Content uploaded by Jorg Huber
Author content
All content in this area was uploaded by Jorg Huber on Jan 10, 2019
Content may be subject to copyright.
Does alcohol consumption really affect asymmetry
perception? A three-armed placebo-controlled
experimental studyadd_3807 1..7
Lewis G. Halsey, Joerg W. Huber & Jennifer C. Hardwick
Department of Life Sciences, University of Roehampton, London, UK
ABSTRACT
Aims A possible explanation for increased levels of attractiveness of faces when under the influence of alcohol is the
reduced ability to perceive bilateral asymmetry. This study tested the degree of preference by alcohol-dosed and
non-alcohol-dosed participants for symmetrical faces and their ability to detect facial symmetry, while controlling for
other explanations. Design Volunteers were recruited to a random allocation experiment with three conditions:
alcoholic drink (alcohol dosed), non-alcoholic drink (placebo) and diluted orange cordial (control). Data on concen-
tration, personality and demographics were collected. Dependent variables were symmetry preference and detection.
Setting Laboratory, University of Roehampton. Participants A total of 101 participants, mainly students (41
alcohol-dosed, 40 placebo, 20 control). Measurements Participants provided verbal responses to images of faces
which were presented on a computer screen for 5 seconds each; the first task required a preference judgement and the
second task consisted of a forced-choice response of whether or not a face was symmetrical. Levels of concentration,
weight and level of alcohol dose were measured, and demographics plus additional psychological and health informa-
tion were collected using a computer-based questionnaire. Findings In contrast to a previous investigation, there
was no difference in symmetry preference between conditions (P=0.846). In agreement with previous findings,
participants who had not drunk alcohol were better at detecting whether a face was symmetrical or asymmetrical
(P=0.043). Measures of concentration did not differ between conditions (P=0.214–0.438). Gender did not affect
ability to detect symmetry in placebo or alcohol-dosed participants (P=0.984, 0.499); however, alcohol-dosed females
were shown to demonstrate greater symmetry preference than alcohol-dosed males (P=0.004). Conclusions People
who are alcohol-dosed are subtly less able to perceive vertical, bilateral asymmetry in faces, with gender being a possible
moderating factor.
Keywords Acute alcohol consumption, attraction, faces, symmetry perception.
Correspondence to: Lewis G. Halsey, Depar tment of Life Sciences, University of Roehampton, London SW15 4JD, UK. E-mail: l.halsey@roehampton.ac.uk
Submitted 17 June 2011; initial review completed 22 July 2011; final version accepted 13 January 2012
INTRODUCTION
A range of facial cues may play important roles in human
mate selection. These include the femininity of face shape
[1], gaze direction [2] and also, perhaps more subtly, bilat-
eral symmetry; ratings of attractiveness by people about
members of the opposite sex are higher when symmetry
is greater [3–5]. Such ratings are also affected positively
by acute alcohol consumption [6,7], at least in certain
circumstances [8]. In part, this may be explained by
alcohol dosing decreasing the ability to detect bilateral
asymmetry [9,10], due presumably to reductions in level
of visual function [11–13].
The studies by Souto et al. [9] and Halsey et al. [10]
were undertaken in the ‘field’. Thus, while they provided
ecological validity [14], they could not include controls
for certain potential confounds such as differences in per-
sonality type or levels of chronic alcohol consumption
between the alcohol-dosed and non-alcohol-dosed par-
ticipants. Indeed, there is evidence that conscientious-
ness, excitement-seeking, boredom susceptibility and
chronic alcohol consumption affect asymmetry detection
bs_bs_banner
RESEARCH REPORT doi:10.1111/j.1360-0443.2012.03807.x
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
[15]. Furthermore, it is possible that the alcohol-dosed
participants experienced changes alongside deterioration
in visual perception that affected performance on the
task, for example reduced concentration.
The present study progresses previous work [10]
by again testing whether acute alcohol consumption
decreases the ability to detect asymmetry in faces and
whether such consumption reduces preference for sym-
metrical faces, but using a three-armed, randomized,
double-blinded trial. Our hypotheses repeat those of the
earlier study: that alcohol-dosed participants are less able
to detect asymmetry in faces and have a reduced prefer-
ence for symmetrical faces over asymmetrical faces.
MATERIALS AND METHODS
Design and procedure
Participants were assigned randomly to one of three con-
ditions defined by type of drink consumed by the partici-
pant prior to undertaking the tasks: an alcoholic drink
(alcohol-dosed), a similar-tasting non-alcoholic drink
(placebo) or diluted orange cordial (control). In the case
of the alcohol-dosed and placebo conditions, each partici-
pant and the experimenter were not made aware until the
end of the experiment whether or not the participant had
drunk alcohol, while in the control condition both parties
knew that the drink was non-alcoholic. Twenty to 40
minutes after finishing their drink, participants under-
took tests of concentration and then two facial percep-
tion tasks. Participants also completed the Neuroticism
Extroversion Openness–Personality Inventory Revised
(NEO PI-R) and the Alcohol Use Disorders Identification
Test (AUDIT) questionnaires. Full details of the methods
are provided in the online supplement; an abridged
version is presented below.
Drinks
Participants in the alcohol-dosed condition received a
drink containing 0.7 g/kg body mass of alcohol. The
drink comprised vodka at 37.5% proof and tonic water.
Thus, a 72 kg participant received about 6 units of
alcohol within approximately 1 pint (0.6 l) of liquid.
The similar-tasting, non-alcoholic drink for the placebo
condition comprised only tonic water. All drinks were
provided chilled and flavoured with either lime or black-
currant cordial at each participant’s choice.
Participants and location
A total of 101 participants (31 males and 70 females)
undertook the tasks. Some data sets include data for
fewer than the total number of participants. The proce-
dures followed were approved by the ethics committee at
the University of Roehampton and were in accordance
with the Helsinki Declaration of 1975, as revised in
1983.
Tests
The facial perception tasks assessed whether and how
much preference participants had for bilaterally sym-
metrical faces over asymmetrical faces (the preference
task) and how well they could perceive asymmetry in a
face (the detection task). In both cases the faces were
presented on a laptop computer screen. For the prefer-
ence task, they were shown a series of 20 paired faces (10
male), where the two faces of a pair were of the same
person and had been manipulated such that one was
symmetrical and the other asymmetrical, and were asked
to choose the more attractive face. For the detection task,
they were shown a series of 20 single faces (10 male) to
which they responded on whether they perceived the
image to be symmetrical or asymmetrical (three male
faces and four female faces were symmetrical, leaving 13
asymmetrical faces). For the preference task the depen-
dent variable was the number of paired faces of which
the participants preferred the symmetrical face, and for
the detection task it was the number of correctly iden-
tified asymmetrical faces. These tasks replicated those
described previously [10], where details of the photo-
graphs used are also provided. Figure 1 shows an
example of the faces used. For the present study, some of
the asymmetrical faces were adjusted subtly to make their
asymmetry less obvious (Fig. 1).
The two concentration tests comprised a standard
letter cancellation task [16] and the rotary pursuit task.
Participants had to repeat the rotary pursuit task eight
times, with a 10-second break between each task. A
simple metric of concentration was derived from these
data by determining any change in the amount of ‘time on’
the target between earlier and later attempts at the task.
Questionnaires
Conscientiousness was assessed using a subscale of
the five-factor personality inventory (NEO PI-R; [17]).
Possible excessive alcohol consumption was assessed by
asking for levels of alcohol consumption and through the
AUDIT [18]. The amount of alcoholic drinks consumed
were obtained retrospectively for the last 7 days in terms
of units of alcohol per day.
Analysis
Values are provided as means ⫾standard error of the
mean (SEM), unless stated as standard deviation (SD). To
test for a preference in each condition for symmetrical
faces, single-sample t-tests were run to compare the
2Lewis G. Halsey et al.
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
number of times in the preference task in which the sym-
metrical face was preferred compared to the number of
times such a face should be preferred if symmetry does
not influence attraction (i.e. by chance and hence 50%
of the time; 10 times). As in Halsey et al. [10], analyses
between conditions were undertaken at the item
level. Repeated-measures one-way analysis of variance
(ANOVA) was used to test for differences between the
three conditions. Where data for two conditions were
pooled, or when testing for differences between the sexes,
paired t-tests were used. A Pearson’s correlation was cal-
culated on preference for the symmetrical face of a face
pair against the ability to detect symmetry or asymmetry
in faces across all participants. Spearman’s correlation
was used to test for relationships between symmetry
preference/symmetry detection and multiple potentially
confounding variables such as personality traits. Statis-
tics were run in Excel 2010 (Microsoft Corp., Washing-
ton, USA) and SPSS version 19 (IBM Inc., NY, USA).
RESULTS
Of the 101 participants, 41 were in the alcohol-dosed
condition, 40 in the placebo condition and 20 in the
control condition. Two-thirds of participants were white
(65%), with the remainder of the sample coming from a
range of ethnic backgrounds. Participants in the alcohol-
dosed condition were given on average 46.2 ⫾SD 9.5 g of
alcohol (5.8 units). The research assistant was accurate
about whether or not the participants had drunk alcohol
70% of the time, which was significantly better than
chance (c2=13.71, P<0.001). The participants were
accurate 77% of the time (c2=24.98, P<0.001). Partici-
pants who thought correctly that they had drunk alcohol
on average rated their level of drunkenness out of 10 at
3.3 ⫾SD 1.8. The second breathalyser test returned
an average reading of 27.3 ⫾SD 6.9 mg/100 ml blood
for alcohol-dosed participants, and the third test 28.3 ⫾
SD 6.5 mg/100 ml. There was no significant difference
between the second and third breathalyser readings
(t38 =1.53, P=0.134). Assessment of conscientiousness
returned a mean t-score for all participants of 38 (range
20–65), generally indicating low scores, given that the
mean is more than 1 SD below the UK norm. Participants
in the three conditions did not differ in levels of conscien-
tiousness (F(2,97) =2.78, P=0.11). Reported alcohol con-
sumption during the previous week ranged from 0 units
(10% of sample) to a maximum of 82 units (mean: males,
20.5 ⫾19.7 units; females, 9.8 ⫾10.1). The AUDIT
indicated that 40% of the sample showed no signs of risk
of alcohol use disorder, 50% showed signs of low risk and
the remainder higher levels. Participants in the three con-
ditions did not differ in levels of alcohol consumption or
disorder risk (F(2,97) =1.07, P=0.35 and F(2,97) =0.27,
P=0.76, respectively).
For each condition, the frequency distribution of the
data both for the preference task and the detection task
and for the two concentration tasks did not differ signifi-
cantly from normal (one-sample Kolmogorov–Smirnov
tests, P>0.05 in all cases).
Comparisons of preference for and detection of
asymmetry between conditions
In all conditions there was a preference for symmetrical
over asymmetrical faces (preference task, single-sample
Figure 1 A pair of faces shown to participants in the preference task (enclosed).The face on the left is asymmetrical, while the face on the
right is symmetrical. Participants were asked which face they found more attractive. The far left face is the asymmetrical version of the face
presented to participants in Halsey et al. [10], and has a forehead that is fairly clearly asymmetrical once the top edge of the image is inspected
Alcohol consumption and asymmetry perception 3
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
t-test, placebo: t39 =7.52, Pⱕ0.001; drunk: t40 =7.22,
Pⱕ0.001; control: t19 =6.86, Pⱕ0.001). There was
no significant difference between conditions in how
much symmetrical faces were preferred over asym-
metrical faces (one-way ANOVA, F(2,98) =0.17,
P=0.846), with an average preference across partici-
pants of 67.3 ⫾1.5%.
When participants were asked to decide whether a
face was symmetrical or asymmetrical (detection task)
across the full set of faces shown, placebo condition
participants made errors with a frequency not signifi-
cantly different to that of control condition participants
(t19 =0.14, P=0.890), and thus these data were com-
bined to compare to the alcohol-dosed data. Placebo- and
control-condition participants combined made errors less
frequently than alcohol-dosed participants (number of
wrong answers: t19 =2.17, P=0.043). Figure 2 shows
the comparison in detection ability per asymmetrical face
for alcohol-dosed versus placebo- and control-condition
participants combined.
There was a significant, positive, although weak,
correlation between preference for symmetry and ability
to perceive symmetry or asymmetry across all partici-
pants (r=0.28, P=0.005).
The participants in the alcohol-dosed group, who
were allowed a relatively short time to digest the alcohol
they had drunk, made fewer errors than alcohol-dosed
participants who were allowed a relatively long time to
digest the alcohol (t19 =2.20, P=0.040; Table 1).
Comparisons of preference for and detection of
asymmetry between sexes
There was no significant difference in the responses of
male and female participants to the preference task in
the placebo or control conditions (placebo: t19 =1.03,
P=0.230; control: t19 =0.59, P=0.328). However,
there was a significant difference in the responses of
alcohol-dosed male and female participants (t19 =3.36,
P=0.004; Fig. 3). There was no significant difference in
Figure 2 The percentage of alcohol-dosed (open circles; n=41) and non-alcohol-dosed (placebo and control condition combined; closed
circles; n=60) participants who perceived an asymmetrical face as being symmetrical.The faces are numbered in the order that they were
shown to participants. The values on the far right are means of all asymmetrical faces, and the associated bars represent 1 standard error of
the mean
Table 1 Mean ⫾standard error of the mean (SEM) number of errors out of 20 in perceiving faces as either symmetrical or
asymmetrical.
Field Laboratory
Non-alcohol-dosed 3.53 ⫾0.36 (36) 4.10 ⫾0.40 (40)
Short digestion time Long digestion time
Alcohol-dosed 4.71 ⫾0.51 (28) 4.00 ⫾0.41 (15) 5.12 ⫾0.47* (26)
Difference 1.18 -0.10 1.02
*P<0.05. Participants were either alcohol-dosed or non-alcohol-dosed, and undertook the experiment in either a field setting [10] or a laboratory setting
(present study). Those in the laboratory experiment either had a relatively short period of time to digest the alcohol consumed (approximately 20
minutes) or a relatively long period of time (approximately 40 minutes). Values in brackets are n.
4Lewis G. Halsey et al.
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
the responses of the sexes to the detection task either in
placebo-condition participants (t19 =0.02, P=0.984) or
alcohol-dosed participants (t19 =0.69, P=0.499).
Comparisons of concentration between conditions
Time taken to complete the letter cancellation task
did not differ significantly between the conditions
(F(2,92) =0.83, P=0.438) or in the number of errors
made (F(2,92) =0.95, P=0.389). There was no significant
difference between the conditions in the change in ‘time
on’ from early to late attempts of the rotary pursuit task
(F(2,60) =1.58, P=0.214).
Personality characteristics
There was no significant difference in conscientiousness
between the three groups (F=2.23, P=0.11). In addi-
tion, facets of the conscientiousness scale (competence,
order, dutifulness, achievement-striving, self-discipline
and deliberation) did not differ significantly (P-value
range 0.07–0.58).
Testing for confounders
Relationships between symmetry preference/symmetry
detection and the following variables were tested for: con-
centration metrics, personality traits, recent alcohol use,
risk of alcohol use disorder, breathalyser readings and
self-assessment of degree of drunkenness. Relationships
were tested across all participants, and also specifically
for the placebo group and the alcohol-dosed group alone.
Across this large number of tests, the absolute values of
the correlation coefficients were typically lower than 0.1,
and nearly always below 0.3. Only a few correlations
returned coefficients greater than 0.5, and they were not
significant. Therefore, these variables were not included
in further analyses as covariates.
DISCUSSION
In general, the main findings of this laboratory study
agree with those of the similar field study [10]. Both
alcohol-dosed and non-alcohol-dosed participants had
a preference for symmetrical faces (see also [3,4]); how-
ever, alcohol-dosed participants were less able to detect
whether faces were symmetrical or asymmetrical (Fig. 2).
Furthermore, symmetry preference and detection corre-
late positively, although only weakly (see also [19]).
The earlier study by Halsey et al. [10] was not able to
control for certain potential confounds, in contrast to the
high level of control of the present study. The current
study was able to check that personality type and scores
of alcohol use disorder were not confounds. However,
unfortunately the conscientiousness scores of all partici-
pants were, on average, somewhat lower than normal,
which might mean that they tended to undertake the
tasks less carefully; 77% of participants ascertained cor-
rectly what experimental condition they were in and thus
the less conscientious participants who knew that they
were alcohol-dosed may have been prone to exaggerat-
ing their state. The present study also found no effect of
acute alcohol consumption on concentration, providing
evidence to rule out the possibility that alcohol-dosed
participants obtained lower scores on the symmetry
detection test simply because they had reduced levels of
Figure 3 The percentage of female (open circles; n=70) and male (closed circles; n=31) participants who preferred the symmetrical face
of pairs of faces. For each pair, the two faces were of the same person, one image being symmetrical, the other asymmetrical. The face pairs
are numbered in the order that they were shown to participants. The values on the far right are means of all face pairs, and the associated
bars represent 1 standard error of the mean
Alcohol consumption and asymmetry perception 5
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
concentration. However, a caveat here is that the experi-
mental design does not rule out the possibility that acute
consumption of greater quantities of alcohol than those
consumed in the present study affects concentration
negatively or that relatively chronic drinking (i.e. over
many hours) can also have such an effect; two factors
relevant in Halsey et al. [10]. The control-condition par-
ticipants did not exhibit a difference in facial perception to
the placebo-condition participants, providing some evi-
dence that it is not simply a belief that alcohol has been
consumed which has an effect. However, as with Parker
et al. [7], the double-blind nature of the study was not
always upheld because more often than by chance the
research assistant and/or the participant had judged cor-
rectly, at least by the end of the study, whether alcohol
had been consumed. Overall, however, the findings of the
present study provide further evidence to support previ-
ous claims [10] in suggesting that alcohol dosing reduces
the ability to detect facial asymmetry and that this may be
an important mechanism underlying higher ratings of
facial attractiveness for members of the opposite sex
when under the influence of alcohol [6].
However, in contrast to [10], there was no difference
in the preference for symmetrical faces between the
placebo condition and alcohol-dosed participants. Along
with the fairly weak correlation between symmetry pref-
erence and detection found in both studies, there is now
quite clear evidence that the impact that reduction in
detection ability has on symmetry preference is limited.
Furthermore, the difference in ability to detect when
faces are symmetrical and when they are asymmetrical
between the placebo condition and alcohol-dosed partici-
pants was smaller in the present study than in the similar
field study [10]. While the number of errors was higher
in the present study in both conditions, which is expected
due to the, on average, slightly reduced asymmetry in
the asymmetric faces in the present study compared to
Halsey et al. [10], the relative increase in errors by the
alcohol-dosed participants was smaller. Perhaps the most
likely reason for this is greater levels of alcohol dosing in
the participants in the field study [10] compared to the
present study, where the mean second and third breath-
alyser readings were lower than the UK legal driving limit
for breath alcohol of 35 ml/100 ml. While it is not pos-
sible to compare the breathalyser readings reported
in the previous study with those in the present study,
because the previous study was only able to report read-
ings with low precision, Halsey et al. [10] provide infor-
mation on the estimated number of units drunk by
alcohol-dosed participants. This is considerably higher
than the measured units provided to participants in
the present study (13.6 ⫾SD 5.0 versus 5.6 ⫾SD 1.1
units). With the caveats that self-reports of alcohol
consumption during binge drinking may tend to be
exaggerations, and that the alcohol-dosed participants in
the previous study drank over longer periods of time than
those in the present study, these data suggest none the
less that the alcohol-dosed participants in the field study
tended indeed to be more inebriated than the alcohol-
dosed participants in the present study. In turn, this
would suggest that there is a dose–time effect of acute
alcohol consumption on the ability to detect asymmetry
in faces.
Further support for this suggestion is provided within
the data set of the present study. The experimental proto-
col was slightly adjusted part-way through the period of
data collection because of concerns early on that the
ingested alcohol was not reaching its full effect, as partici-
pants were not being provided with enough time before
undertaking the tasks subsequent to finishing their drink
(approximately 20 minutes). This resulted in approxi-
mately 60% of the participants in the alcohol-dosed
group being provided with an extended period of ‘diges-
tion time’ (approximately 40 minutes). For alcohol-dosed
participants given a short digestion time, while their
average second breathalyser reading (25.6 ⫾SD 8.4 mg/
100 ml) was similar to that of the alcohol-dosed partici-
pants given a long digestion time (28.2 ⫾SD 5.7 mg/
100 ml), the increase in breathalyser readings from
the second to third measure averaged 2.5 ⫾SD 2.1 mg/
100 ml for the former group, while for the latter group
there was, on average, a slight decrease between the
second and third readings (0.4 ⫾SD 0.3 mg/100 ml).
Together, these data suggest that the latter set of partici-
pants had digested more of the alcohol consumed by the
time of the tasks. Re-analysis of the test data where the
alcohol-dosed condition included only data for partici-
pants given a longer digestion time did not change the
study findings either qualitatively or in terms of statisti-
cally significant patterns. However, the number of errors
made by participants given a longer digestion time was
significantly higher than for those participants given a
shorter digestion time, resulting in a difference in mean
errors between the placebo- and alcohol-dosed conditions
similar to, although still slightly lower than, that found in
the field [10] (Table 1).
Halsey et al. [10] reported an unexpected effect of sex
on the ability of participants to discern whether a face
was symmetrical or asymmetrical, with males making
fewer mistakes than females. In the present study, this
difference was not apparent; however, there was found to
be a greater preference for symmetrical faces by alcohol-
dosed females than by alcohol-dosed males. Although
these data were not reported by Halsey et al. [10], analy-
sis of the data from that study returns a P-value of 0.1.
These findings contradict those of Rhodes et al. [20], who
report that (non-alcohol-dosed) males find the appeal of
symmetrical opposite-sex faces stronger than do females.
6Lewis G. Halsey et al.
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction
An explanation for the present finding is not obvious and
warrants further investigation.
Future studies incorporating a range of alcohol-dose
levels should be valuable for clarifying how level of
alcohol dosing affects the ability to perceive asymmetry
and perhaps also the level of symmetr y preference. It may
also be that alcohol-dose level is found to interact with
gender effects on these measures. While currently it
seems reasonable to conclude that acute alcohol con-
sumption influences facial perception, the significance of
this relatively small effect in explaining the mechanisms
of increased perceptions of attractiveness during periods
of inebriation is unknown.
Declarations of interest
This study was funded by the AERC. The authors of this
paper have no connections with the tobacco, alcohol,
pharmaceutical or gaming industry or anyone funded
substantially by one of these organizations.
Acknowledgements
We thank Dr Jan Stygall and Dr Angela Attwood for
advice on the methods, and Dr Tony Little for providing
face images. This study was funded by an AERC Small
Grant.
References
1. Perrett D. I., Lee K. J., Penton-Voak I. S., Rowland D. R.,
Yoshikawa S., Burt D. M. et al. Effects of sexual dimorphism
on facial attractiveness. Nature 1998; 394: 884–7.
2. Conway C. A., Jones B. C., DeBruine L. M., Little A. C. Evi-
dence for adaptive design in human gaze preference. Proc
Biol Sci 2008; 275: 63–9.
3. Little A. C., Apicella C., Marlowe F. Preferences for symmetry
in human faces in two cultures: data from the UK and the
Hadza, an isolated group of hunter-gatherers. Proc Biol Sci
2007; 274: 3113–7.
4. Perrett D. I., Burt D. M., Penton-Voak I. S., Lee K. J., Rowland
D. A., Edwards R. Symmetry and human facial attractive-
ness. Evol Hum Behav 1999; 20: 295–307.
5. Young S., Sacco D., Hugenberg K. Vulnerability to disease is
associated with a domain-specific preference for symmetri-
cal faces relative to symmetrical non-face stimuli. Eur J Soc
Psychol 2011; 41: 558–68.
6. Jones B. T., Jones B. C., Thomas A. P., Piper J. Alcohol con-
sumption increases attractiveness ratings of opposite-sex
faces: a possible third route to risky sex. Addiction 2003; 98:
1069–75.
7. Parker L., Penton-Voak I. S., Attwood A., Munafo M. Effects
of acute alcohol consumption on ratings of attractivess of
facial stimuli: evidence of long-term encoding. Alcohol
Alcohol 2008; 43: 636–40.
8. Egan V., Cordan G. Barely legal: is attraction and estimated
age of young female faces disrupted by alcohol use, make
up, and the sex of the observer? Br J Psychol 2008; 13:
415–27.
9. Souto A., Bezerra B. M., Halsey L. G. Alcohol intoxication
reduces perception of asymmetry: an explanation for
increased perceptions of facial attractiveness after alcohol
consumption? Perception 2008; 37: 955–8.
10. Halsey L. G., Huber J., Bufton R., Little A. C. An explanation
for enhanced perceptions of attractiveness after alcohol
consumption. Alcohol 2010; 44: 307–13.
11. Andre J. Visual functioning in challenging conditions:
effects of alcohol consumption, luminance, stimulus
motion, and glare on contrast sensitivity. J Exp Psychol Appl
1996; 2: 250–69.
12. Colzato L. S., Erasmus V., Hommel B. Moderate alcohol con-
sumption in humans impairs feature binding in visual per-
ception but not across perception and action. Neurosci Lett
2004; 360: 103–5.
13. Watten R., Lie I. Visual functions and acute ingestion of
alcohol. Ophthalmic Physiol Opt 1996; 16: 460–6.
14. Sheridan C. Methods in Experimental Psychology. Kansas
City: Holt, Rinehart and Winston; 1979.
15. Oinonen K., Sterniczuk R. An inverse relationship between
typical alcohol consumption and facial symmetry detection
ability in young women. J Psychopharmacol 2007; 21: 507–
18.
16. Uttl B., Pilkenton-Taylor C. Letter cancellation performance
across the adult life span. Clin Neuropsychol 2001; 15: 521–
30.
17. Costa P., McCrae R. NEO—Revised Revised NEO Personality
Inventory (NEO-PI-R). Manual (UK edition). Adapted for the
UK by J. Rust and W Lord. Oxford: Hogrefe; 2006.
18. Saunders J. B., Aasland O. B., Babor T. F., De la Fuente J. R.,
Grant M. Development of the alcohol use disorders identi-
fication test (AUDIT): WHO collaborative project on early
detection of persons with harmful alcohol consumption—
II. Addiction 1992; 88: 791–803.
19. Oinonen K., Mazmanian D. Facial symmetry detection
ability changes across the menstrual cycle. Biol Psychol
2007; 75: 136–45.
20. Rhodes G., Proffitt F., Grady J., Sumich A. Facial symmetry
and the perception of beauty. Psychon Bull Rev 1998; 5:
659–69.
Alcohol consumption and asymmetry perception 7
© 2012 The Authors, Addiction © 2012 Society for the Study of Addiction Addiction