The role of serotonin in personality inference: tryptophan depletion impairs the identification of neuroticism in the face

Abstract

Serotonergic mechanisms mediate the expression of personality traits (such as impulsivity, aggression and anxiety) that are linked to vulnerability to psychological illnesses, and modulate the identification of emotional expressions in the face as well as learning about broader classes of appetitive and aversive signals. Faces with neutral expressions signal a variety of socially relevant information, such that inferences about the big five personality traits, including Neuroticism, Extraversion and Agreeableness, can be accurately made on the basis of emotionally neutral facial photographs. Given the close link between Neuroticism and psychological distress, we investigated the effects of diminished central serotonin activity (achieved by tryptophan depletion) upon the accuracy of 52 healthy (non-clinical) adults' discriminations of personality from facial characteristics. All participants were able to discriminate reliably four of the big five traits. However, the tryptophan-depleted participants were specifically less accurate in discriminating Neuroticism than the matched non-depleted participants. These data suggest that central serotonin activity modulates the identification of not only negative facial emotional expression but also a broader class of signals about personality characteristics linked to psychological distress.

Full text

ORIGINAL INVESTIGATION
The role of serotonin in personality inference:
tryptophan depletion impairs the identification
of neuroticism in the face
Robert Ward
1
&Shubha Sreenivas
1
&Judi Read
2
&Kate E. A. Saunders
2
&
Robert D. Rogers
1
Received: 27 September 2016 /Accepted: 27 March 2017 / Published online: 9 May 2017
#The Author(s) 2017. This article is published with open access at Springerlink.com
Abstract Serotonergic mechanisms mediate the expression
of personality traits (such as impulsivity, aggression and anx-
iety) that are linked to vulnerability to psychological illnesses,
and modulate the identification of emotional expressions in
the face as well as learning about broader classes of appetitive
and aversive signals. Faces with neutral expressions signal a
variety of socially relevant information, such that inferences
about the big five personality traits, including Neuroticism,
Extraversion and Agreeableness, can be accurately made on
the basis of emotionally neutral facial photographs. Given the
close link between Neuroticism and psychological distress,
we investigated the effects of diminished central serotonin
activity (achieved by tryptophan depletion) upon the accuracy
of 52 healthy (non-clinical) adults’discriminations of person-
ality from facial characteristics. All participants were able to
discriminate reliably four of the big five traits. However, the
tryptophan-depleted participants were specifically less accu-
rate in discriminating Neuroticism than the matched non-
depleted participants. These data suggest that central serotonin
activity modulates the identification of not only negative facial
emotional expression but also a broader class of signals about
personality characteristics linked to psychological distress.
Keywords Serotonin .Personality .Neuroticism .
Psychological distress
Introduction
Serotonin mechanisms mediate the expression of personality
traits linked to mental illnesses (Takano et al. 2007). High
levels of Neuroticism have been associated with increased 5-
HTT binding within the thalamus (Takano et al. 2007), while a
number of characteristic emotional experiences associated
with Neuroticism, such as anxiety, depression, hopelessness,
somatization, guilt, hostility and affective temperament, have
been linked to the 5-HTTPLR polymorphism of the serotonin
transporter gene (Gonda et al. 2009). However, to date, there
have been no systematic investigations of the role of serotonin
activity in the identification of Neuroticism and other aspects
of personality in other people.
Other evidence attests to the importance of serotoninergic
mechanisms in the recognition of facial expressions of
emotion including, but not limited to, fearful states (Harmer
et al. 2003). Tryptophan depletion—producing temporary re-
ductions in central serotonin activity (Moore et al. 2000)—
impairs the accuracy of fear recognition in (non-clinical)
healthy females and slows latencies for the recognition of
fearful expressions in both healthy male and female adults
(Harmer et al. 2003). By contrast, single doses of the selective
serotonin reuptake inhibitor (SSRI), citalopram, improve fear
recognition in healthy adults (Browning et al. 2007)andnor-
malize recognition of fearful (and positive) facial expressions
in previously depressed individuals (Bhagwagar et al. 2004).
These effects are likely to be mediated by altered signaling
within neural circuitry encompassing the amygdala (Harmer
et al. 2004), suggesting that anti-depressant modulation of
monoaminergic activity within limbic circuits alters sensitivi-
ty to facial emotional signals (as well as broader cognitive
biases) to support delayed therapeutic effects (Anderson
et al. 2011; Booij and Van der Does 2011; Hayward et al.
2005;Harmeretal.2006; Walsh and Harmer 2015).
*Robert D. Rogers
r.rogers@bangor.ac.uk
1
School of Psychology, Bangor University, Brigantia Building,
Penrallt Road, LL57 2AS Bangor, UK
2
Department of Psychiatry, University of Oxford, Warneford Hospital,
OX3 7JX Oxford, UK
Psychopharmacology (2017) 234:2139–2147
DOI 10.1007/s00213-017-4619-4
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

However, beyond emotional expressions, the face transmits a
variety of other informative cues to social traits. In this study,
we examined whether serotoninergic neuromodulation influ-
ences the accuracy of personality inferences, as cued by neu-
tral facial expressions.
Research on trait inferences shows that observers can
identify personality and other important social traits, with
modest levels of accuracy, using only impoverished cues.
For example, Borkenau and Liebler (1992)establishedthat,
even at ‘zero-acquaintance’, observers could draw accurate
personality inferences from short video clips of an individual
walking into a room, or quietly facing the camera. Such
nonverbal ‘thin slices’of behaviour can be famously
diagnostic of consensual longer-term impressions; for
example, teachers’end-of-semester ratings can be predicted
from judges’reactions to a few seconds of silent video clips of
those teachers (Ambady and Rosenthal 1992). While accurate
trait inferences can be drawn on the basis of very short
exposures (Borkenau et al. 2009;Ruleetal.2009), accuracy
is also affected by the quantity and quality of cues available
(Funder and Colvin 1988; Carney et al. 2007). In particular,
the availability of other cues under individuals’control—such
as their hairstyles, clothing, cosmetics and facial
expressions—can improve accuracy of trait judgements
(Naumann et al. 2009).
However, even when controllable cues are minimal, facial
appearance under constrained and standardized conditions
(such as in a passport photograph) can drive accurate trait
inferences. Emotionally neutral facial appearances, without
evident clothing, cosmetics, or hairstyle cues, can allow accu-
rate judgements of socially relevant traits as diverse as trust-
worthiness (Stirrat and Perrett 2010), women’sreproductive
success (Pflüger et al. 2012) and even clinically diagnosed
borderline personality disorder (Daros et al. 2016). The accu-
racy of trait inferences based upon neutral faces suggests an
important, but under-explored, channel of social communica-
tion in which individuals continuously and involuntarily
broadcast cues about behavioural predispositions to observers
and potential social partners.
Of primary interest for the present study are findings that
factor-analytic personality traits, and in particular trait
Neuroticism, can be identified from composite images of neu-
tral facial appearance (e.g. Little and Perrett 2007; Kramer and
Ward 2010; Jones et al. 2012). Neuroticism is marked by a
tendency towards fearful emotions including anxiety, depres-
sion, fear and worry (Vinberg et al. 2014; McWilliams 2003).
Neurotic traits are also related to the frequency and intensity of
negative emotions (Verduyn and Brans 2012). Of all the big
five personality traits, Neuroticism would be the most closely
related to fear (Nettle 2006). Given that serotonin plays an
important role in the recognition of emotional expressions
and, in particular, fearful states (e.g. Harmer et al. 2003), we
hypothesised that serotonin activity plays a role in identifying
trait Neuroticism from the face. We therefore investigated, for
the first time, the role of serotonergic activity in making accu-
rate personality discriminations from facial cues. We tested the
specific prediction that tryptophan depletion impairs accurate
identification of Neuroticism from neutral facial expressions
in healthy adult volunteers.
Method
The experiment was approved by a National Health Service
(England) Research Ethics Committee. All participants pro-
vided written, informed consent.
Participants
Fifty-three males and females were recruited from the student
population of Oxford University and the local community.
One female was excluded on the basis of 0% accuracy in
discriminating Neuroticism (3.7 SDs from her group mean).
No other participant had 0% accuracy. Participants were
assessed by an experienced psychiatrist against explicit exclu-
sion criteria using the Structured Clinical Interview for DSM-
IV-TR Axis I Disorders (First et al. 2002).
Exclusion criteria included (i) the presence or history of
serious physical illness, (ii) history of neurological disorder
or head injury, (iii) current, previous or family history of
mood-related illness (unipolar or bipolar disorder), (iv) current
or previous substance misuse or dependence, (v) other signif-
icant psychiatric illness, (vi) any illness or indicators that pre-
clude blood-sampling and (vii) pregnancy or breast-feeding.
Design
The study consisted of a between-subjects, double-blind de-
sign. Twenty-five participants (10 men, 15 women) were ran-
domly selected to consume an amino-acid drink without l-
tryptophan (T−treatment), and 27 participants (16 men, 11
women) were randomly selected to consume an amino-acid
drink with l-tryptophan (T+ treatment).
Materials
Amino acids AminoacidsweresuppliedbyNutriciaor
Cambridge Bioscience. Amounts for males and females, re-
spectively, were the following: l-alanine (5.5 g; 4.58 g), l-
arganine (4.9 g; 4.08 g), l-cystine (2.7 g; 2.25 g), glycine
(3.2 g; 2.67 g), l-isoleucine (8.0 g; 6.67 g), l-leucine (13.5 g;
11.25 g), l-lysine monohydrochloride (11.0 g; 9.17 g), l-me-
thionine (3.0 g; 2.5 g), histidine (3.2 g; 2.67 g); l-phenylalinine
(5.7 g;4.75 g), l-proline (12.2 g; 10.17 g), l-serine (6.9 g;
5.75 g), l-threonine (6.5 g; 5.42 g), l-tyrosine (6.9 g; 5.75 g)
and l-valine (8.9 g; 7.42 g). The T+ drink contained l-
2140 Psychopharmacology (2017) 234:2139–2147
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

tryptophan (2.3 g; 1.92 g). Tastes were masked with 5 g
(15 cals; 1.3 g carb.) of citric (or malic) acid (cherry-and-
vanilla or grapefruit) and artificial sweetener.
Psychometric assessments Following the screening inter-
view, participants completed the (i) trait and state Positive
and Negative Affect Scales (Watson et al. 1988), (ii) Beck’s
Depression Inventory (BDI-II) (Beck et al. 1996), and (iii)
Buss-Perry Aggression Questionnaire (Buss and Perry
1992). The PANAS consists of two subscales of 20 items,
rated using a 5-point Likert scale, to measure positive and
negative aspects of emotional experience. Cronbach’sαcoef-
ficients for the state (i.e. momentary) version of the positive
affect and negative affect subscales have been reported as .90
and .87, respectively (Watson et al. 1988).
The BDI-II consists of 21 statements that capture the fre-
quency of depressive symptoms over the previous 14 days,
with total scores tested against cut-offs to indicate minimal to
severe depression (Beck et al. 1996). It has strong convergent
and criterion reliability. A Cronbach’sαcoefficient of 0.94
indicates high internal consistency (Arnau et al. 2001).
Finally, our participants completed the Standard Progressive
Raven’s Standard Progressive Matrices to assess non-verbal
cognitive ability (Raven et al. 2004).
Personality trait discrimination task
The trait discrimination task used composites of female faces,
reflecting the correlated facial appearance of women who
share similar levels of personality traits (e.g. Kramer and
Ward 2010; Jones et al. 2012). We used the full-face stimulus
set created by Kramer and Ward (2010), which was made by
averaging neutral face photos from women scoring highest
and lowest on self-report measures of each big five personality
traits (Agreeableness, Openness, Conscientiousness,
Extraversion and Neuroticism), as measured on the Mini
International Personality Item Pool (Mini-IPIP, Donnellan
et al. 2006). Thus, the composite face-pairs reflected regular-
ities in facial appearance which are correlated with specific
personality traits. In addition, the stimuli set included two
composites which were not based upon the Big Five traits:
physical health (used by Kramer and Ward 2010) and depres-
sive symptoms (as described by Scott et al. 2013). These two
additional composites were presented only once in each stim-
ulus set, providing insufficiently reliable patterns of discrimi-
nations; they are not discussed further.
On each presentation (hereafter, trial), one high and one low
composite for a personality trait were presented, randomly posi-
tioned on either side of centre on a standard computer display,
along with a discrimination statement relevant to that trait (e.g.
‘More prone to mood swings’for trait Neuroticism, see Fig. 1).
Participants made an unspeeded mouse-click on the face best
matching the discrimination statement. Each trait was presented
four times, using four different discrimination statements (items)
adapted directly from the Mini-IPIP items (Donnellan et al. 2006).
Procedure
All participants followed a low-protein diet for 24 h before
ingesting an amino-acid drink on the morning of the experi-
ment. On arrival at the laboratory, participants completed
baseline ratings of state positive and negative affect using
the PANAS (Watson et al. 1988). Blood samples (6 ml) were
taken for baseline measurements of plasma tryptophan levels
and tryptophan/large-neutral-amino-acid (LNAA) ratios.
Participants then ingested their allocated amino acid drinks.
Five hours later (+5 h), participants provided follow-up rat-
ings of state affect with the PANAS, as well as a second blood
sample. Participants then completed cognitive assessments
and social exchange games that included the computerized
trait discrimination task reported here.
Fig. 1 Face stimuli and example discrimination statements for a
Agreeableness, bExtraversion and cNeuroticism. In each case, the face
on the left is a composite of 15 women scoring highest on the trait
measure, and the face on the right a composite of 15 women scoring
lowest on that measure (see Kramer and Ward 2010 for details).
Participants were asked to identify which of the two faces best matched
the accompanying discrimination statement
Psychopharmacology (2017) 234:2139–2147 2141
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Data analysis
Differences in the number of males and female participants in
the T−and the T+ groups were tested with a χ
2
test. Group
matching of age, cognitive ability (scored with the Raven’s
Standard Progressive Matrices; Raven et al. 2004), trait affect
(scored with the Positive and Negative Affect Scale; Watson
et al. 1988) and recent depressive symptoms (scored with the
Beck Depression Inventory II; Beck et al. 1996)weretested
with two-sample ttests. Changes in total plasma tryptophan,
the ratio of plasma tryptophan to LNAAs, and state positive
and negative affect (as measured by the PANAS) were
assessed by separate repeated measures ANOVAs with the
single between-subject factor of treatment (T−vs T+) and
the within-subject factor of time (baseline vs +5 h).
Using the same full-face stimuli as here, Kramer and
War d (2010) demonstrated that healthy young adults
(comparable to our participants) are able to discriminate
reliably Neuroticism, Extraversion and Agreeableness but
not Openness or Conscientiousness. Similar accuracy
levels of trait discrimination are reported by Kramer and
War d (2011) using hemi-face versions of these stimuli
(that is, presenting just the left or right half of the stimuli).
Therefore, we focused our analyses upon the most robust-
ly discriminated traits of Neuroticism, Extraversion, and
Agreeableness. First, we used one-sample ttests to assess
all participants’discrimination accuracy against chance
(0.5) and two-sample ttests to test the prediction that T
−participants were less accurate in discriminating
Neuroticism and, perhaps, other traits compared to T+
participants. Fifty-two participants provide a statistical
power of approximately 0.8 at α= .05 (one-sided) to
detect medium effect sizes (0.5) in the one-sample ttests
and medium-to-large effect sizes (0.7) in the two-sample
tests.
Since this is the first study of tryptophan depletion and
the discrimination of personality traits from the face, we
also used a mixed-effects binomial model (Bates 2010)to
test whether the inclusion of the above variables in the
model moderated the reliability of treatment effects. The
model included random intercept terms for participant and
for item, and a random slope for treatment by item, and
fixed effects for the remaining predictors, comprising
tryptophan depletion, and the demographic and
psychometric variables above. All non-categorical predic-
tors (i.e. other than sex and treatment) were zero-centred
and scaled to variance of 1.
Results
The demographic and psychometric characteristics of the T−
and T+ participants are shown in Table 1. As per standard
practice (Rogers et al. 2003; Bilderbeck et al. 2014), the
groups were closely matched for age, cognitive ability (as
measured by the Raven’s Progressive Matrices (Raven et al.
2004)), recent depressive symptoms (as measured by the BDI
(Beck et al. 1996)), trait positive and negative affect (as mea-
sured with the trait version of the PANAS (Watson et al.
1988)) and trait aggression (Buss and Perry 1992), all
t(50) < 2.0.
As expected (Moore et al. 2000), plasma total trypto-
phan concentrations showed significant divergence be-
tween baseline and +5 h in the T−and T+ participants
(see Table 1), F(1,45) = 170.1, p< .00001. Specifically,
total tryptophan decreased in the T−participants from
10.2 to 4.0 μg/ml, t(22) = 7.20, p< .00001, but increased
in the T+ participants from 10.4 to 21.2 μg/ml,
t(23) = 11.09, p< .00001 (note: 3 T+ and 2 T−partici-
pants were excluded from these tests because of missing
blood samples in the morning or afternoon). The ratio of
plasma tryptophan to other LNAAs also showed different
changes following the T−and T+ drinks (see Table 1),
F(1, 45) = 69.6, p< .00001. Here, however, only the
tryptophan/LNAA ratios following the T−treatment were
reduced from .14 to .03, t(22) = 10.4, p< .00001, but the
change from .14 to .16 following the T+ treatment was
not significant, t(23) = 1.28, p= .215. At +5 h, both
Tabl e 1 Demographic and psychometric characteristics, plus plasma
tryptophan measurements, of 25 healthy adults who completed a
tryptophan depletion protocol and consumed an amino acid drink
without tryptophan (T−participants) and 27 healthy adults who
consumed an amino acid drink that did contain tryptophan (T+
participants)
T−participants T+ participants
Sex (male/female) 10:15 16:11
Age 25.2 (1.21) 24.1 (1.20)
Depressive symptoms (BDI) 0.68 (0.34) 1.30 (0.30)
Trait +ve affect (PANAS) 37.5 (1.38) 35.9 (1.15)
Trait −ve affect (PANAS) 11.4 (0.39) 12.9 (0.83)
Total Aggression (Buss-Perry) 44.4 (2.16) 50.3 (2.18)
Raven’s Matrices 53.4 (0.69) 54.4 (0.76)
Plasma trypt. (μg/ml) 0 h 10.2 (0.39) 10.4 (0.38)
Plasma trypt. (μg/ml) +5 h 4.0 (0.68) 21.2 (1.11)
Tryptophan LNAA ratio 0 h 0.14 (0.01) 0.14 (0.01)
Tryptophan LNAA ratio + 5 h 0.03 (0.01) 0.16 (0.01)
State +ve affect (PANAS) +0 h 28.2 (1.66) 29.0 (1.41)
State +ve affect (PANAS) +5 h 27.2 (1.58) 29.0 (1.59)
State −ve affect (PANAS) +0 h 12.2 (0.57) 12.6 (0.45)
State −ve affect (PANAS) +5 h 11.5 (0.58) 11.1 (0.27)
Beck’s Depression Inventory (BDI-II; Beck et al. 1996), trait and state
Positive and Negative Affect Schedule (PANAS; Watson et al. 1988),
Raven’s Matrices (Raven et al. 2004) and Aggression Questionnaire
(Buss and Perry 1992)
2142 Psychopharmacology (2017) 234:2139–2147
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

plasma total tryptophan and the tryptophan/LNAA ratios
were reduced in the T−participants compared to the T+
participants, t(45) = 12.4, p< .00001, and t(45) = 11.0,
p< .00001, respectively.
Overall, participants’state negative affect (as scored by the
PANAS; Watson et al. 1988) showed a significant decline
between baseline and +5 h (see Table 1), F(1, 50) = 12.0,
p< .002. State positive affect was unchanged, F(1, 50) < 1.
Tryptophan depletion did not produce any marked differential
changes in either state positive or negative affect (Table 1); all
Fs(1, 50) < 1.6, all ps > 0.2.
Trait accuracy Overall, and consistent with Kramer and Ward
(2010), participants discriminated Neuroticism with significant-
ly greater accuracy than chance by a one-sample ttest, M = .70,
SE = .035, t(51) = 5.7, p< .0001, d= .79. This was also true of
Extraversion, M = .75, SE = .038, t(51) = 6.65, p< .0001,
d= .92, and Agreeableness, M = .63, SE = .042, t(51) = 3.09,
p= .003, d= .43. Like Kramer and Ward (2010), our partici-
pants were unable to discriminate Openness reliably, M = .50,
SE = .037, t(51) = .13, p=.9,d= .02, but, unlike Kramer and
War d (2010), they were able to discriminate Conscientiousness,
M = .60, SE = .036, t(51) = 2.8, p= .008, d= .38.
Separate comparisons showed that the T−participants
showed significantly poorer discrimination of Neuroticism
compared to the T+ participants by a two-sample ttest (see
Fig. 2), t(50) = 2.23, p=.030,d= .62. By contrast, the dis-
crimination accuracies of the T−participants showed no sig-
nificant changes compared to the T+ participants for both
Extraversion and Agreeableness (see Fig. 2)o
rindeed
Openness and Conscientiousness (see Table 2), all ts(50) < 1,
ps > .3, d<.26).
Discrimination accuracy for Neuroticism was examined in
detail with a binomial mixed-effects model (Bates 2010)(see
Table 3). Female participants showed lower accuracy than
male participants, β=−0.796, SE = 0.360, p=0.027,and
the T−participants showed significantly lower accuracy in
identifying neurotic faces, β=−0.934, SE = .358,
p= 0.0091). No significant effect was found for other predic-
tors (all βs<0.33). The effect of treatment was numerically
larger for women than men (T+: women = .75, SE = .058;
men = .78, SE = .055; T−: women = .55, SE = .065,
men = .725, SE = .087). However, an additional treat-
ment × sex interaction term was not significant, β=0.332,
SE = 0.699, and did not improve the model, χ
2
(1) = 0.21.
Discussion
To our knowledge, this is the first demonstration that serotonin
function supports the perceptual discrimination of personality.
These findings show that tryptophan depletion—producing
temporary reductions in central serotonin activity (Moore
et al. 2000; Bilderbeck et al. 2014)—diminishes the capacity
of healthy adults to discriminate Neuroticism, but not other
traits, in other people on the basis of facial cues. The T−and
the T+ participants were closely matched in terms of sex, age
and cognitive ability. Consistent with previous studies involv-
ing healthy adult volunteers with no personal or family history
of depression (Rogers et al. 2003; Bilderbeck et al. 2014;
Wood et al. 2006), there were no marked differences between
our T−participants and T+ participants in terms of their state
positive or negative affect, either before or following con-
sumption of the amino acid drinks.
Fig. 2 Mean proportionate
accuracy on the two-alternative
forced-choice trait discrimination
of Agreeableness, Extraversion,
and Neuroticism in 27 healthy
participants who had consumed
an amino acid drink that did
contain tryptophan (T+
participants) and 25 healthy adult
participants who had consumed
an amino acid drink without
tryptophan (T−participants).
Error bars represent 1 standard
error
Psychopharmacology (2017) 234:2139–2147 2143
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Our findings expand our understanding of serotoninergic
mechanisms in processing of emotional and social signals in
other people. Substantial evidence attests to the monoaminer-
gic modulation of emotional recognition in the face (Walsh
and Harmer 2015) and the processing of other appetitive and
aversive signals (Boureau and Dayan 2011; Crockett et al.
2012; Hindi et al. 2012). Neuroticism is strongly associated
with fearfulness and vulnerability to anxiety, depression and
negative rumination (Fergusson et al. 1989;Nettle2006;
Vinberg et al. 2014; Verduyn and Brans 2012). Thus, our
finding that tryptophan depletion impairs the accuracy of
Neuroticism discrimination suggests that serotonergic activity
mediates the ability to discriminate subtle cues that signal
vulnerability to psychological distress in the absence of
expressed negative emotional states.
Serotonergic mechanisms influence the expression of so-
cial behaviours in both humans (Knutson et al. 1998;
Moskowitz et al. 2001;TseandBond2002; Young and
Leyton 2002) and non-human primates (Raleigh et al. 1981;
Raleigh et al. 1985) and, in particular, a wide variety of social
exchanges with partners and group members (Colzato et al.
2013; Higley et al. 1996; Crockett et al. 2012; Crockett et al.
2008;Woodetal.2006;Bilderbecketal.2014). Twelve days
administration of tryptophan in healthy (non-clinical) adults
decreases (perceived) quarrelsomeness and increases
dominance (Moskowitz et al. 2001,2003). The present find-
ings suggest that serotonergic influences over social behav-
iours also moderates sensitivity to the facial signals of vulner-
ability to distress in (potential or actual) social partners.
Important social behaviours—for example, the selection of
mates and the pursuit of affiliative relationships—are influ-
enced by the perception and then the appraisal of facial ap-
pearances (e.g. Rhodes 2006). The present data indicate that
serotonergic activity mediates these behaviours by its contri-
bution to perceptual sensitivity to cues of psychological
vulnerability.
These results also raise the possibility that psychological dis-
orders, such as depression, whose pathophysiology includes se-
rotonergic dysfunction (Bhagwagar and Cowen 2008;
Bhagwagar and Cowen 2008;Cowen2008) are associated with
mistaken inferences about other peoples’personalities, especially
around vulnerability to psychological distress and illness.
Accurate discrimination of mental health can be accompanied
by other negative and potentially incorrect inference. For
example, Scott et al. (2013)andDarosetal.(2016,Study2A)
report high levels of accuracy in judging depressive symptoms
from facial expressions, but importantly, the faces of individuals
with poor mental health were mistakenly judged as lower in
warmth and friendliness (Scott et al. 2013), and more negatively
in their emotional expressions (Daros et al. 2016). To the extent
that depressive symptoms are mediated by serotonergic dysfunc-
tion (Bhagwagar and Cowen 2008;Cowen2008), there may be a
cycle of psycho-pathogenic effects: (1) Healthy observers mis-
takenly appraise depressive individuals with negative character-
istics (Coyne 1976); (2) serotonin dysfunction in depression trig-
gers misidentification of personality of others. Together, these
misperceptions may disrupt social relationships in depressed
people.
Tryptophan depletion had no marked effects upon the
discrimination of other big five traits shown to our
participants (Agreeableness, Extraversion, Openness and
Conscientiousness), suggesting that serotonin activity plays a
particular role in the discrimination of Neuroticism. The
specificity of the effects of tryptophan depletion reported here
also rules out a single mechanism for personality discrimination
such as, for example, the selection of the more attractive of the
two composites presented as the more desirable characteristic
(i.e. highly Agreeable, Extraverted, and Stable). Rather, our
data suggest that, like the discrimination of emotional
expressions (e.g. Calder and Young 2005), trait inference based
upon facial characteristics depends upon dissociable mecha-
nisms. Future investigation will need to delineate the ex-
tent to which discrimination of personality from the face
share cognitive and neurochemical substrates with the dis-
crimination of emotional expressions (c.f. serotonin and
noradrenaline; Harmer et al. 2004).
The present study used the same stimuli and, broadly, rep-
licates the findings of Kramer and Ward (2010)and,in
Tab l e 3 Mixed-effects binomial modelling of discrimination on
Neuroticism trials
(Intercept) 1.00 (.375)
Sex (male = 0; female = 1) −.796 (.360)*
Age .131 (.190)
Buss-Perry total −.327 (.210)
BDI .061 (.193)
Trait PANAS +ve −.047 (.166)
Trait PANAS −ve −.267 (.189)
Ravens .004 (.178)
Treatment(T+=0;T−=1) −.934 (.358)**
Standardized weighting (standard error) for non-categorical factors
Tabl e 2 Mean discrimination accuracy (together with standard errors)
for Agreeableness, Extraversion, Neuroticism, Openness and
Conscientiousness in 25 healthy adults who had completed a tryptophan
depletion protocol and consumed an amino acid drink without tryptophan
(T−participants) and 27 matched healthy adults who had consumed an
amino acid drink that did contain tryptophan (T+ participants)
T−participants T+ participants
Agreeableness 0.63 (0.056) 0.63 (0.063)
Extraversion 0.75 (0.056) 0.75 (0.052)
Neuroticism 0.62 (0.054) 0.77 (0.040)
Conscientiousness 0.57 (0.059) 0.63 (0.045)
Openness 0.47 (0.056) 0.54 (0.048)
2144 Psychopharmacology (2017) 234:2139–2147
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

particular, the accurate discrimination of full-face composites
for the traits Agreeableness, Extraversion, and Neuroticism.
However, in contrast to Kramer and Ward, our participants
had completed a protocol that included prior dietary
(protein) restriction, consumption of amino acid drinks and a
full day of testing. Collectively, these two datasets, therefore,
suggest some robustness in the accuracy of individuals’dis-
crimination of these stimuli. However, in contrast to Kramer
and Ward (2010), our participants showed significantly better
than chance discrimination of Conscientiousness. Therefore,
at present, we cannot be certain about the consistency of
Conscientiousness identification across study protocols or
participant populations, and are reluctant to draw strong con-
clusions about the absence of changes in Conscientiousness
discrimination following tryptophan depletion.
In these data, women were significantly less accurate than
men in the discrimination of Neuroticism. This is somewhat
unusual, as women are often found to be more sensitive than
men in trait inference tasks, and are rarely found to be worse
(e.g. Hall and Mast 2008). This outcome probably reflects the
differential effects of the tryptophan depletion across sexes.
Tryptophan depletions’cognitive and emotional effects are
often greater in women than men (Booij et al. 2002), including
effects upon the recognition of fearful expressions (Harmer
et al. 2003). Sex differences involving tryptophan depletion
may reflect differences in hormonal activity or the distribution
of receptors (Fehr et al. 2000; Zhang et al. 1999). Here,
although not statistically significant, impaired discrimination
of Neuroticism following tryptophan depletion was
numerically more marked in the women than men.
Therefore, the significant main effect of sex upon may
reflect only the greater impact of depletion upon women’s
accuracy for Neuroticism.
Finally, changes in the recognition of facial expressions of
emotion following single treatments of serotonergic or norad-
renergic anti-depressants (Bhagwagar et al. 2004; Browning
et al. 2007) appear to form part of a complex of cognitive
adjustments that presage the therapeutic effects upon mood
following continued treatment in clinical populations (Walsh
and Harmer 2015). Further experimentation with tryptophan
loading protocols (Murphy et al. 2006,2009), single or sub-
chronic treatments with anti-depressants could indicate
whether the accuracy of Neuroticism inferences from the face
canbeimprovedbyenhancing serotonergic activity and other
personality inferences in treated and untreated depressed pa-
tients. Similarly, trait inferences about potential social partners
may regulate approach and withdrawal behaviours, highlight-
ing the issue of how serotonergic modulation of these judge-
ments relates to the activity of the inter-dependent peptide
systems implicated in both mood and social bonding, such
as oxytocin (Dolen 2015;Flight2013).
Notwithstanding these future possibilities, these data
provide preliminary evidence that transient reductions in
serotonin activity, produced by tryptophan depletion, impair
the discrimination between high versus low Neuroticism on
the basis of facial cues. These findings suggest that serotonin
activity modulates the detection of physical signals of
vulnerability to depression and psychological distress.
Acknowledgements The study was supported by a grant from the
Centre for Reputation Studies at the Said Business School, Oxford
University. Rogers holds a consultancy agreement with Pfizer and grants
from the Medical Research Council.
Open Access This article is distributed under the terms of the Creative
Commons Attribution 4.0 International License (http://
creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided you give appro-
priate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made.
References
Ambady N, Rosenthal R (1992) Thin slices of expressive behavior as
predictors of interpersonal consequences: A meta-analysis.
Psychological Bulletin. doi:10.1037/0033-2909.111.2.256
Anderson IM, Juhasz G, Thomas E, Downey D, McKie S, Deakin JFW,
Elliott R (2011) The effect of acute citalopram on face emotion
processing in remitted depression: a pharmacoMRI study. Eur
Neuropsychopharmacol 21(1):140–148. doi:10.1016/j.euroneuro.
2010.06.008
Arnau RC, Meagher MW, Norris MP, Bramson R (2001) Psychometric
evaluation of the Beck Depression Inventory-II with primary care
medical patients. Health Psychol 20:112–119
Bates, D. M. (2010). lme4: Mixed-effects modeling with R. URL http://
lme4.r-forge.r-project.org/book
Beck A, Steer R, Brown G (1996) Beck depression Inventory manual,
2nd edn. The Psychological Corporation, San Antonio
Bhagwagar Z, Cowen PJ, Goodwin GM, Harmer CJ (2004)
Normalization of enhanced fear recognition by acute SSRI treatment
in subjects with a previous history of depression. Am J Psychiatry
161(1):166–168
Bhagwagar Z, Cowen PJ (2008) ‘It’s not over when it’sover’: persistent
neurobiological abnormalities in recovered depressed patients.
Psychol Med 38:307–313
Bilderbeck AC, Brown GDA, Read J, Woolrich M, Cowen PJ, Behrens
TE, Rogers RD (2014) Serotonin and social norms tryptophan de-
pletion impairs social comparison and leads to resource depletion in
a multiplayer harvesting game. Psychol Sci 25(7):1303–1313
Booij L, Van der Does AJW (2011) Emotional processing as a predictor
of symptom change: An acute tryptophan depletion study in de-
pressed patients. European Neuropsychopharmacology 21(5):379–
383
Booij L, Van der Does W, Benkelfat C, Bremner JD, Cowen PJ, Fava M,
Gillin C, Leyton L, Moore P, Smith KA, Van der Kloot WA (2002)
Predictors of mood response to acute tryptophan depletion: a reanal-
ysis. Neuropsychopharmacology 27:852–861
Borkenau P, Liebler A (1992) Trait Inferences: Sources of Validity at Zero
Acquaintance. Journal of Personality and Social Psychology 62(4):
645–657
Borkenau P, Brecke S, Möttig C, Paelecke M (2009) Extraversion is
accurately perceived after a 50-ms exposure to a face. Journal of
Research in Personality 43(4):703–706. doi:10.1016/j.jrp.2009.03.
007
Psychopharmacology (2017) 234:2139–2147 2145
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Boureau YL, Dayan P (2011) Opponency revisited: competition and cooper-
ation between dopamine and serotonin. Neuropsychopharmacology 36:
74–97
Browning M, Reid C, Cowen PJ, Goodwin GM, Harmer CJ (2007) A
single dose of citalopram increases fear recognition in healthy sub-
jects. J Psychopharmacol 21:684–690
Buss AH, Perry M (1992) The aggression questionnaire. J Pers Soc
Psychol 63:452–459
Calder AJ, Young AW (2005) Understanding the recognition of facial
identity and facial expression. Nat Rev Neurosci 6(8):641–651
Carney DR, Colvin CR, Hall JA (2007) A thin slice perspective on the
accuracy of first impressions. Journal of Research in Personality
41(5):1054–1072. doi:10.1016/j.jrp.2007.01.004
Colzato LS, Steenbergen L, de Kwaadsteniet EW, Sellaro R, Liepelt R,
Hommel B (2013) Tryptophan promotes interpersonal trust. Psychol
Sci. doi:10.1177/0956797613500795
Cowen PJ (2008) Serotonin and depression: pathophysiological mecha-
nism or marketing myth? Trends Pharmacol Sci 29:433–436
Coyne JC (1976) Depression and the response of others. J Abnorm
Psychol 85(2):186–193
Crockett MJ, Clark L, Tabibnia G, Lieberman MD, Robbins TW (2008)
Serotonin modulates behavioral reactions to unfairness. Science
320:1739
Crockett MJ, Clark L, Apergis-Schoute AM, Morein-Zamir S, Robbins
TW (2012) Serotonin modulates the effects of Pavlovian aversive
predictions on response vigor. Neuropsychopharmacology 37:
2244–2252
Daros AR, Ruocco AC, Rule NO (2016) Identifying mental disorder from
the faces of women with borderline personality disorder. J
Nonverbal Behav 40(4):255–281. doi:10.1007/s10919-016-0237-9
Dolen G (2015) Autism: oxytocin, serotonin, and social reward. Soc
Neurosci 10:450–465
Donnellan MB, Oswald FL, Baird BM, Lucas RE (2006) The mini-IPIP
scales: tiny-yet-effective measures of the Big Five factors of person-
ality.PsycholAssess18(2):192
Fehr C, Szegedi A, Anghelescu I, Klawe C, Hiemke C, Dahmen N (2000)
Sex differences in allelic frequencies of the 5-HT2C Cys23Ser poly-
morphism in psychiatric patients and healthy volunteers: findings
from an association study. Psychiatr Genet 10:59–65
Fergusson DM, Horwood LJ, Lawton JM (1989) The relationship be-
tween neuroticism and depressive symptoms. Soc Psychiatry
Psychiatr Epidemiol 24(6):275–281. doi:10.1007/BF01788029
First MB, Spitzer RL, Gibbon M, Williams JBW (2002) Structured clin-
ical interview for DSM-IV-TR axis I disorders - patient edition (re-
search version). Biometrics Research, New York
Flight MH (2013) Cognitive neuroscience: oxytocin and serotonin make
it worthwhile. Nat Rev Neurosci 14:740–741
Funder DC, Colvin CR (1988) Friends and Strangers: Acquaintanceship,
Agreement, and the Accuracy of Personality Judgment. Journal of
Personality and Social Psychology 55(1):149–158. doi:10.1037/
0022-3514.55.1.149
Gonda X, Fountoulakis K, Juhasz G, Rihmer Z, Lazary J, Laszik A,
Bagdy G (2009) Association of the allele of the 5-HTTLPR with
neuroticism-realted traits and temperaments in a psychiatrically
healthy population. Eur Arch Psychiatry Clin Neurosci 259:106–
113. doi:10.1007/s00406-008-0842-7
Hall JA, Schmid Mast M (2008) Are women always more interpersonally
sensitive than men? Impact of goals and content domain. Personality
and Social Psychology Bulletin 34(1):144–155
Harmer CJ, Rogers RD, Tunbridge E, Cowen PJ, Goodwin GM (2003)
Tryptophan depletion decreases the recognition of fear in female
volunteers. Psychopharmacology 167(4):411–417. doi:10.1007/
s00213-003-1401-6
Harmer CJ, Shelley NC, Cowen PJ, Goodwin GM (2004) Increased
positive versus negative affective perception and memory in healthy
volunteers following selective serotonin and norephinephrine reup-
take inhibition. Am J Psychiatry 161(7):1256–1263
Harmer CJ, Mackay CE, Reid CB, Cowen PJ, Goodwin GM (2006)
Antidepressant drug treatment modifies the neural processing of
nonconscious threat cues. Biol Psychiatry 59(9):816–820. doi:10.
1016/j.biopsych.2005.10.015
Hayward G, Goodwin GM, Cowen PJ, Harmer CJ (2005) Low-dose
tryptophan depletionin recovered depressed patients induces chang-
es in cognitive processing without depressive symptoms. Biol
Psychiatry 57(5):517–524. doi:10.1016/j.biopsych.2004.11.016
Higley JD, Mehlman PT, Poland RE, Taub DM, Vickers J, Suomi SJ,
Linnoila M (1996) CSF testosterone and 5-HIAA correlate with differ-
ent types of aggressive behaviors. Biol Psychiatry 40:1067–1082
Hindi AC, Finckh B, Buchel C (2012) The influence of serotonin on fear
learning. PLoS One 7:e42397
Jones AL, Kramer RS, Ward R (2012) Signals of personality and health:
the contributions of facial shape, skin texture, and viewing angle. J
Exp Psychol Hum Percept Perform 38(6):1353
Knutson B, Wolkowitz OM, Cole SW, Chan T, Moore EA, Johnson RC,
Terpstra J, Turner RA, Reus VI (1998) Selective alteration of per-
sonality and social behavior by serotonergic intervention. Am J
Psychiatry 155:373–379
Kramer RSS, Ward R (2010) Internal facial features are signals of per-
sonality and health. Q J Exp Psychol 63(11):2273–2287. doi:10.
1080/17470211003770912
Kramer RSS, Ward R (2011) Different signals of personality and health
from the two sides of the face. Perception 40(5):549–562. doi:10.
1068/p6856
Little AC, Perrett DI (2007) Using composite images to assess accuracy
in personality attribution to faces. Br J Psychol 98:111–126. doi:10.
1348/000712606X109648
McWilliams L (2003) Neuroticism and depression. Br J Psychiatry
182(1):80–80. doi:10.1192/bjp.182.1.80
Moore P, Landolt HP, Seifritz E, Clark C, Bhatti T, Kelsoe J, Rapaport M,
Gillin JC (2000) Clinical and physiological consequences of rapid
tryptophan depletion. Neuropsychopharmacology 23:601–622
Moskowitz DS, Pinard G, Zuroff DC, Annable L, Young SN (2001) The
effect of tryptophan onsocial interactionin everyday life: a placebo-
controlled study. Neuropsychopharmacology 25:277–289
Moskowitz DS, Pinard G, Zuroff DC, Annable L, Young SN (2003)
Tryptophan, serotonin and human social behavior. Adv Exp Med
Biol 527:215–224
Murphy SE, Longhitano C, Ayres RE, Cowen PJ, Harmer CJ (2006)
Tryptophan supplementation induces a positive bias in the process-
ing of emotional material in healthy female volunteers.
Psychopharmacology 187:121–130
Murphy SE, Longhitano C, Ayres RE, Cowen PJ, Harmer CJ, Rogers RD
(2009) The role of serotonin in nonnormative risky choice: the ef-
fects of tryptophan supplements on the Breflection effect^in healthy
adult volunteers. J Cogn Neurosci 21:1709–1719
Naumann LP, Vazire S, Rentfrow PJ, Gosling SD (2009) Personality
judgments based on physical appearance. Personality and Social
Psychology Bulletin 35(12):1661–1671. doi:10.1177/
0146167209346309
Nettle D (2006) The evolution of personality variation in humans and
other animals. Am Psychol 61(6):622–631. doi:10.1037/0003-
066X.61.6.622
Pflüger LS, Oberzaucher E, Katina S, Holzleitner IJ, Grammer K (2012)
Cues to fertility: Perceived attractiveness and facial shape predict
reproductive success. Evolution and Human Behavior 33(6):708–
714. doi:10.1016/j.evolhumbehav.2012.05.005
Raleigh MJ, Brammer GL, McGuire MT, Yuwiler A (1985) Dominant
social status facilitates the behavioral effects of serotonergic ago-
nists. Brain Res 348:274–282
Raleigh MJ, Yuwiler A, Brammer GL, McGuire MT, Geller E, Flannery
JW (1981) Peripheral correlates of serotonergically-influenced
2146 Psychopharmacology (2017) 234:2139–2147
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

behaviors in vervet monkeys (Cercopithecus Aethiops Sabaeus).
Psychopharmacology 72:241–246
Raven J, Raven J, Court J (2004) Manual for Raven’s Progressive
Matrices and vocabulary scales. Harcourt Assessment, San Antonio
Rhodes G (2006) The evolutionary psychology of facial beauty. Annu
Rev Psychol 57:199–226
Rogers RD, Tunbridge EM, Bhagwagar Z, Drevets WC, Sahakian BJ,
Carter CS (2003) Tryptophan depletion alters the decision-making
of healthy volunteers through altered processing of reward cues.
Neuropsychopharmacology 28:153–162
Rule NO, Macrae CN, Ambady N (2009) Ambiguous Group
Membership Is ExtractedAutomatically From Faces. Psychological
Science 20(4):441–443. doi:10.1111/j.1467-9280.2009.02314.x
Scott NJ, Kramer RSS, Jones AL, Ward R (2013) Facial cues to depres-
sive symptoms and their associated personality attributions.
Psychiatry Res 208(1):47–53. doi:10.1016/j.psychres.2013.02.027
Stirrat M, Perrett DI (2010) Valid facial cues to cooperation and trust male
facial width and trustworthiness. Psychol Sci 21(3):349–354. doi:
10.1177/0956797610362647
Takano A, Arakawa R, Hayashi M, Takahashi H, Ito H, Suhara T (2007)
Relationship between neuroticism personality trait and serotonin
transporter binding. Biol Psychiatry 62(6):588–592. doi:10.1016/j.
biopsych.2006.11.007
Tse WS, Bond AJ (2002) Serotonergic intervention affects both social
dominance and affiliative behaviour. Psychopharmacology 161:
324–330
Verduyn P, Brans K (2012) The relationship between extraversion, neu-
roticism and aspects of trait affect. Personal Individ Differ 52(6):
664–669. doi:10.1016/j.paid.2011.12.017
Vinberg M, Miskowiak K, Kessing LV (2014) Serotonin transporter ge-
notype, salivary cortisol, neuroticism and life events: impact on
subsequent psychopathology in healthy twins at high and low risk
for affective disorder. Prog Neuro-Psychopharmacol Biol Psychiatry
48(0):193–198. doi:10.1016/j.pnpbp.2013.10.007
Walsh AEL, Harmer CJ (2015) The cognitive neuropsychological model
of antidepressant response. Curr Opin Psychol 4:124–130
Watson D, Clark LA, Tellegen A (1988) Development and validation of
brief measures of positive and negative affect: the PANAS scales. J
Pers Soc Psychol 54:1063–1070
Wood RM, Rilling JK, Sanfey AG, Bhagwagar Z, Rogers RD (2006) Effects
of tryptophan depletion on the performance of an iterated Prisoner’s
dilemma game in healthy adults. Neuropsychopharmacology 31:
1075–1084
Young SN, Leyton M (2002) The role of serotonin in human mood and
social interaction. Insight from altered tryptophan levels. Pharmacol
Biochem Behav 71:857–865
Zhang L, Ma W, Barker JL, Rubinow DR (1999) Sex differences in
expression of serotonin receptors (subtypes 1A and 2A) in rat brain:
a possible role for testosterone. Neuroscience 94:251–259
Psychopharmacology (2017) 234:2139–2147 2147
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