Content uploaded by Martha Newson
Author content
All content in this area was uploaded by Martha Newson on Apr 30, 2020
Content may be subject to copyright.
SOCCER & CORTISOL
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1002/smi.2924
Devoted fans release more cortisol when watching live soccer matches
Newson, Martha1*, Shiramizu, Victor2, Buhrmester, Michael1, Hattori, Wallisen3, Jong, Jon4,
Yamamoto, Emilia5, Whitehouse, Harvey1
1) Centre for Anthropology and Mind, University of Oxford, Oxford, UK
2) Institute of Neuroscience & Psychology, University of Glasgow, UK
3) Department of Public Health, Federal University of Uberlândia, Uberlândia ,Brazil
4) Belief, Brain, and Behaviour Lab, Coventry University, Coventry, UK
5) Psychobiology Graduation Program, Federal University of Rio Grande do Norte, Natal, Brazil
* Corresponding author
Martha Newson
ICEA, School of Anthropology & Museum Ethnography
51-53 Banbury Road,
Oxford
OX2 6PE
Email: Martha.newson@anthro.ox.ac.uk
Phone: +44 7845 044 291
Running Head: SOCCER & CORTISOL
Acknowledgements
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Thank you to previous reviewers for their feedback and our team of hard-working research
assistants at UFRN, Natal, Brazil.
Author contributions
M.N., HW, JJ, and EY designed the experiment. M.N., V.S., and W.H conducted the experiment.
M.N, V.S., W.H., and M.B. ran analyses. M.N., H.W., J.J., and M.B wrote the paper.
There are no conflicts of interest.
The full dataset will be available on the Open Science Framework when the article is published.
Devoted fans release more cortisol when watching live soccer matches
Why do some sports fans exhibit such strong emotions when watching live matches? Identity fusion
is a strong form of group alignment in which personal and group identities synergise to produce a
visceral sense of ‘oneness’ with one’s team. This paper investigates its role using a three-item state
measure with high internal validity in elevating salivary cortisol levels during intense group events
(n = 41). Our evidence was gathered at field laboratories during the 2014 soccer (football) World
Cup in Natal, Brazil, with live screenings of two Brazilian victories (Colombia, 2-1; Chile, 1-1 with
penalties), and the historic semi-final loss to Germany (1-7). We replicated previous studies
showing that salivary cortisol concentrations fluctuate during live football events and are related to
group membership, and extending them by suggesting that identity fusion is even more strongly
related to cortisol concentrations than identification. There was an interaction between match
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
outcome and cortisol, such that watching a loss, i.e. dysphoria, was associated with particularly
high cortisol concentrations. While women were more fused to the team than men, there were no
other gender effects. Taken together, these findings suggest that identity fusion modulates
physiological reactivity, resulting in distinct psycho-physiological profiles during stressful events.
Keywords: Identity fusion, stress response system, soccer, football, cortisol, physiological profiles
1 Introduction
Attending live sports matches at stadia or public viewing events can produce intense shared
emotions, commonly expressed by ritualised chanting and singing, synchronous stamping and
clapping, and even inter-group violence (Brown, 2007; Giulianotti, 1995; King, 2002, 2003;
Pearson, 2012; Stone, 2007; Stroeken, 2002; Weed, 2007). Soccer (or football) is the most watched
sport in the world, with nearly half the globe – 3.2 billion viewers - tuning in for the 2014 FIFA
World Cup (FIFA, 2015). But why do so many people tune in to football, and what are the effects
of watching a live sporting event? Recent research into international soccer has identified a number
of physiological responses associated with spectatorship, including: heightened cortisol and
testosterone levels on match days (van der Meij et al., 2012); increased rate of myocardial infarction
(heart attacks) following major games (Carroll, Ebrahim, Tilling, Macleod, & Smith, 2002; Wilbert-
Lampen et al., 2008); heightened emotions among spectators (Jones, Coffee, Sheffield, Yangüez, &
Barker, 2012; Sullivan, 2014); and, in some cases, aggressive or violent altercations among fans
(Stott, Adang, Livingstone, & Schreiber, 2007; Stott, Hutchison, & Drury, 2001; Stott & Pearson,
2007).
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
To what extent, do group identities and bonding contribute to these reactions? Using field
laboratories for maximum ecological validity, and the collection of salivary cortisol samples under
carefully controlled conditions, we offer replication evidence for van der Meij et al.’s (2012)
finding that group identity predicts increased cortisol concentrations regarding sports events, but
gender does not. We also measure a particularly strong form of group bonding, ‘identity fusion’
(Buhrmester & Swann, 2015; W. B. Swann, Gómez, Seyle, Morales, & Huici, 2009) and test for its
relationship to cortisol concentrations both during live matches and up to a week after the game.
1.1 The stress response system
The main role of the stress response system is to coordinate behavioural and physiological
responses when an individual faces physical and/or psychosocial challenges (Del Giudice, Ellis, &
Shirtcliff, 2011; Sapolsky, Romero, & Munck, 2000). Non-invasive biomarkers of the stress
response system have been widely used in stress research, of which salivary cortisol is the most
representative glucocorticoid and has been a major biomarker of physical and/or psychosocial
challenges (Ulrich-Lai & Herman, 2009). Past studies have shown that psychosocial stress or
situations that threaten the social self, promote an increase in cortisol secretion (Dickerson &
Kemeny, 2004), suggesting that cortisol helps the individual to cope with psychosocially
demanding scenarios.
Increases in salivary cortisol have also been implicated in the physical and social demands
of numerous sports (Hayes, Grace, Baker, & Sculthorpe, 2015). For instance, peak cortisol levels
have been found in badminton players who have just lost a game (González Jiménez et al., 2012).
These effects also apply to players who are not involved in anaerobic exercise, such as pairs of
Caribbean domino players competing against a neighbouring town (Wagner, Flinn, & England,
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
2002). Spectators, who experience vicarious competition also experience an increase in cortisol
levels, as shown by the higher cortisol secretions found among Spanish soccer fans during the final
soccer match of the 2010 FIFA World Cup, compared to those found on a control day (van der Meij
et al., 2012). The social component of soccer seems integral to cortisol production, as the increases
observed in Spanish participants were highest among participants who reported the highest levels of
fandom.
1.2 The stress response system and identity fusion
Little is known about the impact of group bonding on salivary cortisol release and still less is
known about how strong forms of alignment, such as identity fusion, affect physiological stress
when the welfare of the group is at stake. Mazur (2005) hypothesised that stress-induced alliances
form due to a stressor promoting a physiological response, which encourages an alliance with
immediate others (e.g. other allies or fans), this in turn reduces the physiological response, thus
weakening the initial alliance or coalition. The closest research to test such a hypothesis was
conducted by W. B. Swann, Gómez, Huici, Morales, and Hixon (2010). It showed that heightened
arousal (i.e. increased heart rate via physical exercise) translates into pro-group activity for people
reporting high fusion scores but not for people simply reporting high identification scores. This is
because, once fused, an individual’s group identity is completely merged with their personal
identity and therefore increasing agency (via arousal) also increases pro-group behaviours.
Indeed, it is well established that ‘fused’ individuals demonstrate extraordinarily high levels
of group commitment (W. Swann, Jetten, Gómez, Whitehouse, & Bastian, 2012). While
identification refers to social identities that are context dependent and can effectively be switched
‘off’ or ‘on’ (Tajfel & Turner, 1979; Whitehouse & Lanman, 2014), fusion reflects a group identity
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
that is always ‘on’ (Buhrmester & Swann, 2015). Fusion to one’s soccer club is associated with
lifelong loyalty to the club (Newson, Buhrmester, & Whitehouse, 2016), violence toward rival fans
(Newson et al., 2018), willingness to fight and die for one’s fellow fans (Bortolini, Newson,
Natividade, Vázquez, & Gómez, 2018; Newson et al., 2018) and even a willingness to sacrifice
one’s life for fellow fans who are in danger (Whitehouse et al., 2017). If heart attacks are among the
health risks for World Cup spectators, and fusion is a strong predictor of negative stress, then
perhaps identity fusion could contribute to better understandings of stress and other factors that
predict myocardial infarctions and coronary heart disease (Steptoe & Kivimäki, 2012).
As yet, fusion’s relationship with psychophysiological measures has not yet been
extensively studied (W. B. Swann et al., 2010), an issue that we address in this paper. Using field
laboratories where we broadcasted live 2014 FIFA World Cup matches in Brazil, we examined
whether experiences of a major global sporting event can have physiologically arousing (i.e.
stressful), yet socially cohesive effects.
2. Present Study
Field sites provide higher levels of ecological validity than laboratory studies (Wilson &
Whitehouse, 2016), give access to hard to reach populations, and generate data that is rich in
variation which helps to move overarching theories forwards (Newson, Buhrmester, Xygalatas, &
Whitehouse, under review). This is especially important when quantifying social situations, which
are challenging to authentically replicate in the lab. However, most research on human psychology
has been carried out in non-ecologically valid contexts, using ‘samples, WEIRD’, i.e. samples that
are Western, Educated, Industrialised, Rich, and Democratic, and usually under-graduate
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
psychology students seeking course credits (Henrich, Heine, & Norenzayan, 2010a). These WEIRD
participants appear to be outliers in numerous cognitive and behavioral domains, from visual
perception, spatial reasoning, and inferential induction, to moral reasoning, decision-making, and
the heritability of IQ. As such, WEIRD participants are ‘the worst population from which to
generalize’ (Henrich et al., 2010a: 79). Over 90% of articles in 2018 articles still reported studies of
European and English-speaking nations, and less than 1% focused on Latin-American or African
nations (Newson et al., under review).
Some fusion research has taken considerable steps to move away from the confines of
recruiting students online and in the laboratory (e.g. the empirical studies associated with the
dysphoric pathway to fusion mentioned above), but more can be done. To move beyond WEIRD
realms, we ran a longitudinal study using Brazilian spectators during the 2014 World Cup. Held in
Brazil, this was a sporting catastrophe for the hosts (Sullivan, 2014) who, despite a promising start,
suffered the worst home defeat in World Cup history, considered by many as a ‘historical disaster’
(Bowman, 2014). Although this paper’s sample size is small, our findings must be considered in the
context of exploratory, naturalistic research, with rigorous controls on physiological measures.
First, we predicted that salivary cortisol concentrations would increase during live World
Cup games, and that stress would be particularly acute under dysphoric match outcomes (H1). Next,
we predicted that cortisol concentrations would correlate with both fusion and identification (H2).
We also predicted that the relationship between cortisol and fusion would be stronger than its
relationship with identification (H3).
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
3 Materials and Method
3.1 Participants and procedure
We organised free, high-quality, live screenings of three matches near a commercial district of the
city of Natal, North-East Brazil. We screened two wins (Brazil v. Chile, 1-1 with penalties /
Colombia, 2-1: n = 32) and one loss (Brazil v. Germany, 1-7: n = 10). While the winning games
were euphoric events and enabled the team to progress through group-stages against national rivals,
the lost game was acutely dysphoric; their worst national defeat since 1920. All those attending the
screenings supported Brazil. Ethics were obtained from the School of Anthropology, University of
Oxford. Any participants who selected a nationality other than the nationality under study or were
under the age of 18 were not accepted for this study.
An advert of the screening and study was posted in local universities, hostels and shopping
districts, with the chance to win a lottery prize worth R$400 (approximately £100) as well as free
food and drink for taking part following the game (bottled water was provided during the game).
This was supported by a snowball technique (viz., through social media and WhatsApp groups that
are highly popular in Brazil). Potential participants were made aware of our eligibility criteria in
advance.
We sought to control baseline cortisol concentrations, more extensively than we have seen
in other field studies (van der Meij et al., 2012), with exclusion criteria preventing participants from
being admitted to the study for: habitual smoking (Kirschbaum & Hellhammer, 1994); caffeine
intake up to two hours before the experiment (Kudielka, Hellhammer, & Wüst, 2009); alcohol
consumption up to 24 hours before the experiment (Badrick, Kirschbaum, & Kumari, 2007); and
the use of oral contraceptives or being in the follicular phase of their cycle according to the last date
of their period (Kirschbaum, Kudielka, Gaab, Schommer, & Hellhammer, 1999). We estimated
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
menstrual cycle phases based on backward method (Gangestad et al., 2016), which counts days
from the next menstrual onset backward to the day of assessment. For this purpose, we asked
women to report the date of their last three menses onset.
A pre-screening survey determined whether participants who came to the field laboratory
were permitted to take part in the cortisol measures, or only the psychometric measures (n = 70). N
= 29 were excluded for the effects their cigarette / caffeine / alcohol consumption or whose
hormones might have affected the cortisol analyses. Only participants who completed both cortisol
and survey measures are included here (n = 41, females = 19, males = 22; Mage = 27.66, SD = 6.51,
range = 19-48).
Physiological studies often use sample sizes with n < 50, compared to the n > 100 or even
1000 of social psychology, e.g., van der Meij et al.’s 2012 study of the cortisol and testosterone
levels of 50 World Cup fans; Porges, Smith, and Decety (2015) study on vagal regulation and
testosterone using a sample of 43 students; or Yoo et al. (2016) study on salivary cortisol and social
anxiety among 42 children.
3.2 Measures
Psychometric data was collected before kick-off (up to 20 minutes in advance). Participants were
invited to complete the psychometric measures again up to a week later but these are excluded due
to attrition (61.9%, see SI 1). All participants were invited to complete salivary cortisol measures,
for which they were required to provide a saliva sample by chewing on a Salivette for one minute
during each of the three time points. Psychophysiological measures were taken at three time points:
20 minutes pre-match; during the half-time interval; and 20 minutes post-match. Thus, sample 2
was taken 50m – 90m following Sample 1 and Sample 3 was taken 50-90m following Sample 2.
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Saliva sampling took a total of 4 minutes on each occasion, i.e. 1 minute to collect the salivette, 2
minutes to chew it, and 1 minute to return to the researcher and record their details.
To control for circadian effects, saliva samples were collected after midday (Pruessner et al.,
1997). All samples were stored at -20ºC until the assaying day. Before the assay, samples were
thawed and centrifuged at 3000 rpm for 10 minutes to produce a clear supernatant of low viscosity.
Free salivary cortisol concentrations were determined in duplicate aliquots using a commercial
enzyme-linked immunosorbent assay (ELISA) kit with a high sensitivity of 0.012 ng/mL (DRG
Instruments, Marburg, Germany). The intra- and inter-assay coefficients of variation were 3.6% and
8%, respectively.
As we were working at live events we used a reduced version of the verbal fusion scale
(Gómez et al., 2011) comprising 3 state-type items, assessed on a 7-item Likert scale (I am ‘one’
with the Brazilian national team and my fellow fans; I make the Brazilian national team and my
fellow fans strong; The Brazilian national team and my fellow fans make me strong). Fusion items
were selected that represented the cornerstones of the fusion construct: the porous boundary
between self and group identities and reciprocal strength (W. Swann et al., 2012). The reduced scale
demonstrated good internal validity (Cronbach’s alpha = .89). We also used a single-item measure
of identification that has been previously validated (Postmes, Haslam, & Jans, 2013); ‘I identify
with the Brazilian national team and my fellow fans]. Both fusion and identification were assessed
in state form, by adding ‘Right now’ at the start of each item. Measures were translated into
Brazilian Portuguese by three English/Brazilian Portuguese bilingual speakers. They were then
back translated and checked for validity by a native English speaker. Psychometric measures were
also recorded in a second wave of data collection 1 day – 1 week post-match but suffered from
severe attrition and are thus included in SI 2.
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
3.3 Analysis
All analyses reported below were conducted using the IBM SPSS statistical Software package
(V22). Prior to analyses, all variables were examined for missing and extreme values. Skew and
kurtosis were observed for cortisol (i.e. leptokurtic, highly positively skewed data see SI), but
deemed manageable by OLS regressions. To estimate the magnitude of cortisol response, the area
under the curve with respect to ground (AUCg) was calculated. This metric reflects the total cortisol
concentration (nmol/L) released in a time interval, using the first measure as the reference (i.e. as
zero) (Pruessner, Kirschbaum, Meinlschmid, & Hellhammer, 2003). We ran the models using the
raw values of cortisol to calculate the AUCg, since recent discussion shows log transformation
overestimates biological phenomena, e.g. conception probability (Roney, 2019). Therefore, we
deem raw values a better proxy for hormonal mechanisms, though log transformation results are
reported in full in SI 3. Only results pertaining to one of the hypotheses had their significance
changed in the transformation analyses, which is reported below.
For consistency with endocrinological literature we checked for AUCg outliers that were 3
SDs above the mean (n = 2) (Pollet & van der Meij, 2017). See Fig. 1 for data distribution.
Excluding these outliers resulted in slightly larger p values, likely as an effect of reducing the
already small sample size, but the significance of the results was not changed (see SI 4). Analyses
for all data points are thus included below. Outliers were not found for psychometric measures.
[INSERT FIG. 1 ABOUT HERE]
4 Results
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Descriptives and correlations are reported in Table 1. Across win (n = 31) and loss games (n = 10),
participants were quite similar in terms of gender, pre-game cortisol and pre-game fusion (p >
.162). However, participants were significantly older in the loss game (M = 31.50, SD = 9.59)
compared to the win games (M = 26.42, SD = 4.72), t = -2.25, p = .030. We found that females in
this sample (M = 4.85, SD = 1.47) were significantly more fused than males (M = 3.64, SD = 1.74),
t(38) = -2.36, p = .024, while males and females did not differ for identification (p = .243). We thus
re-ran all analyses including gender as a covariate, but this did not affect any of the results (SI 5).
Cortisol concentrations fluctuate during live sporting events. A repeated-measures
ANOVA including T1, T2, and T3 cortisol as within-subject factors and game outcome as a
between-subject factor showed an overall change in salivary cortisol before, during and after the
match, F(2, 38) = 3.87, p = .030, partial η2 = .17. There was also a significant cortisol x result
interaction, F(2, 38) = 10.13, p < .001, partial η2 = .35. Mauchly’s test of sphericity was not violated
(p = .272). Post-hoc tests revealed that at half time (T2) the loss game was associated with
significantly more cortisol release (M = 17.98, SD = 15.17) than the win games (M = 5.55, SD =
4.83), t(40) = -4.10, p < .001, but there were not significant differences between win/loss outcomes
on other occasions. This is likely because the loss game involved the opposition leading 5-0 by half
time (Fig. 2). We log-transformed the data (SI 3) and found no significant change in cortisol over
time (p = .112). However, the interaction reported above, and subsequent results reported below,
did not change significance in the transformation analyses.
[INSERT TABLE 1 ABOUT HERE]
[INSERT FIG. 2 ABOUT HERE]
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Increased cortisol concentrations correlate with fusion. Spectators who reported higher levels of
fusion and identification to their team and fellow fans also experienced the greatest increases in
cortisol concentration over the course of the game (see Table 1 for correlations, r > .37, p < .014). A
step-wise linear regression with fusion and identification entered as independent variables, and
cortisol AUCg entered as the dependent variable indicated that fusion (β = .45, p = .004) was the
best predictor of cortisol AUCg (R2 = .20, F(1, 38) = 9.59, p = .004. In this model, identification
was excluded (β = .16, p = .390). However, when entered alone, identification also related to
cortisol AUCg, but not as well as fusion, R2 = .15, F(1, 38) = 6.74, p = .013. See Fig. 1 for a
scatterplot of fusion and identification plotted against cortisol AUCg.
4 Discussion
Across win and loss outcomes, analyses provided some evidence that participants who experienced
the greatest stress response system activation over the course of a World Cup match also tended to
be the most fused. We expected dysphoric outcomes to induce more cortisol release, particularly
given the scale of stress experienced by Brazilians in their final 2014 World Cup game at home (5-0
defeat by half time), and this was tentatively supported by the results albeit sample sizes were
particularly small when split by game outcome. Previous research has only looked at euphoric
events, hence earlier reports of salivary cortisol concentrations increasing on match days compared
to non-match control days, but not increasing during the (win) match itself (van der Meij et al.,
2012).
A distinct pattern in cortisol secretion for highly fused individuals emerged. This
relationship has never before been demonstrated and fusion was more strongly related to cortisol
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
AUCg than identification. Though grounded in well-supported theory (W. B. Swann et al., 2010;
W. B. Swann et al., 2009; Whitehouse, 1996, 2004), this was an exploratory study and these initial
results must be treated cautiously. Fusion appears to modulate physiological reactivity, resulting in
distinct psycho-physiological profiles. This research, using a naturalistic setting, empirically tested
how stress (i.e. salivary cortisol concentrations) relates to important self-group identity relations,
paving the way for future research.
This study focused on spectators, but a question remains as to whether highly fused active
participants in competitive sport experience heightened cortisol concentrations during matches in a
similar way, or perhaps to an even greater extent. The stress levels exhibited by the sample, which
correlated with a measure of extreme group bonding, support theories of stress-induced alliances
that would have been relevant in the ancestral past, i.e. during inter-group conflict (Choi & Bowles,
2007; Mazur, 2005; McDonald, Navarrete, & Van Vugt, 2012). As identity fusion was a significant
predictor of stress in a context that is already associated with a higher risk for heart attacks, we
suggest that it should be considered as a factor when determining risk groups for the purposes of
health research and policy, for instance hardcore (highly fused) fans may be offered cardiac exams
by their clubs as part of community programmes and provided with education around heart health
and managing high stress events.
We did not find significant sex differences in this sample, aside from Brazilian women
being significantly more fused to their World Cup team and fellow fans than men. This supports
van der Meij et al.’s findings concerning salivary cortisol concentrations of fans watching their
teams play during World Cup soccer events. This is perhaps surprising given evolutionary research
that places males at the forefront of sports spectatorship (Deaner, Balish, & Lombardo, 2015;
Lombardo, 2012). As World Cup sport is explicitly tied up with national and cultural identities,
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
further work is needed to explore whether club-level competitive sport induces more stress (and
potentially alliance) among men than women, and whether this is responsible for the
disproportionate number of men who engage with regional soccer compared to women.
A major limitation concerns our small sample size. Post-hoc power analyses were
conducted, revealing that for medium effect sizes of F2 = .25 and r = .45 for correlations, sample
sizes of 44, 39, and 64 were needed for the repeated measures ANOVA (H1), correlations (H2), and
linear regressions (H3) respectively. (Faul, Erdfelder, Buchner, & Lang, 2009). Nonetheless,
consistent with psychophysiological research, our sample consisted of n = 41, with just 10 in the
loss game due to many participants sobbing or walking out before half time. Participants in the loss
game were also older, perhaps due to younger participants spontaneously deciding to leave the
research early due to the game’s outcome. Presumably the cortisol scores of those deciding to quit
the research were even higher. Furthermore, the dysphoric game occurred later in the World Cup
tournament, with the euphoric games occurring at the start. As Brazil won all of its opening games
there was not an opportunity to compare dysphoric and euphoric games from the same subsection of
the tournament, though it would be a more controlled comparison for future research.
The study with perhaps the most parallels is van der Meij’s field study on cortisol and
testosterone secretion among soccer fans (2012). We commend the authors’ ambitions and strong
results, and aimed to extend their research by using a challenging field environment coupled with
even more rigorous controls (e.g. smoking / drinking / eating rules were stricter for cortisol
analyses). Even laboratory studies including hormonal and other physiological measures tend to
have samples of around 50 participants, such as Diekhof et al.’s 2014 study with soccer fans on
testosterone and parochial altruism (n = 50) (Diekhof, Wittmer, & Reimers, 2014) or van der Meij’s
other fan study into basal cortisol (n = 74) (van der Meij et al., 2015). This demonstrates how
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
challenging it is to recruit and retain participants for such measures in the field. Both time and
financial costs influence sample sizes in such studies. In addition, our small sub-sample sizes are
balanced by the fact that there are fewer uncontrolled social and background variables in
physiological measures. Indeed, health research using physiological measures often comprises only
eight subjects (Reinard, 2006). Nonetheless, the issue of sample sizes should be addressed in future
studies – potentially by visiting participants’ homes or running controlled laboratory studies as per
van der Meij et al.
In the present study, sample sizes were too small to assess the effect of loss versus win
outcomes on the cortisol - fusion relationship. Nonetheless, it appears that watching one’s team lose
produces a significant increase in salivary cortisol concentrations, whilst watching a victory does
not. It may be, as per van der Meij et al.’s paper, that cortisol concentrations also rose on victorious
match days in relation to control non-match days, but we did not detect an increase during the game
itself. Given the time constraints of collecting data in the field, the survey was kept very brief. The
original survey was administered to >400 participants during live games in stadia and at FanFest
sites, as well as at our field laboratories in homes and community centres. The short nature of the
survey thus reflects this need to attract and retain participants in a challenging environment. Of
particular value for future inclusion would be a measure of state anxiety, which could be used as a
control for results. Furthermore, identification should be measured beyond the single-item measure
we used here.
Previous research has not found significant associations between negative affect and fusion,
but relationships between positive affect and fusion have been demonstrated (Kavanagh, Jong,
McKay, & Whitehouse, 2017). An issue with such self-reported measures then, is that while an
event may be painful, embarrassing, or traumatic, participants may still rate it as ‘positive’ because
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
of the potent social ties it facilitates. Further work is needed to tie together objective physiological
measures, self-reported affect measures, and qualitative participant testimonies. Testing how
frequently participants spectate or participate in a given event may also be important – does regular
exposure to highly emotive crowd rituals such as football matches produce cumulative stress or
does it reduce stress via familiarity or increased bonding to the extended group?
This study highlights the challenges of collecting longitudinal data in the field and is a
methodological step toward less WEIRD, and more WILD (Worldly, Independent, Local, and
Distinctive), participants (Henrich, Heine, & Norenzayan, 2010b; Newson et al., under review). The
criteria for physiological measures were strict, so only participants who did not smoke, agreed not
to eat nor drink prior to, or consume any alcohol during the experiment, and were willing to wear
the equipment during live matches were included. To combat threats to external validity of forgoing
alcohol during a World Cup game, we included local beers in the post-match buffet, which
participants were aware of – so it was a matter of delaying rather than forgoing alcohol
consumption. It is also important to note that the relationship between alcohol and football
consumption is not quite as high in Brazil as it often is in Northern Europe, with respect to
differences in ‘binging’ culture more generally. A particular obstacle in this research was
persuading participants to chew on an unpleasant salivette for a minute, three times, after
experiencing the worst World Cup defeat in living memory. At a cost to our relatively high levels of
ecological validity (i.e. natural groups attending a live event in a natural setting) are issues of
internal validity. Specifically, there is a possibility that the data is non-independent (i.e. the game
was watched as a group, with the highs and lows of the game shared among fans). As such, ingroup
variability may be biased downward, increasing the likelihood of a Type 1 error. This points to the
merits of mixing methods and analyses to triangulate results, whereby carefully controlled
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
laboratory studies can confirm or disprove our ecologically valid results (Newson et al., under
review).
5 Conclusions
Although sample sizes could be improved in future studies, these exploratory results are
encouraging. Taken together, analyses suggest that highly fused people have a distinct
physiological profile, such that there seems to be an interaction between the stress response system
and fusion during stressful group events. The behavioural consequences of fusion have the potential
to be some of the most extreme and dangerous social behaviours we know, with fused group
members risking life and limb in the name of their group (Buhrmester, Fraser, Lanman,
Whitehouse, & Swann, 2015; Newson et al., 2018; Sheikh, Gómez, & Atran, 2016; W. Swann,
Gómez, Dovidio, Hart, & Jetten, 2010; Whitehouse et al., 2017; Whitehouse, McQuinn,
Buhrmester, & Swann, 2014). When viewed in this way, a more thorough understanding of extreme
forms of group alignment, such as fusion, and how this bonding relates to stress and stressful
environments could inform practices around crowd control and group relations more broadly.
References
Badrick, E., Kirschbaum, C., & Kumari, M. (2007). The relationship between smoking status
and cortisol secretion. The Journal of Clinical Endocrinology & Metabolism, 92(3), 819-
824.
Bortolini, T., Newson, M., Natividade, J., Vázquez, A., & Gómez, Á. (2018). Identity fusion
predicts pro-group behaviours: targeting nationality, religion or football in Brazilian
samples. British Journal of Social Psychology. doi:10.1111/bjso.12235
Bowman, J. (2014). Brazil's newspapers cry 'Shame,' 'Grief,' 'Disgrace' after stunning World
Cup loss. Retrieved from CBC News website:
http://www.cbc.ca/newsblogs/yourcommunity/2014/07/brazils-newspapers-cry-
shame-grief-disgrace-after-stunning-world-cup-loss.html
Brown, S. (2007). Football fans around the world: From supporters to fanatics. Oxford:
Routledge.
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Buhrmester, M., Fraser, W. T., Lanman, J. A., Whitehouse, H., & Swann, W. B. (2015). When
terror hits home: Identity fused Americans who saw Boston bombing victims as
“family” provided aid. Self and Identity, 14(3), 253-270.
Buhrmester, M., & Swann, W. B. (2015). Identity Fusion. In Emerging Trends in the Social and
Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource (pp. 1–15).
Carroll, D., Ebrahim, S., Tilling, K., Macleod, J., & Smith, G. D. (2002). Admissions for myocardial
infarction and World Cup football: database survey. BMJ: British Medical Journal,
325(7378), 1439.
Choi, J.-K., & Bowles, S. (2007). The coevolution of parochial altruism and war. Science,
318(5850), 636-640.
Deaner, R. O., Balish, S. M., & Lombardo, M. P. (2015). Sex differences in sports interest and
motivation: An evolutionary perspective.
Del Giudice, M., Ellis, B. J., & Shirtcliff, E. A. (2011). The adaptive calibration model of stress
responsivity. Neuroscience & Biobehavioral Reviews, 35(7), 1562-1592.
Dickerson, S. S., & Kemeny, M. E. (2004). Acute stressors and cortisol responses: a theoretical
integration and synthesis of laboratory research. Psychological bulletin, 130(3), 355.
Diekhof, E. K., Wittmer, S., & Reimers, L. (2014). Does competition really bring out the worst?
Testosterone, social distance and inter-male competition shape parochial altruism in
human males. PloS one, 9(7), e98977.
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(4), 1149-1160.
FIFA. (2015). 2014 FIFA World Cup™ reached 3.2 billion viewers, one billion watched final.
Retrieved from FIFA website: 2014 FIFA World Cup™ reached 3.2 billion viewers, one
billion watched final
Giulianotti, R. (1995). Football and the politics of carnival: An ethnographic study of Scottish
fans in Sweden. International Review for the Sociology of Sport, 30(2), 191-220.
Gómez, Á., Brooks, M. L., Buhrmester, M. D., Vázquez, A., Jetten, J., & Swann Jr, W. B. (2011). On
the nature of identity fusion: Insights into the construct and a new measure. Journal of
Personality and Social Psychology, 100(5), 918-933. doi:10.1037/a0022642
González Jiménez, E., Aguilar Cordero, M., García García, C. J., García López, P., Álvarez Ferre, J.,
Padilla López, C., & Ocete Hita, E. (2012). Influencia del entorno familiar en el
desarrollo del sobrepeso y la obesidad en una población de escolares de Granada
(España). Nutrición Hospitalaria, 27(1), 177-184.
Hayes, L. D., Grace, F. M., Baker, J. S., & Sculthorpe, N. (2015). Exercise-induced responses in
salivary testosterone, cortisol, and their ratios in men: a meta-analysis. Sports
Medicine, 45(5), 713-726.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010a). Beyond WEIRD: Towards a broad-based
behavioral science. Behavioral and Brain Sciences, 33(2-3), 111-135.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010b). The weirdest people in the world?
Behavioral and Brain Sciences, 33(2-3), 61-83.
Jones, M. V., Coffee, P., Sheffield, D., Yangüez, M., & Barker, J. B. (2012). Just a game? Changes in
English and Spanish soccer fans’ emotions in the 2010 World Cup. Psychology of Sport
and Exercise, 13(2), 162-169. doi:http://dx.doi.org/10.1016/j.psychsport.2011.10.008
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Kavanagh, C., Jong, J., McKay, R., & Whitehouse, H. (2017). High arousal martial arts rituals
predict identity fusion and prosocial behavior.
King, A. (2002). End of the terraces: The transformation of English football. London, New York:
Leicester University Press.
King, A. (2003). The European ritual: Football in the new Europe. Oxford: Ashgate Publishing,
Ltd.
Kirschbaum, C., & Hellhammer, D. H. (1994). Salivary cortisol in psychoneuroendocrine
research: recent developments and applications. Psychoneuroendocrinology, 19(4),
313-333.
Kirschbaum, C., Kudielka, B. M., Gaab, J., Schommer, N. C., & Hellhammer, D. H. (1999). Impact
of gender, menstrual cycle phase, and oral contraceptives on the activity of the
hypothalamus-pituitary-adrenal axis. Psychosomatic medicine, 61(2), 154-162.
Kudielka, B. M., Hellhammer, D., & Wüst, S. (2009). Why do we respond so differently?
Reviewing determinants of human salivary cortisol responses to challenge.
Psychoneuroendocrinology, 34(1), 2-18.
Lombardo, M. P. (2012). On the evolution of sport. Evolutionary Psychology, 10(1).
Mazur, A. (2005). Biosociology of dominance and deference: Rowman & Littlefield Publishers.
McDonald, M. M., Navarrete, C. D., & Van Vugt, M. (2012). Evolution and the psychology of
intergroup conflict: The male warrior hypothesis. Phil. Trans. R. Soc. B, 367(1589), 670-
679.
Newson, M., Bortolini, T., Buhrmester, M., da Silva, S., Acquino, J., & Whitehouse, H. (2018).
Brazil's Football Warriors: Social bonding and inter-group violence. Evolution and
Human Behavior, 39(6), 675-683.
doi:https://doi.org/10.1016/j.evolhumbehav.2018.06.010
Newson, M., Buhrmester, M., & Whitehouse, H. (2016). Explaining Lifelong Loyalty: The Role
of Identity Fusion and Self-Shaping Group Events. PloS one, 11(8), e0160427.
Newson, M., Buhrmester, M., Xygalatas, D., & Whitehouse, H. (under review). To avoid WEIRD
participants, go WILD. Journal for the Cognitive Science of Religion
.
Pearson, G. (2012). An ethnography of English football fans. Cans, Cops and Carnivals.
Pollet, T. V., & van der Meij, L. (2017). To remove or not to remove: the impact of outlier
handling on significance testing in testosterone data. Adaptive Human Behavior and
Physiology, 3(1), 43-60.
Porges, E. C., Smith, K. E., & Decety, J. (2015). Individual differences in vagal regulation are
related to testosterone responses to observed violence. Frontiers in Psychology, 6, 19.
Postmes, T., Haslam, S. A., & Jans, L. (2013). A single-item measure of social identification:
Reliability, validity, and utility. British Journal of Social Psychology, 52(4), 597-617.
doi:10.1111/bjso.12006
Pruessner, J. C., Kirschbaum, C., Meinlschmid, G., & Hellhammer, D. H. (2003). Two formulas
for computation of the area under the curve represent measures of total hormone
concentration versus time-dependent change. Psychoneuroendocrinology, 28(7), 916-
931.
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Pruessner, J. C., Wolf, O. T., Hellhammer, D. H., Buske-Kirschbaum, A., Von Auer, K., Jobst, S., . . .
Kirschbaum, C. (1997). Free cortisol levels after awakening: a reliable biological
marker for the assessment of adrenocortical activity. Life sciences, 61(26), 2539-2549.
Roney, J. R. (2019). On the use of log transformations when testing hormonal predictors of
cycle phase shifts: Commentary on. Evolution and Human Behavior, 40(6), 526-530.
Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids influence stress
responses? Integrating permissive, suppressive, stimulatory, and preparative actions.
Endocrine reviews, 21(1), 55-89.
Sheikh, H., Gómez, Á., & Atran, S. (2016). Empirical evidence for the devoted actor model.
Current Anthropology, 57(S13), S000-S000.
Steptoe, A., & Kivimäki, M. (2012). Stress and cardiovascular disease. Nature Reviews
Cardiology, 9(6), 360.
Stone, C. (2007). The role of football in everyday life. In S. Brown (Ed.), Football fans around
the world: From supporters to fanatics: Routledge.
Stott, C., Adang, O., Livingstone, A., & Schreiber, M. (2007). Variability in the collective
behaviour of England fans at Euro2004:‘Hooliganism’, public order policing and social
change. European Journal of Social Psychology, 37(1), 75-100.
Stott, C., Hutchison, P., & Drury, J. (2001). ‘Hooligans’ abroad? Inter-group dynamics, social
identity and participation in collective ‘disorder’ at the 1998 World Cup Finals. British
Journal of Social Psychology, 40(3), 359-384. doi:10.1348/014466601164876
Stott, C., & Pearson, G. (2007). Football'hooliganism': policing and the war on the'English
disease': Citeseer.
Stroeken, K. (2002). Why ‘the world’loves watching football (and ‘the Americans’ don’t).
Anthropology today, 18(3), 9-13.
Sullivan, G. B. (2014). Collective emotions and the World Cup 2014: The relevance of theories
and research on collective pride and shame. Psicologia e Saber Social, 3(1), 112-117.
Swann, W., Gómez, Á., Dovidio, J. F., Hart, S., & Jetten, J. (2010). Dying and Killing for One’s
Group: Identity Fusion Moderates Responses to Intergroup Versions of the Trolley
Problem. Psychological Science, 21(8), 1176-1183. doi:10.1177/0956797610376656
Swann, W., Jetten, J., Gómez, Á., Whitehouse, H., & Bastian, B. (2012). When group membership
gets personal: A theory of identity fusion. Psychological review, 119(3), 441-456.
doi:10.1037/a0028589
Swann, W. B., Gómez, Á., Huici, C., Morales, J. F., & Hixon, J. G. (2010). Identity fusion and self-
sacrifice: Arousal as a catalyst of pro-group fighting, dying, and helping behavior.
Journal of Personality and Social Psychology, 99(5), 824-841. doi:10.1037/a0020014
Swann, W. B., Gómez, Á., Seyle, D. C., Morales, J. F., & Huici, C. (2009). Identity fusion: The
interplay of personal and social identities in extreme group behavior. Journal of
Personality and Social Psychology, 96(5), 995-1011. doi:10.1037/a0013668
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin &
S. Worchel (Eds.), The social psychology of intergroup relations (Vol. 33, pp. 33-47).
Monterey, CA: Brooks / Cole.
Ulrich-Lai, Y. M., & Herman, J. P. (2009). Neural regulation of endocrine and autonomic stress
responses. Nature Reviews Neuroscience, 10(6), 397-409.
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
van der Meij, L., Almela, M., Hidalgo, V., Villada, C., IJzerman, H., van Lange, P. A., & Salvador, A.
(2012). Testosterone and cortisol release among Spanish soccer fans watching the
2010 World Cup final. PloS one, 7(4), e34814.
van der Meij, L., Klauke, F., Moore, H. L., Ludwig, Y. S., Almela, M., & van Lange, P. A. (2015).
Football Fan aggression: the importance of low basal cortisol and a fair referee. PloS
one, 10(4), e0120103.
Wagner, J. D., Flinn, M. V., & England, B. G. (2002). Hormonal response to competition among
male coalitions. Evolution and Human Behavior, 23(6), 437-442.
Weed, M. (2007). The Pub as a Virtual Football Fandom Venue: An Alternative to ‘Being
there’? In S. Brown (Ed.), Football fans around the world: From supporters to fanatics:
Routledge.
Whitehouse, H. (1996). Rites of terror: Emotion, metaphor and memory in Melanesian
initiation cults. Journal of the Royal Anthropological Institute, 703-715.
Whitehouse, H. (2004). Modes of religiosity: A cognitive theory of religious transmission:
Rowman Altamira.
Whitehouse, H., Jong, J., Buhrmester, M., Gomez, A., Bastian, B., Kavanagh, C., . . . Gavrilets, S.
(2017). The evolution of extreme cooperation via shared dysphoric experiences.
Nature: Scientfic reports(7, p.44292).
Whitehouse, H., & Lanman, J. A. (2014). The Ties That Bind Us: Ritual, fusion, and
identification. Current Anthropology, 55(6), 674-695.
Whitehouse, H., McQuinn, B., Buhrmester, M., & Swann, W. B. (2014). Brothers in arms: Libyan
revolutionaries bond like family. Proceedings of the National Academy of Sciences,
111(50), 17783-17785.
Wilbert-Lampen, U., Leistner, D., Greven, S., Pohl, T., Sper, S., Völker, C., . . . Steinbeck, G.
(2008). Cardiovascular Events during World Cup Soccer. New England Journal of
Medicine, 358(5), 475-483. doi:doi:10.1056/NEJMoa0707427
Wilson, D. S., & Whitehouse, H. (2016). Developing the Field Site Concept for the Study of
Cultural Evolution. Cliodynamics: The Journal of Quantitative History and Cultural
Evolution, 7(2).
Yoo, Y.-G., Lee, D.-J., Lee, I.-S., Shin, N., Park, J.-Y., Yoon, M.-R., & Yu, B. (2016). The effects of
mind subtraction meditation on depression, social anxiety, aggression, and salivary
cortisol levels of elementary school children in South Korea. Journal of pediatric
nursing, 31(3), e185-e197.
Tables
Table 1. Principle (M,SD or %) and demographic variables and their correlations (Spearman’s
Rho)
T1
Fusion
Cortisol
AUCg
Age Gender
T1 Ident.
T1 Fusion
4.18
(1.71)
This article is protected by copyright. All rights reserved.
SOCCER & CORTISOL
Cortisol AUCg
.40**
16.34
(13.50)
Age
-.03 .07
27.66
(6.51)
Gender
.35* .17 -.06
53.7%
M
T1 Ident.
.62***
.35*
-.02
.14
5.05 (1.61)
* p < .05
** p < .005
*** p < .001
Figure legends
Fig. 1. Higher fusion levels are associated with higher salivary cortisol concentrations (AUCg)
Fig. 2. Salivary cortisol concentrations peak during half time, compared to an hour before or
immediately after a match in loss game (mean SE). * denotes p < .05
This article is protected by copyright. All rights reserved.
This article is protected by copyright. All rights reserved.
Accepted Article
This article is protected by copyright. All rights reserved.
Accepted Article
