated by higher p-values. Thus, all hormonal
measurements were log-transformed prior to data analysis in order to make their
distributions closer to Gaussian. We provide further technical details of the
procedure and analysis of hormonal changes following the treatment in
Supplementary Table 3.
Task 1: Testosterone’s effect on preference for brands high in social rank.Ina
pretest, we presented 184 students of a private Southern California college (with
similar demographic characteristics as our participants) with the logos of 15
familiar apparel brands in a randomized, counterbalanced order. Participants rated
each brand’s association with quality and social rank using 100-point
subscales (0 =not at all to 100 =very much). Social rank was constructed by
averaging three items related to status (status, conspicuousness, prestige60,61) and
three items related to power (power, performance, control62–64).
Importantly, the first task did not allow us to directly disentangle status
enhancement and power enhancement motives, as we could not identify (based on
the pretest data) any brand pairs that were perceived differently with respect to
their status and power associations, and thus we combined the average of the six
items to a general measure of social rank associations. Our data indicated that
brands high in social rank were typically also perceived as high in quality. However,
we were able to identify five pairs of brands for which the difference in social rank
associations was significantly greater than the difference in quality associations.
Supplementary Table 5summarizes these pairs, along with their perceived social
rank and quality associations among the experiment participants. In order to
mitigate the possibility that participants would guess the study’s purpose, the task
included an additional pair for which both brands were associated with lower social
rank (Gap vs. H&M).
In the experimental task, we presented participants with the five brand pairs in
a randomized, counterbalanced order (Fig. 2a). One brand appeared on the left side
of the screen and the other on the right (sides were randomized and
counterbalanced). Participants indicated which of the two brands they preferred
using a 10-point Likert rating scale (1 =strongly left brand, 10 =strongly right
brand). For standardization, we z-scored the ratings at the question level; all of the
results are robust to this analytical choice.
We followed the behavioral task with a survey that examined the participants’
associations with the brands used. We showed participants all brands in a
randomized order and asked them to rate their associations with quality and social
rank (i.e., power and status) using 100-point scales. We constructed a social rank
scale by averaging their power and status ratings in a similar fashion to the pretest.
This scale allowed us to examine whether T affects preferences for social-rank-
enhancing brands rather than forming social rank associations.
Using the post-experimental survey, we conducted a manipulation check
verifying that our pairs of brands differed in social rank associations more than
they differed in quality associations, for the participants of our main study
(Supplementary Table 5). Paired t-tests of the difference in difference between
social rank and quality associations showed that the differences in social rank
associations were greater than the differences in quality associations for all of the
brand pairs (all p’s < 0.001).
To rule out the potential effect of T administration on brand associations rather
than a difference in preference for these brands, we tested for effects of T on the
perceived quality and social rank associations of the brands, using two-sample t-
tests (Supplementary Table 6). We found no reliable treatment effects on any of the
brands’rank perceptions (all p’s > 0.25). Only one of the seven brands showed a
significant difference (at the α=0.05 level, uncorrected) in quality perception.
Thus, T did not influence the brands’perception among our participants.
Task 2: Testosterone’s effect on attitudes toward goods associated with
status, power, or quality. For each of six goods we composed three different text
descriptions, describing the goods as either power-enhancing, status-enhancing, or
high in quality. The descriptions included the goods’images accompanied by the
text (all ads are available online in the project’s open science framework page). We
pretested the different text descriptions in two separate online surveys (N=714
and N=744 Amazon Mechanical Turk users). Participants saw one of the three
descriptions for each of the goods in a counterbalanced randomized order and
reported to what extent the descriptions and the goods conveyed status, power, and
quality on a 10-point Likert scale (0 =not at all, 10 =very much). As in the pretest
for Task 1, respondents rated the descriptions’and goods’associations with quality,
three items related to power, and three items related to status. The pretest results
are summarized in Supplementary Fig. 1.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0
6NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

In the experimental task, we manipulated social rank (i.e., power and status)
and quality associations for identical goods and investigated whether T
administration altered attitudes toward these goods. (We included high quality as a
third condition to conceptually replicate the findings of study 1 and assess the
extent to which social rank-promoting behaviors might stem from a preference for
characteristics typically associated with high-end options, such as quality, as
opposed to deeper psychological motives directly tied to social rank promotion.
This is important because perceived quality is often influenced by price and brand
effects.)
We presented each participant with one of the three text descriptions of each
good. Each participant saw the text descriptions for all six of the goods, such that
two of the descriptions focused on quality, two on power, and two on the status
features of the goods. We randomly assigned each participant to one of three
groups that saw a different combination of goods × text description interaction (i.e.,
a third of the participants saw the status description, another third the power
description, and another third the quality description for each of the goods, such
that two out of the six descriptions were in each description condition for each
participant). This resulted in a 2 (T/placebo treatment, between participants) × 3
(description condition, between participants) factorial design repeated over six
good categories (within participant).
Participants reported their attitudes toward each good (e.g., “What is your
attitude toward Alpina watches?”) using three 10-point Likert scales
(1 =unfavorable, 10 =favorable; 1 =bad, 10 =good; 1 =negative, 10 =positive)
that were averaged to create a single attitude score. The attitude score was z-scored
at the text description level (all results are robust to this analytical choice). In
addition, we asked participants to report hypothetical purchase intentions (10-
point Likert scale) and WTP (open text entry); the two measures were z-scored and
averaged to create an index for hypothetical purchasing behavior. We found that
the attitude ratings predicted both hypothetical purchase intentions (r(1,446) =
0.329, 95% CI =[0.282 0.374], p< 0.001) and WTP (r(1,441) =0.249, 95% CI =
[0.200 0.297], p< 0.001).
To account for two sources of variance in purchase intentions, we included two
task-related questions in our post-experiment survey, after completion of the full
experimental battery. First, we asked participants whether they already owned
goods in the target category. Second, we asked participants about their general
buying intentions for the category (i.e., participants were asked “Within the next
month, how likely are you to purchase goods of the following categories?”
measured on 1–10 Likert scales). Our regression models included controls for these
two factors, both of which were highly significant predictors (p’s < 0.001) of
hypothetical purchasing behavior.
Data analysis. Data were analyzed using linear regression mixed models with
item-specific and participant-specific random effects65. All estimated models and
their detailed results across experimental tasks are available in the Supplementary
Information online.
Availability of materials and data. Materials, data, and analysis scripts are
available on the project’s Open Science Framework (OSF) page: https://osf.io/
jqmnx.
Received: 17 August 2017 Accepted: 1 June 2018
References
1. Sapolsky, R. M. The influence of social hierarchy on primate health. Science
308, 648–652 (2005).
2. Sidanius, J. & Pratto, F. Social Dominance: An Intergroup Theory of Social
Oppression and Hierarchy (Cambridge University Press, Cambridge, 1999).
3. Anderson, C. & Brown, C. E. The functions and dysfunctions of hierarchy.
Res. Organ. Behav. 30,55–89 (2010).
4. Leary, M. R., Jongman-Sereno, K. P. & Diebels, K. J. in The Psychology of
Social Status (eds Cheng, J. T., Tracy, J. L. & Anderson, C.) 159–178 (Springer,
New York, 2014).
5. Ridgeway, C. L. in Advances in Group Processes, Vol. 1(ed. Lawler, E.) 59–93
(JAI, Greenwich, CT, 1984).
6. Heffetz, O. & Frank, R. H. in Handbook of Social Economics (eds Benhabib, J.,
Bisin, A. & Jackson, M.) 69–91 (North-Holland, Amsterdam, 2008).
7. Marmot, M. Status syndrome. Significance 1, 150–154 (2004).
8. Magee, J. C. & Galinsky, A. D. Social hierarchy: the self‐reinforcing nature of
power and status. Acad. Manag. Ann. 2, 351–398 (2008).
9. Maner, J. K. & Case, C. R. Dominance and prestige: dual strategies for
navigating social hierarchies. Adv. Exp. Soc. Psychol. 54, 129–180 (2016).
10. Nelissen, R. M. A. & Meijers, M. H. C. Social benefits of luxury brands as
costly signals of wealth and status. Evol. Hum. Behav. 32, 343–355 (2011).
11. Saad, G. The Evolutionary Bases of Consumption (Lawrence Erlbaum,
Mahwah, NJ, 2007).
12. Veblen, T. The Theory of the Leisure Class (Macmillan, New York, 1899).
13. Bagwell, L. S. & Bernheim, B. D. Veblen effects in a theory of conspicuous
consumption. Am. Econ. Rev. 86, 349–373 (1996).
14. Frank, R. H. Choosing the Right Pond: Human Behavior and the Quest for
Status (Oxford University Press, New York, 1985).
15. Heffetz, O. A test of conspicuous consumption: visibility and income
elasticities. Rev. Econ. Stat. 93, 1101–1117 (2011).
16. Han, Y. J., Nunes, J. C. & Drèze, X. Signaling status with luxury goods: the role
of brand prominence. J. Mark. 74,15–30 (2010).
17. Landis, B. & Gladstone, J. J. Personality, income, and compensatory
consumption: low-income extraverts spend more on status. Psychol. Sci. 28,
1518–1520 (2017).
18. Moav, O. & Neeman, Z. Saving rates and poverty: the role of conspicuous
consumption and human capital. Econ. J. 122, 933–956 (2012).
19. Brown, P. H., Bulte, E. & Zhang, X. Positional spending and status seeking in
rural China. J. Dev. Econ. 96, 139–149 (2011).
20. Ordabayeva, N. & Chandon, P. Getting ahead of the Joneses: when equality
increases conspicuous consumption among bottom-tier consumers. J.
Consum. Res.38,27–41 (2011).
21. Griskevicius, V., Tybur, J. M. & Van den Bergh, B. Going green to be seen:
status, reputation, and conspicuous conservation. J. Pers. Soc. Psychol. 98,
392–404 (2010).
22. Sundie, J. M. et al. Peacocks, Porsches, and Thorstein Veblen: conspicuous
consumption as a sexual signaling system. J. Pers. Soc. Psychol. 100, 664–680
(2011).
23. Zahavi, A. Mate selection—a selection for a handicap. J. Theor. Biol. 53,
205–214 (1975).
24. Eisenegger, C., Haushofer, J. & Fehr, E. The role of testosterone in social
interaction. Trends Cogn. Sci. 15, 263–271 (2011).
25. Mazur, A. & Booth, A. Testosterone and dominance in men. Behav. Brain Sci.
21.3, 353–363 (1998).
26. Lincoln, G. A., Guinness, F. & Short, R. V. The way in which testosterone
controls the social and sexual behavior of the red deer stag (Cervus elaphus).
Horm. Behav. 3, 375–396 (1972).
27. Kern, M. D. & King, J. R. Testosterone-induced singing in female white-
crowned sparrows. Condor 74, 204–209 (1972).
28. Folstad, I. & Karter, A. J. Parasites, bright males, and the immunocompetence
handicap. Am. Nat. 139, 603–622 (1992).
29. Zilioli, S. & Bird, B. M. Functional significance of men’s testosterone reactivity
to social stimuli. Front. Neuroendocrinol. 47,1–18 (2017).
30. Saad, G. & Vongas, J. G. The effect of conspicuous consumption on men’s
testosterone levels. Organ. Behav. Hum. Dec. 110,80–92 (2009).
31. Dabbs, J. M., Carr, T. S., Frady, R. L. & Riad, J. K. Testosterone, crime, and
misbehavior among 692 male prison inmates. Pers. Indiv. Differ. 18, 627–633
(1995).
32. Vongas, J. G. & Al Hajj, R. The effects of competition and implicit power
motive on men’s testosterone, emotion recognition, and aggression. Horm.
Behav. 92,57–71 (2017).
33. Stanton, S. J. The role of testosterone and estrogen in consumer behavior and
social and economic decision making: a review. Horm. Behav. 92, 155–163
(2016).
34. Dreher, J.-C. et al. Testosterone causes both prosocial and antisocial status-
enhancing behaviors in human males. Proc. Natl. Acad. Sci. USA 113,
11633–11638 (2016).
35. Eisenegger, C., Naef, M., Snozzi, R., Heinrichs, M. & Fehr, E. Prejudice and
truth about the effect of testosterone on human bargaining behaviour. Nature
463, 356–359 (2010).
36. van Honk, J., Montoya, E. R., Bos, P. A., van Vugt, M. & Terburg, D. New
evidence on testosterone and cooperation/Eisenegger et al. reply. Nature 485,
E4–E5 (2012).
37. Wibral, M., Dohmen, T., Klingmüller, D., Weber, B. & Falk, A. Testosterone
administration reduces lying in men. PLoS ONE 7, e46774 (2012).
38. Berenbaum, S. A., Bryk, K. K., Nowak, N., Quigley, C. A. & Moffat, S. Fingers
as a marker of prenatal androgen exposure. Endocrinology 150, 5119–5124
(2009).
39. Nepomuceno, M., Saad, G., Stenstrom, E., Mendenhall, Z. & Iglesias, F.
Testosterone and gift giving: mating confidence moderates the association
between digit ratios (2D4D and rel2) and erotic gift giving. Pers. Indiv. Differ.
91,27–30 (2016).
40. Nepomuceno, M., Saad, G., Stenstrom, E., Mendenhall, Z. & Iglesias, F.
Testosterone at your fingertips: digit ratios (2D:4D and rel2) as predictors of
courtship-related consumption intended to acquire and retain mates. J.
Consum. Psychol. 26, 231–244 (2016).
41. Eisenegger, C., von Eckardstein, A., Fehr, E. & von Eckardstein, S.
Pharmacokinetics of testosterone and estradiol gel preparations in healthy
young men. Psychoneuroendocrinology 38, 171–178 (2013).
42. Nave, G., Nadler, A., Zava, D. & Camerer, C. Single-dose testosterone
administration impairs cognitive reflection in men. Psychol. Sci. 28,
1398–1407 (2017).
NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0 ARTICLE
NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications 7
Content courtesy of Springer Nature, terms of use apply. Rights reserved

43. Margittai, Z. et al. Exogenous cortisol causes a shift from deliberative to
intuitive thinking. Psychoneuroendocrinology 64, 131–135 (2016).
44. Schultheiss, O. C., Wirth, M. M. & Stanton, S. J. Effects of affiliation and
power motivation arousal on salivary progesterone and testosterone. Horm.
Behav. 46, 592–599 (2004).
45. Schultheiss, O. C. et al. Effects of implicit power motivation on men’s and
women’s implicit learning and testosterone changes after social victory or
defeat. J. Pers. Soc. Psychol. 88, 174–188 (2005).
46. Saad, G. On the method of evolutionary psychology and its applicability to
consumer research. J. Mark. Res. 54, 464–477 (2016).
47. Wright, N. D. et al. Testosterone disrupts human collaboration by increasing
egocentric choices. Proc. Biol. Sci. 279, 2275–2280 (2012).
48. Gettler, L. T., McDade, T. W., Feranilc, A. B. & Kuzawa, C. W. Longitudinal
evidence that fatherhood decreases testosterone in human males. Proc. Natl.
Acad. Sci. USA 108, 16194–16199 (2011).
49. Ronay, R. & von Hippel, W. The presence of an attractive woman elevates
testosterone and physical risk taking in young men. Soc. Psychol. Pers. Sci. 1,
57–64 (2010).
50. Miller, K. M., Hofstetter, R. & Krohmer, H. How should consumers’
willingness to pay be measured? An empirical comparison of state-of-the-art
approaches. J. Mark. Res 48, 172–184 (2011).
51. Kang, M. J., Rangel, A., Camus, M. & Camerer, C. F. Hypothetical and real
choice differentially activate common valuation areas. J. Neurosci. 31, 461–468
(2011).
52. Durante, K. M., Griskevicius, V., Cantú, S. M. & Simpson, J. A. (2013). Money,
status, and the ovulatory cycle. J. Mark. Res 51,27–39 (2014).
53. Durante, K. M., Griskevicius, V., Hill, S. E., Perilloux, C. & Li, N. P. Ovulation,
female competition, and product choice: hormonal influences on consumer
behavior. J. Consum. Res. 37, 921–934 (2011).
54. Saad, G. & Stenstrom, E. Calories, beauty, and ovulation: the effects of the
menstrual cycle on food and appearance-related consumption. J. Consum.
Psychol. 22, 102–113 (2012).
55. O’Cass, A. & McEwen, H. Exploring consumer status and conspicuous
consumption. J. Consum. Behav. 4,25–39 (2004).
56. Treiman, D. J. Occupational Prestige in Comparative Perspective (Academic
Press, New York, NY, 1977).
57. Welker, K. M. et al. Preliminary evidence that testosterones association with
aggression depends on self-construal. Horm. Behav. 92, 117–127 (2016).
58. Zethraeus, N. et al. A randomized trial of the effect of estrogen and
testosterone on economic behavior. Proc. Natl. Acad. Sci. USA 106, 6535–6538
(2009).
59. Kocoska-Maras, L. et al. A randomized trial of the effect of testosterone and
estrogen on verbal fluency, verbal memory, and spatial ability in healthy
postmenopausal women. Fertil. Steril. 95, 152–157 (2011).
60. Dubois, D. & Ordabayeva, N. in The Cambridge Handbook of Consumer
Psychology (eds Norton, M. I., Rucker, D. D. & Lamberton, C.) 332–367
(Cambridge University Press, Cambridge, UK, 2015).
61. Anderson, C., Hildreth, J. A. D. & Howland, L. Is the desire for status a
fundamental human motive? A review of the empirical literature. Psychol.
Bull. 141, 574–601 (2015).
62. Salancik, G. R. & Pfeffer, J. Who gets power—and how they hold on to it: a
strategic-contingency model of power. Organ. Dyn. 5,3–21 (1977).
63. Fiske, S. T. in Handbook of Social Psychology Vol. 2 (eds Fiske, S. T., Gilbert,
D. T. & Lindzey, G.) 941–982 (Wiley, New York, 2010) .
64. Keltner, D., Gruenfeld, D. H. & Anderson, C. Power, approach, and inhibition.
Psychol. Rev. 110, 265–284 (2003).
65. Barr, D. J., Levy, R., Scheepers, C. & Tily, H. J. Random effects structure for
confirmatory hypothesis testing: keep it maximal. J. Mem. Lang. 68, 255–278
(2013).
Acknowledgements
We would like to acknowledge funding for this research project from INSEAD Research
and Development funds to H.P. and D.D., the MacArthur Foundation, Ivey Business
School, IFREE, the Russell Sage Foundation, the Wharton Neuroscience Initiative and
the Wharton-INSEAD alliance. Special thanks to Jorge Barraza, Austin Henderson, and
Garrett Thoelen for research assistance.
Author contributions
G.N., A.N., D.D., D.Z., C.C., and H.P conceived and designed the study and wrote the
manuscript. G.N. and A.N. conducted the experiments and analyzed the data. G.N. and
A.N contributed equally to this work.
Additional information
Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467-
018-04923-0.
Competing interests: The authors declare no competing interests.
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
© The Author(s) 2018
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-04923-0
8NATURE COMMUNICATIONS | (2018) 9:2433 | DOI: 10.1038/s41467-018-04923-0 | www.nature.com/naturecommunications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com