ArticlePDF Available

Abstract and Figures

Over the past 10 years, Oosterhof and Todorov’s valence–dominance model has emerged as the most prominent account of how people evaluate faces on social dimensions. In this model, two dimensions (valence and dominance) underpin social judgements of faces. Because this model has primarily been developed and tested in Western regions, it is unclear whether these findings apply to other regions. We addressed this question by replicating Oosterhof and Todorov’s methodology across 11 world regions, 41 countries and 11,570 participants. When we used Oosterhof and Todorov’s original analysis strategy, the valence–dominance model generalized across regions. When we used an alternative methodology to allow for correlated dimensions, we observed much less generalization. Collectively, these results suggest that, while the valence–dominance model generalizes very well across regions when dimensions are forced to be orthogonal, regional differences are revealed when we use different extraction methods and correlate and rotate the dimension reduction solution.
Content may be subject to copyright.
RegisTeRed RepoRT
https://doi.org/10.1038/s41562-020-01007-2
People quickly and involuntarily form impressions of others
based on their facial appearance13. These impressions then
influence important social outcomes4,5. For example, people
are more likely to cooperate in socioeconomic interactions with
individuals whose faces are evaluated as more trustworthy6, vote for
individuals whose faces are evaluated as more competent7, and seek
romantic relationships with individuals whose faces are evaluated as
more attractive8. Facial appearance can even influence life-or-death
outcomes. For example, untrustworthy-looking defendants are
more likely to receive death sentences9. Given that such evaluations
influence profound outcomes, understanding how people evaluate
others’ faces can provide insight into a potentially important route
through which social stereotypes impact behaviour10,11.
Over the past decade, Oosterhof and Todorov’s valence–domi-
nance model12 has emerged as the most prominent account of how
we evaluate faces on social dimensions5. Oosterhof and Todorov
identified 13 different traits (aggressiveness, attractiveness, car-
ingness, confidence, dominance, emotional stability, unhappiness,
intelligence, meanness, responsibility, sociability, trustworthiness
and weirdness) that perceivers spontaneously use to evaluate faces
when forming trait impressions12. From these traits, they derived
a two-dimensional model of perception: valence and dominance.
Valence, best characterized by rated trustworthiness, was defined
as the extent to which the target was perceived as having the inten-
tion to harm the viewer12. Dominance, best characterized by rated
dominance, was defined as the extent to which the target was per-
ceived as having the ability to inflict harm on the viewer12. Crucially,
the model proposes that these two dimensions are sufficient to
drive social evaluations of faces. As a consequence, the majority of
research on the effects of social evaluations of faces has focused on
one or both of these dimensions4,5.
Successful replications of the valence–dominance model have
only been conducted in Western samples13,14. This focus on the West
is consistent with research on human behaviour more broadly, which
typically draws general assumptions from analyses of Western par-
ticipants’ responses15. Kline etal.16 recently termed this problematic
practice the Western centrality assumption and argued that regional
variation, rather than universality, is probably the default for
human behaviour.
Consistent with Kline etal.’s notion that human behaviour is
best characterized by regional variation, two recent studies of social
evaluation of faces by Chinese participants indicate that different
factors underlie their impressions17,18. Both studies reported that
Chinese participants’ social evaluations of faces were underpinned
by a valence dimension similar to that reported by Oosterhof and
Todorov for Western participants, but not by a corresponding
dominance dimension. Instead, both studies reported a second
dimension, referred to as capability, which was best characterized
by rated intelligence. Furthermore, the ethnicity of the faces rated
only subtly affected perceptions17. Research into potential cultural
differences in the effects of experimentally manipulated facial char-
acteristics on social perceptions has also found little evidence that
cultural differences in social perceptions of faces depend on the eth-
nicity of the faces presented1921. Collectively, these results suggest
that the Western centrality assumption may be an important barrier
to understanding how people evaluate faces on social dimensions.
Crucially, these studies also suggest that the valence–dominance
model is not necessarily a universal account of social evaluations
of faces and warrants further investigation in the broadest set of
samples possible.
Although the studies described above demonstrate that the
valence–dominance model is not perfectly universal, to which spe-
cific world regions it does and does not apply are open and impor-
tant questions. Demonstrating differences between British and
Chinese raters is evidence against the universality of the valence–
dominance model, but it does not adequately address these ques-
tions. Social perception in China may be unique in not fitting the
valence–dominance model because of the atypically high general
importance placed on status-related traits, such as capability, during
social interactions in China22,23. Indeed, Tan etal.24 demonstrated
face-processing differences between Chinese participants living in
mainland China and Chinese participants living in nearby coun-
tries, such as Malaysia. Insights regarding the unique formation of
social perceptions in other cultures and world regions are lacking.
To which world regions does the valence–
dominance model of social perception apply?
Over the past 10 years, Oosterhof and Todorov’s valence–dominance model has emerged as the most prominent account of
how people evaluate faces on social dimensions. In this model, two dimensions (valence and dominance) underpin social
judgements of faces. Because this model has primarily been developed and tested in Western regions, it is unclear whether
these findings apply to other regions. We addressed this question by replicating Oosterhof and Todorov’s methodology across
11 world regions, 41 countries and 11,570 participants. When we used Oosterhof and Todorov’s original analysis strategy,
the valence–dominance model generalized across regions. When we used an alternative methodology to allow for correlated
dimensions, we observed much less generalization. Collectively, these results suggest that, while the valence–dominance
model generalizes very well across regions when dimensions are forced to be orthogonal, regional differences are revealed
when we use different extraction methods and correlate and rotate the dimension reduction solution.
Protocol registration
The stage 1 protocol for this Registered Report was accepted in principle on 5 November 2018. The protocol, as accepted by the
journal, can be found at https://doi.org/10.6084/m9.figshare.7611443.v1.
A full list of author affiliations appears at the end of the paper.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 159
RegisTeRed RepoRT Nature HumaN BeHaviour
Only a large-scale study investigating social perceptions in many
different world regions can provide such insights.
To establish the world regions to which the valence–dominance
model applies, we replicated Oosterhof and Todorov’s methodol-
ogy12 in a wide range of world regions (Africa, Asia, Australia and
New Zealand, Central America and Mexico, Eastern Europe, the
Middle East, the United States and Canada, Scandinavia, South
America, the United Kingdom and Western Europe; see Table 1).
Our study is the most comprehensive test of social evaluations of
faces to date, including more than 11,000 participants. Participating
research groups were recruited via the Psychological Science
Accelerator project2527. Previous studies compared two cultures to
demonstrate regional differences17,18. In contrast, the scale and scope
of our study allows us to generate the most comprehensive picture
of the world regions to which the valence–dominance model does
and does not apply.
We tested two specific competing predictions: (1) the valence–
dominance model applies to all world regions; and (2) the valence–
dominance model applies in Western-world regions, but not other
world regions.
Results
Analysed dataset. Following the planned data exclusions (see the
Supplementary Information for a breakdown of these exclusions;
code 1.5), the analysed dataset is summarized in Table 2.
Main analysis (principal component analysis (PCA); code 2.1).
Oosterhof and Todorov reported the results of a PCA with orthogo-
nal components, no rotation and retaining components with eigen-
values of >1. We conducted an identical analysis and report: (1)
the number of components extracted per the registered criteria; (2)
whether the first and second components had the same primary pat-
tern as Oosterhof and Todorov reported; and (3) the similarity of the
first and second factors as quantified with a congruence coefficient.
We extrac ted the same number of components (two) as Oosterhof
and Todorov in two world regions (Africa and South America)
and a different number of components (three) in the other world
regions (see Fig. 1). In the world regions where a third compo-
nent was extracted, the trait ratings of unhappy and weird tended
to have the highest loadings on that component, but those ratings
also crossloaded on the first component. We hesitate to interpret or
describe this component with any authority because it varied across
world regions, consisted of crossloaded traits and explained only a
small proportion of additional variance.
The primary pattern reported by Oosterhof and Todorov (a first
component that correlated strongly with rated trustworthiness but
not with rated dominance and a second component that correlated
strongly with rated dominance but not with rated trustworthiness)
was present in all world regions except Eastern Europe. In Eastern
Europe, dominance was correlated with the first component more
strongly than our registered criterion (i.e., that dominance would
correlate weakly with the first component; r < 0.5). Figure 1 shows
the full loading matrices for each region and Table 3 shows how
these relate to our registered criteria.
We report Tucker’s coefficient of congruence, ϕ, which quantifies
the loading similarity of Oosterhof and Todorov’s reported compo-
nent to the corresponding component we extracted. However, it is
important to interpret ϕ with caution when the numbers of compo-
nents differ across the solutions being compared. When comparing
loadings across solutions, an assumption is that the configuration of
the traits to components is the same (that is, configural invariance).
To the extent that the structures of the loading matrices differ across
solutions, the comparability of the loadings is compromised (that is,
loadings estimated from different dimensional spaces are not on the
same scale). For world regions that did not have the same configura-
tion of traits to components (that is, those with a different number
of components extracted or a different primary pattern observed), ϕ
was uninterpretable. This is because the differences in configuration
across the two solutions were conflated with the loading differences.
Our analyses indicated that the first component was equal to
the first component in Oosterhof and Todorov’s original study for
all world regions (ϕ > 0.95). The second component was equal to
(ϕ > 0.95) or fairly similar to (ϕ > 0.85) the second component
reported by Oosterhof and Todorov in all of the world regions
except Asia (ϕ = 0.848). Table 4 summarizes these results.
Together, these results suggest that the valence–dominance
model generalizes across world regions when using an identical
analysis to that used in Oosterhof and Todorov’s original study.
Thus, the results of our PCA support prediction 1 (that the valence–
dominance model will apply to all world regions) but not prediction
2 (that the valence–dominance model will apply in Western-world
regions but not other world regions). However, we note here that in
most world regions we extracted a third component not extracted
in the original study: that Eastern Europe did not demonstrate the
same primary pattern and that ϕ should be interpreted with caution
for all world regions except Africa and South America.
Robustness analyses (exploratory factor analysis (EFA); code
2.2). Following our analysis plan, we conducted additional robust-
ness analyses that directly addressed criticisms of the type of sta-
tistical analyses used by Oosterhof and Todorov (see ref. 28 for a
discussion of these criticisms). These robustness analyses employed
EFA with an oblimin rotation as the model and used parallel analy-
sis to identify the number of factors to extract. The goal of an EFA
with an oblimin rotation is to simplify the loading matrix and yield
interpretable factors.
We conducted this analysis on Oosterhof and Todorov’s original
data and found a similar result to their PCA solution: two factors
extracted, with factor 1 characterized by a high loading for trustwor-
thiness and factor 2 characterized by a high loading for dominance.
Table 1 | World regions, countries and localities of data
collection
World region Countries and localities
Africa Kenya, (Nigeria) and South Africa
Asia China, India, Malaysia, Taiwan and Thailand
Australia and New
Zealand Australia and New Zealand
Central America and
Mexico El Salvador and Mexico
Eastern Europe Hungary, Lithuania, Poland, Russia, Serbia and
Slovakia
The Middle East Iran, Israel and Turkey
United States and
Canada Canada and the United States
Scandinavia Denmark, (Finland), Norway and (Sweden)
South America Argentina, Brazil, Chile, Colombia and
Ecuador
United Kingdom England, Scotland and Wales
Western Europe Austria, Belgium, France, Germany, (Greece),
Italy, the Netherlands, Portugal, Spain and
Switzerland
We collected data from a minimum of 350 raters per world region based on the simulations
described in the Methods. Countries in parentheses were added to the list after acceptance in
principle of the stage 1 protocol. Ecuador was incorrectly classified as Central America and Mexico
in our stage 1 submission, but has been classified as South America for analyses and in our stage
2 submission.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav
160
RegisTeRed RepoRT
Nature HumaN BeHaviour
However, for all other world regions, we extracted more than two
factors using parallel analysis. Full EFA loading matrices for each
region and Oosterhof and Todorov’s original data are shown in
Fig. 2. The four-factor solution for the USA and Canada did not
converge. We did not register a contingency for nonconvergence,
but because parallel analysis can lead to over extraction, we reran
the EFA with one fewer than the number of suggested factors. The
model converged when estimating three factors.
In contrast with the PCA, the results of our robustness analyses
showed less evidence that the valence–dominance model generalizes
across world regions. For example, we extracted a different number
of factors than the original solution for all world regions. A summary
of the results for our replication criteria is given in Table 5.
Because the number of factors differed from the original solu-
tion in all world regions and the loading matrices were differen-
tially rotated from the original solution, it is not valid to compare
the differences in the loadings from the original solution with those
observed in the world regions reported here, as we had initially
planned. Loadings quantify the relationship of traits to a factor. To
compare loadings across samples, we must first determine whether
we extracted the same factor in each sample (that is, satisfied the
assumption of configural invariance). Our registered analyses
included the calculation of Tucker’s coefficient of congruence, ϕ in
order to compare the first factor from the original study with the
first factor we extracted in a given world region, and to compare
the second factor from the original study with the second factor
extracted in a given world region. However, because we extracted
a different number of factors from the original solution in all world
regions, it is not valid to compare the loadings across these different
factors, or to quantify their differences using ϕ.
The congruence coefficient is only appropriate to report when
we can ensure that the factors are comparable across samples. That
the number of factors extracted did not replicate the original pattern
and that the EFAs were rotated differently across world regions
negates the comparability of the loadings. Consistent with our
registered analysis code, we reported ϕ for the relationship of the
first factor from Oosterhof and Todorov to the factor with the most
explained variance in a world region, and ϕ for the relationship of
the second factor from Oosterhof and Todorov to the factor with
the second most explained variance in a world region only in the
Supplementary Information. However, we stress that these coeffi-
cients are quantifying loadings that link to different factors from
different dimensional spaces and are not necessarily comparable.
In summary, the results of our EFA support neither prediction 1
(that the valence–dominance model will apply to all world regions)
nor prediction 2 (that the valence–dominance model will apply to
Western-world regions but not other world regions).
Discussion
Our primary analyses—PCAs identical to those reported by
Oosterhof and Todorov—suggested that the valence–dominance
model of social perception of faces generalizes well across world
regions. Although most world regions showed a third component
not discussed in the original work12, this third component is actually
similar to the third component in Oosterhof and Todorov’s origi-
nal study. In Oosterhof and Todorov’s original study, they did not
interpret the third component because its eigenvalue was below 1,
whereas in our analyses the eigenvalues of the third components
in most of the regions were just above 1. Nonetheless, the third
component in each region had a factor congruence between 0.77
and 0.90 with the third component for Oosterhof and Todorov’s
data. However, we emphasize here that many of these dimensions
accounted for a relatively small proportion of the variance explained
and, thus, may be of limited theoretical importance.
In contrast with the results of our PCAs, an alternative analysis
that addressed common criticisms of the type of analysis Oosterhof
and Todorov employed showed much less generalization across
world regions. We used modern extraction techniques and EFAs
with correlated factor rotations. The correlated rotation meth-
ods aim to simplify the loading matrix with the goal of estimating
Table 2 | Number of participants per region and Cronbach’s α values following data quality checks and exclusions
Region Aggressive Attractive Caring Confident Dominant Emotionally
stable
Intelligent Mean Responsible Sociable Trustworthy Unhappy Weird
Western Europe α=0.978
n=152
α=0.991
n=147
α=0.976
n=136
α=0.985
n=156
α=0.973
n=150
α=0.981
n=141
α=0.975
n=141
α=0.969
n=120
α=0.978
n=138
α=0.988
n=188
α=0.978
n=141
α=0.983
n=140
α=0.982
n=113
United States and
Canada
α=0.983
n=248
α=0.991
n=224
α=0.986
n=257
α=0.989
n=303
α=0.977
n=246
α=0.986
n=270
α=0.979
n=239
α=0.984
n=270
α=0.984
n=269
α=0.988
n=246
α=0.984
n=263
α=0.985
n=252
α=0.987
n=226
United Kingdom α=0.879
n=16
α=0.949
n=22
α=0.936
n=34
α=0.93
n=30
α=0.886
n=34
α=0.9
n=30
α=0.911
n=34
α=0.87
n=27
α=0.892
n=37
α=0.932
n=28
α=0.92
n=27
α=0.937
n=24
α=0.899
n=18
South America α=0.948
n=97
α=0.982
n=108
α=0.944
n=112
α=0.968
n=108
α=0.957
n=121
α=0.949
n=100
α=0.938
n=110
α=0.949
n=95
α=0.937
n=117
α=0.974
n=110
α=0.952
n=107
α=0.961
n=87
α=0.973
n=116
Scandinavia α=0.95
n=48
α=0.969
n=44
α=0.949
n=46
α=0.96
n=56
α=0.941
n=49
α=0.955
n=67
α=0.958
n=54
α=0.912
n=36
α=0.915
n=37
α=0.969
n=64
α=0.949
n=58
α=0.952
n=55
α=0.952
n=39
Middle East α=0.912
n=32
α=0.949
n=32
α=0.934
n=42
α=0.943
n=39
α=0.9
n=35
α=0.903
n=33
α=0.896
n=48
α=0.901
n=36
α=0.87
n=34
α=0.944
n=41
α=0.895
n=42
α=0.943
n=57
α=0.896
n=32
Eastern Europe α=0.941
n=59
α=0.971
n=58
α=0.926
n=56
α=0.946
n=60
α=0.952
n=74
α=0.923
n=56
α=0.939
n=64
α=0.937
n=68
α=0.953
n=65
α=0.955
n=68
α=0.937
n=54
α=0.964
n=74
α=0.956
n=53
Central America
and Mexico
α=0.845
n=26
α=0.93
n=25
α=0.788
n=24
α=0.89
n=32
α=0.859
n=33
α=0.835
n=23
α=0.832
n=33
α=0.817
n=23
α=0.824
n=22
α=0.882
n=28
α=0.851
n=27
α=0.771
n=27
α=0.842
n=15
Australia and
New Zealand
α=0.956
n=77
α=0.98
n=88
α=0.964
n=90
α=0.972
n=93
α=0.936
n=66
α=0.957
n=88
α=0.951
n=81
α=0.947
n=71
α=0.937
n=68
α=0.972
n=95
α=0.953
n=72
α=0.948
n=85
α=0.962
n=70
Asia α=0.932
n=59
α=0.957
n=52
α=0.948
n=73
α=0.959
n=72
α=0.917
n=55
α=0.908
n=55
α=0.927
n=64
α=0.909
n=51
α=0.931
n=63
α=0.952
n=65
α=0.93
n=61
α=0.937
n=61
α=0.942
n=49
Africa α=0.808
n=45
α=0.873
n=38
α=0.865
n=44
α=0.805
n=31
α=0.79
n=38
α=0.779
n=38
α=0.756
n=37
α=0.889
n=51
α=0.811
n=36
α=0.819
n=34
α=0.867
n=49
α=0.795
n=43
α=0.889
n=37
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 161
RegisTeRed RepoRT Nature HumaN BeHaviour
interpretable factors, and in our data revealed more regional varia-
tion. These results suggest that, if the dimensions of face percep-
tion are indeed correlated, using analytical techniques that force
these dimensions to be uncorrelated may be obscuring important
regional differences in the structure of face perceptions.
A necessary next step for moving forward in person percep-
tion research is to address which analysis model (PCA or EFA)
best aligns with theory, so that models and theories can be revised
and expanded appropriately in future research. Crucially, the two
models make different assumptions about trait ratings of faces.
Table 3 | Replication criteria for the PCA for each region
Region Component 1 Component 2 Replicated
Trustworthy Dominant Dominant Trustworthy
Oosterhof and Todorov12 0.941 0.244 0.929 0.060 Yes
Africa 0.924 0.271 0.843 0.065 Yes
Asia 0.922 0.370 0.863 0.006 Ye s
Australia and New Zealand 0.943 0.257 0.907 0.076 Yes
Central America and Mexico 0.918 0.007 0.915 0.050 Yes
Eastern Europe 0.938 0.599 0.755 0.113 No
Middle East 0.831 0.490 0.810 0.382 Ye s
Scandinavia 0.953 0.392 0.881 0.121 Yes
South America 0.898 0.309 0.905 0.151 Ye s
United Kingdom 0.944 0.331 0.851 0.121 Yes
United States and Canada 0.966 0.406 0.841 0.073 Yes
Western Europe 0.957 0.357 0.875 0.166 Yes
Oosterhof and Todorov’s valence–dominance model was judged to have been replicated in a given world region if the first component had a loading of >0.7 with trustworthiness and <0.5 with dominance,
and if the second component had a loading of >0.7 with dominance and <0.5 with trustworthiness.
–0.71
0.81
0.90
0.67
–0.24
0.93
0.72
–0.76
0.91
0.91
0.94
–0.71
–0.87
0.66
0.32
–0.29
0.65
0.93
0.19
0.13
0.55
0.11
0.20
–0.06
0.01
–0.22
0.63 0.18
–0.62
0.67
0.78
0.74
0.01
0.92
0.76
–0.75
0.85
0.84
0.92
–0.74
–0.71
0.69
0.46
–0.41
0.45
0.92
0.06
0.27
0.48
0.36
–0.14
–0.05
0.07
–0.15
0.16
–0.30
0.07
0.36
0.07
0.10
–0.28
0.14
–0.03
0.33
–0.14
–0.58
0.50
0.56 0.18 0.08
–0.60
0.71
0.76
0.79
0.31
0.95
0.77
–0.73
0.82
0.90
0.90
–0.79
–0.77
0.76
0.48
–0.55
0.50
0.91
0.02
0.24
0.64
0.18
–0.05
–0.15
0.01
–0.22
0.09
–0.21
0.12
0.31
0.08
0.09
–0.32
0.09
–0.21
0.37
–0.24
–0.57
0.42
0.59 0.21 0.08
–0.58
0.78
0.81
0.69
0.27
0.82
0.75
–0.70
0.78
0.89
0.92
–0.75
–0.82
0.75
0.34
–0.41
0.53
0.84
0.13
0.24
0.63
0.34
–0.05
–0.07
0.25
–0.10
0.57 0.19
–0.63
0.88
0.87
0.77
0.60
0.90
0.87
–0.64
0.92
0.90
0.94
–0.60
–0.81
0.74
0.23
–0.28
0.53
0.76
0.05
0.08
0.72
0.19
–0.11
–0.11
0.11
–0.12
–0.02
0.24
0.11
–0.30
–0.01
–0.28
0.21
–0.02
0.14
–0.25
0.20
0.76
–0.41
0.65 0.16 0.09
–0.62
0.73
0.87
0.78
0.33
0.89
0.80
–0.67
0.87
0.92
0.94
–0.65
–0.76
0.71
0.48
–0.31
0.41
0.85
0.06
0.15
0.63
0.24
–0.11
–0.12
0.24
–0.23
0.18
–0.31
0.12
0.39
0.23
0.18
–0.32
0.12
–0.23
0.25
–0.10
–0.66
0.46
0.60 0.18 0.10
–0.65
0.76
0.91
0.81
0.37
0.88
0.83
–0.80
0.91
0.93
0.92
–0.73
–0.76
0.63
0.51
–0.30
0.26
0.86
–0.14
0.42
0.50
0.25
–0.24
–0.01
0.36
–0.27
0.33
–0.24
0.09
0.47
0.22
0.22
–0.08
0.15
–0.05
0.16
–0.27
–0.53
0.52
0.64 0.18 0.09
–0.55
0.66
0.83
0.81
0.49
0.94
0.72
–0.44
0.77
0.87
0.83
–0.72
–0.70
0.77
0.48
–0.41
0.45
0.81
–0.06
0.52
0.84
0.39
–0.28
–0.38
0.10
–0.10
0.13
–0.38
0.06
0.26
0.11
0.04
–0.19
0.09
0.11
0.28
–0.21
–0.58
0.61
0.54 0.25 0.09
–0.70
0.72
0.89
0.81
0.41
0.95
0.83
–0.70
0.92
0.95
0.97
–0.73
–0.79
0.67
0.48
–0.30
0.42
0.84
0.01
0.17
0.68
0.13
–0.04
–0.07
0.19
–0.20
0.18
–0.27
0.14
0.37
0.16
0.16
–0.27
0.05
–0.23
0.21
–0.17
–0.62
0.45
0.66 0.17 0.08
–0.74
0.75
0.90
0.79
0.26
0.92
0.79
–0.67
0.91
0.93
0.94
–0.68
–0.77
0.62
0.44
–0.25
0.45
0.91
0.10
0.14
0.65
0.16
–0.10
–0.08
0.13
–0.13
0.13
–0.30
0.17
0.36
0.11
0.22
–0.27
0.07
–0.21
0.24
–0.19
–0.69
0.52
0.63 0.17 0.10
–0.63
0.80
0.88
0.77
0.39
0.95
0.79
–0.69
0.91
0.94
0.95
–0.74
–0.75
0.72
0.35
–0.30
0.50
0.88
0.04
0.25
0.68
0.12
–0.11
–0.12
0.07
–0.15
0.03
–0.31
0.08
0.34
0.11
0.13
–0.26
0.06
–0.12
0.20
–0.14
–0.62
0.52
0.64 0.18 0.08
–0.69
0.82
0.86
0.83
0.36
0.95
0.81
–0.64
0.90
0.93
0.96
–0.74
–0.78
0.70
0.36
–0.36
0.45
0.88
0.08
0.24
0.73
0.30
–0.19
–0.17
0.06
–0.15
0.01
0.31
0.04
–0.30
–0.15
–0.17
0.24
–0.05
0.11
–0.22
0.14
0.61
–0.50
0.65 0.19 0.08
South America United Kingdom United States and Canada Western Europe
Central America and Mexico Eastern Europe Middle East Scandinavia
Oosterhof and Todorov (2008)
Africa Asia Australia and New Zealand
Component 1 Component 2 Component 3 Component 1 Component 2 Component 3 Component 1 Component 2 Component 3 Component 1 Component 2 Component 3
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Fig. 1 | PCA loading matrices for each region. Positive loadings are shaded red and negative loadings are shaded blue. Darker colours correspond to
stronger loadings. The proportion of variance (Prop.Var) explained by each component is included at the top of each table.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav
162
RegisTeRed RepoRT
Nature HumaN BeHaviour
The PCA model does not assume that a latent factor causes the
trait ratings of the faces. The component captures linear combi-
nations of the original variables, maximized to explain variance.
Furthermore, in the original valence–dominance model, those
components were assumed to be orthogonal. In contrast, the the-
ory underlying the EFA model is that a latent factor causes the trait
ratings and any unexplained variance in that rating is measure-
ment error. Additionally, our EFA models allowed for the factors to
be correlated.
Theory can guide which model we use to analyse person per-
ception data. A person perception theory that aligns with a PCA
model would state that there are no underlying latent factors that
cause a person to rate a face in a particular way. There are, instead,
perceptual processes that vary across contexts, those doing the rat-
ing and those being rated, and the differential processes give rise
to components that can be used to reduce the data. This theory of
person perception would move forward with identifying the shared
processes across contexts, raters and ratees to see whether there are
predictable patterns in how the data are reduced.
A person perception theory that aligns with an EFA model makes
different assumptions about the processes that give rise to face rat-
ings. This theory would state that latent factors (for example, valence
or dominance) cause the trait ratings and, once we account for the
correct latent factors, any variability left in the ratings is measure-
ment error. We suggest that more careful and explicit consideration
of how theory connects to these approaches, and of which approach
is best suited to different research questions, will benefit the field.
Our study is one of several recent studies that have begun to
utilize different statistical models and to explore more dynamic
theories of trait ratings21,29,30 by exploring how the structures of
trait ratings vary systematically. This growing body of work cata-
logues variations in trait ratings by target demographic21,29,31, target
status32, target age33, perceiver knowledge34 and cultural factors17,18.
Furthermore, this growing body of work proposes dynamic theories
of person perception and more flexible statistical models for captur-
ing them21,29,30,35.
Table 4 | Factor congruence for each region’s PCA
Region Component 1 Component 2
Loading Congruence Loading Congruence
Africa 0.980 Equal 0.947 Fairly similar
Asia 0.974 Equal 0.843 Not similar
Australia and
New Zealand 0.982 Equal 0.959 Equal
Central America
and Mexico 0.992 Equal 0.935 Fairly similar
Eastern Europe 0.953 Equal 0.948 Fairly similar
Middle East 0.952 Equal 0.859 Fairly similar
Scandinavia 0.973 Equal 0.960 Equal
South America 0.976 Equal 0.953 Equal
United Kingdom 0.976 Equal 0.938 Fairly similar
United States and
Canada 0.972 Equal 0.952 Equal
Western Europe 0.975 Equal 0.936 Fairly similar
–0.32
0.86
0.68
0.92
0.23
0.94
0.67
–0.42
0.88
0.92
0.83
–0.61
–0.88
0.84
0.10
–0.50
0.47
0.97
–0.04
–0.06
0.72
–0.12
–0.02
–0.29
0.17
0.01
0.56 0.23
–0.05
0.04
0.43
0.75
0.11
0.51
0.04
–0.14
0.34
0.81
0.27
–0.93
0.11
0.90
0.19
–0.46
0.31
0.79
–0.23
–0.16
0.74
0.02
–0.14
–0.31
0.10
0.17
–0.07
0.83
0.33
0.15
0.24
0.19
0.40
–0.12
0.38
0.25
0.59
0.22
–0.76
–0.09
0.05
–0.22
0.24
0.23
0.37
0.55
–0.23
0.46
–0.25
0.06
–0.10
–0.12
0.26 0.210.23 0.11
0.91
0.17
–0.70
0.34
0.78
–0.21
–0.02
0.83
–0.07
–0.23
–0.46
0.08
0.12
–0.07
0.10
0.39
0.71
0.26
0.47
–0.02
–0.13
0.07
0.80
0.06
–0.96
0.10
–0.17
0.06
–0.05
0.21
0.28
0.41
0.80
–0.18
0.86
–0.13
0.25
–0.14
–0.27
0.03
0.82
0.19
0.23
0.28
0.16
0.12
–0.12
0.01
0.30
0.54
0.20
–0.72
0.25 0.24 0.20 0.19
–0.24
0.96
0.34
0.31
0.20
0.28
0.59
–0.17
0.76
0.68
0.79
0.07
–0.94
–0.42
–0.16
0.57
0.45
0.07
0.62
0.18
–0.65
0.05
0.28
0.21
–0.94
0.09
0.68
0.12
–0.28
0.63
0.80
0.27
0.17
0.48
0.21
–0.08
–0.13
–0.06
0.09
0.37 0.25 0.16
–0.21
0.95
0.64
0.57
0.75
0.50
0.83
–0.23
0.89
0.47
0.84
0.17
–0.92
0.85
0.02
–0.43
0.41
0.61
–0.09
–0.12
0.82
–0.02
–0.25
–0.32
0.12
0.09
–0.16
–0.03
0.12
0.54
0.15
0.54
0.03
–0.16
0.09
0.51
0.03
–1.01
0.17
0.49 0.19 0.19
–0.15
–0.01
0.64
0.67
0.25
0.47
–0.08
–0.23
0.06
0.73
0.34
–0.95
0.09
–0.08
0.82
0.48
0.13
0.27
0.03
0.31
–0.08
0.45
0.32
0.51
0.23
–0.81
0.85
0.22
–0.32
0.39
0.77
–0.09
–0.19
0.72
–0.06
–0.13
–0.30
0.07
0.06
–0.18
0.16
–0.18
0.34
0.21
0.58
0.68
–0.21
0.54
0.04
0.19
–0.06
–0.17
0.26 0.25 0.18 0.18
–0.31
0.90
0.28
0.18
0.49
0.26
0.77
–0.31
0.69
0.26
0.76
0.25
–1.01
–0.26
0.01
0.69
0.83
0.11
0.72
0.23
–0.46
0.36
0.75
0.24
–1.06
0.17
0.74
0.20
–0.29
0.41
0.79
–0.09
0.21
0.57
0.08
–0.20
–0.25
0.06
0.14
0.37 0.36 0.16
0.92
0.10
–0.63
0.21
0.60
–0.36
0.16
0.93
0.10
–0.47
–0.69
0.18
0.30
–0.06
0.07
0.45
0.80
0.50
0.54
0.27
–0.03
0.56
0.74
0.18
–0.89
0.18
–0.06
0.82
0.16
0.29
0.45
0.38
0.71
0.06
0.41
–0.01
0.44
0.21
–0.89
0.280.29 0.26
–0.32
0.86
0.31
0.35
0.49
0.47
0.80
–0.19
0.87
0.39
0.77
0.22
–0.97
–0.12
–0.00
0.54
0.76
0.26
0.57
0.07
–0.27
0.11
0.64
0.19
–1.05
0.15
0.84
0.22
–0.41
0.38
0.72
–0.12
–0.07
0.79
–0.11
–0.14
–0.30
0.13
0.08
0.41 0.28 0.19
–0.50
0.63
0.74
0.12
0.16
0.17
–0.09
–0.31
0.54
0.64
0.73
–0.00
–0.42
–0.11
–0.12
0.39
0.66
0.12
0.57
0.13
–0.19
0.06
0.50
0.06
–0.98
0.22
–0.21
0.31
–0.07
0.28
0.07
0.43
0.94
–0.25
0.44
–0.01
0.28
0.03
–0.64
0.63
0.39
–0.16
0.45
0.87
0.05
–0.06
0.64
0.11
–0.03
–0.08
0.07
–0.02
0.30 0.220.23 0.15
–0.22
0.91
0.42
0.32
0.43
0.50
0.78
–0.31
0.75
0.39
0.74
0.18
–0.98
–0.21
–0.01
0.43
0.72
0.29
0.54
0.07
–0.20
0.23
0.61
0.21
–1.04
0.22
0.80
0.15
–0.42
0.41
0.81
–0.10
0.05
0.79
–0.06
–0.23
–0.30
0.10
0.06
0.41 0.27 0.20
–0.17
0.55
–0.00
0.26
0.11
0.34
1.03
–0.13
0.58
0.03
0.40
0.02
–0.68
–0.29
–0.06
0.21
0.68
0.26
0.56
0.11
–0.18
0.22
0.57
0.16
–0.99
0.20
0.71
0.30
–0.21
0.42
0.87
0.03
–0.08
0.70
0.19
–0.10
–0.17
0.06
–0.03
–0.31
0.49
0.80
0.11
0.14
0.24
–0.18
–0.38
0.29
0.51
0.56
0.06
–0.42
0.27 0.24 0.170.24
South America United Kingdom United States and Canada Western Europe
Central America and Mexico Eastern Europe Middle East Scandinavia
Oosterhof and Todorov (2008) Africa Asia Australia and New Zealand
Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Factor 2 Factor 3 Factor 4 Factor 1 Factor 2 Factor 3 Factor 4
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Dominant
Aggressive
Mean
Unhappy
Weird
Confident
Intelligent
Attractive
Caring
Sociable
Responsible
Emotionally Stable
Trustworthy
Prop.Var
Fig. 2 | EFA loading matrices for each region. Positive loadings are shaded red and negative loadings are shaded blue. Darker colours correspond to
stronger loadings. The proportion of variance explained by each factor is included at the top of each table.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 163
RegisTeRed RepoRT Nature HumaN BeHaviour
Our results are consistent with this recent work in that they do
not provide strong evidence that there are a few generalizable latent
factors that cause the trait ratings across world regions. However,
they do suggest a dynamic process of person perception and eluci-
date the differential patterns of ratings across world regions. We can
use these data, representing impressions formed on a global scale,
to expand or refine our theories and to guide the selection of sta-
tistical models to represent those theories. Given the accumulating
evidence for variation in trait ratings, it is important that the con-
nection between the statistical models used to represent theories of
person perception are explicit and can accommodate the complexi-
ties of the impression formation process.
Methods
Ethics. Each research group had approval from their local ethics committee or
institutional review board to conduct the study, had explicitly indicated that their
institution did not require approval for the researchers to conduct this type of
face-rating task or had explicitly indicated that the current study was covered by
a pre-existing approval. Although the specics of the consent procedure diered
across research groups, all participants provided informed consent. All data were
stored centrally on University of Glasgow servers.
Procedure. Oosterhof and Todorov derived their valence–dominance model from
a PCA of ratings (by US raters) of 66 faces for 13 different traits (aggressiveness,
attractiveness, caringness, confidence, dominance, emotional stability, intelligence,
meanness, responsibility, sociability, trustworthiness, unhappiness and
weirdness)12. Using the criteria of the number of components with eigenvalues
greater than 1.0, this analysis produced two principal components. The first
component explained 63% of the variance in trait ratings, strongly correlated
with rated trustworthiness (r = 0.94) and weakly correlated with rated dominance
(r = 0.24). The second component explained 18% of the variance in trait ratings,
strongly correlated with rated dominance (r = 0.93) and weakly correlated with
rated trustworthiness (r = 0.06). We replicated Oosterhof and Todorov’s method12
and primary analysis in each world region we examined.
Stimuli in our study came from an open-access, full-colour face image set36
consisting of images of the faces of 60 men and 60 women taken under standardized
photographic conditions (Mage = 26.4 years; s.d. = 3.6 years; range = 18–35 years).
These 120 images consisted of 30 Black (15 male; 15 female), 30 White (15 male; 15
female), 30 Asian (15 male; 15 female) and 30 Latin faces (15 male; 15 female). As
reported by Oosterhof and Todorov’s study12, the individuals photographed posed
looking directly at the camera with a neutral expression, and the background,
lighting and clothing (here, a grey t-shirt) were constant across images.
In our study, adult raters were randomly assigned to rate the 13 adjectives
tested by Oosterhof and Todorov using scales ranging from 1 (not at all) to 9
(very) for all 120 faces in a fully randomized order at their own pace. Because all
researchers collected data through an identical interface (except for differences in
instruction language), data collection protocols were highly standardized across
laboratories. Each participant completed the block of 120 face-rating trials twice
so that we could report test–retest reliabilities of ratings; ratings from the first and
second blocks were averaged for all analyses (see code 1.5.5 in the Supplementary
Information).
Raters also completed a short questionnaire requesting demographic
information (sex, age and ethnicity). These variables were not considered in
Oosterhof and Todorov’s analyses but were collected in our study so that other
researchers could use them in secondary analyses of the published data. The data
from this study comprise the largest and most comprehensive open-access set of
face ratings with open stimuli from around the world, providing an invaluable
resource for further research addressing the Western centrality assumption in
person perception research.
Raters completed the task in a language appropriate for their country (see
below). To mitigate potential problems with translating single-word labels,
dictionary definitions for each of the 13 traits were provided. Twelve of these
dictionary definitions had previously been used to test for effects of social
impressions on the memorability of face photographs37. Dominance (not included
in that study) was defined as strong and important.
Participants. Simulations determined that we should obtain at least 25 different
raters for each of the 13 traits in every region (see https://osf.io/x7fus/ for
code and data). We focused on ratings of attractiveness and intelligence for the
simulations because they showed the highest and lowest agreement among the
traits analysed by Oosterhof and Todorov, respectively. First, we sampled from a
population of 2,513 raters, each of whom had rated the attractiveness of 102 faces;
these simulations showed that more than 99% of 1,000 random samples of 25
raters produced good or excellent inter-rater reliability coefficients (Cronbach’s α
values > 0.80). We then repeated these simulations, sampling from a population
of 37 raters, each of whom rated the intelligence of 100 faces, showing that 93% of
1,000 random samples of 25 raters produced good or excellent inter-rater reliability
coefficients (Cronbach’s α values > 0.80). Thus, averages of ratings from 25 or more
raters will have produced reliable dependent variables in our analyses; we planned
to test at least 9,000 raters in total.
In addition to rating the faces for the 13 traits examined by Oosterhof and
Todorov, 25 participants in each region were randomly assigned to rate the
targets’ age in light of Sutherland etal.’s results showing that a youth/attractiveness
dimension emerged from analyses of a sample of faces with a very diverse age
range38. Age ratings were not included in analyses relating to replications of
Oosterhof and Todorov’s valence–dominance model. These age ratings were
collected to allow for planned exploratory analyses including rated age, but we did
not perform these analyses.
Analysis plan. The code used for our analyses is included in the Supplementary
Information and publicly available from the Open Science Framework (https://osf.
io/87rbg/). The specific sections of code are cited below.
Ratings from each world region were analysed separately and anonymous raw
data have been published on the Open Science Framework. Our main analyses
directly replicated the PCA reported by Oosterhof and Todorov to test their
theoretical model in each region sampled (code 2.1). First, we calculated the average
rating for each face separately for each of the 13 traits (code 2.1.2). We then subjected
Table 5 | Replication criteria for the EFA for each region
Region Factor 1 Factor 2 Replicated
Trustworthy Dominant Dominant Trustworthy
Oosterhof and Todorov12 0.826 0.228 0.970 0.288 Ye s
Africa 0.786 0.200 0.069 0.214 No
Asia 0.761 0.487 0.110 0.236 No
Australia and New Zealand 0.730 0.157 0.071 0.281 No
Central America and Mexico 0.268 0.108 0.241 0.591 No
Eastern Europe 0.843 0.750 0.609 0.322 No
Middle East 0.177 0.502 0.600 0.686 No
Scandinavia 0.744 0.428 0.293 0.211 No
South America 0.458 0.778 0.261 0.058 No
United Kingdom 0.338 0.249 0.265 0.510 No
United States and Canada 0.768 0.491 0.264 0.189 No
Western Europe 0.398 0.111 0.256 0.164 No
Oosterhof and Todorov’s valence–dominance model was judged to have been replicated in a given world region if the first factor had a loading >0.7 with trustworthiness and <0.5 with dominance and the
second factor had a loading >0.7 with dominance and <0.5 with trustworthiness.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav
164
RegisTeRed RepoRT
Nature HumaN BeHaviour
these mean ratings to PCA with orthogonal components and no rotation, as
Oosterhof and Todorov did (code 2.1.3). Using the criteria they reported, we retained
and interpreted components with eigenvalues greater than 1.0 (code 2.1.3.1).
Criteria for replicating Oosterhof and Todorov’s valence–dominance model. We
used multiple sources of evidence to judge whether Oosterhof and Todorov’s
valence–dominance model replicated in a given world region. First, we examined
the solution from the PCA conducted in each region and determined whether
Oosterhof and Todorov’s primary pattern replicated according to three criteria: (1)
the first two components had eigenvalues greater than 1.0; (2) the first component
(that is, the one explaining more of the variance in ratings) correlated strongly with
trustworthiness (r > 0.7) and weakly with dominance (r < 0.5); and (3) the second
component (that is, the one explaining less of the variance in ratings) correlated
strongly with dominance (r > 0.7) and weakly with trustworthiness (r < 0.5). If the
solution in a world region met all three of these criteria, we concluded that the
primary pattern of the model replicated in that region (code 2.1.3.3).
In addition to reporting whether the primary pattern was replicated in each
region, we also reported Tucker’s coefficient of congruence39,40. The congruence
coefficient, ϕ, ranges from 1 to 1 and quantifies the similarity between two
vectors of loadings41. It is:
ϕx;yð Þ¼
Px
i
y
i
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Px
2
iPy
2
i
p
where xi and yi are the loadings of variable i (i = 1, , n number of indicators
in the analysis) onto factors x and y, respectively. For the purposes of the current
research, we compared the vector of loadings from the first component from
Oosterhof and Todorov with the vector of loadings from the first component
estimated from each world region. We repeated this analysis for the second
component. This produced a standardized measure of component similarity for
each component in each world region that was not sensitive to the mean size of the
loadings42. Furthermore, this coefficient was fitting for the current study because
it does not require an apriori specification of a factor structure for each group, as
would be needed if we were to compare the factor structures in a multiple-group
confirmatory factor analysis. Following previous guidelines42, we concluded that
the components reported by Oosterhof and Todorov were not similar to those
estimated in a given world region if the coefficient was <0.85, were fairly similar if
it was between 0.85 and 0.94 and were equal if it was >0.95 (code 2.1.4).
Thus, we reported whether the solution had the same primary pattern that
Oosterhof and Todorov found and quantified the degree of similarity between
each component and the corresponding component from Oosterhof and Todorov’s
work. This connects to our competing predictions.
Prediction 1 (the valence–dominance model applies to all world regions) was
supported if the solution from the PCA conducted in each region satisfied all of the
criteria described above. Specifically, the primary pattern was replicated and the
components had at least a fair degree of similarity as quantified by a value of ϕ of
0.85 or greater.
Prediction 2 (the valence–dominance model applies in Western-world
regions but not other world regions) was supported if the solutions from the
PCA conducted in Australia and New Zealand, the United States and Canada,
Scandinavia, the United Kingdom and Western Europe, but not Africa, Asia,
Central America and Mexico, Eastern Europe, the Middle East or South America,
satisfied the criteria described above.
Exclusions. Data from raters who failed to complete all 120 ratings in the first
block of trials or who provided the same rating for 75% or more of the faces were
excluded from the analyses (codes 1.5.1, 1.5.3 and 1.5.5).
Data quality checks. Following previous research testing the valence–dominance
model1214, data quality was checked by separately calculating the inter-rater
agreement (indicated by Cronbach’s α and test–retest reliability) for each trait in every
world region (code 2.1.1). A trait was only included in the analysis for that region if
the coefficient exceeded 0.70. Cases in which the coefficient did not exceed 0.70 are
reported and discussed. There were no cases in which the coefficient did not exceed
0.70. Test–retest reliability of traits was not used to exclude traits from analysis.
Power analysis. Simulations showed that we had more than 95% power to detect the
key effect of interest (that is, two components meeting the criteria for replicating
Oosterhof and Todorov’s work, as described above). We used the open data from
Morrison etal.’s replication13 of Oosterhof and Todorov’s research to generate a
variance–covariance matrix representative of typical inter-relationships among the
13 traits tested in our study. We then generated 1,000 samples of 120 faces from
these distributions and ran our planned PCA (which is identical to that reported
by Oosterhof and Todorov) on each sample (see https://osf.io/87rbg/ for code and
data). The results of >99% of these analyses matched our criteria for replicating
Oosterhof and Todorov’s findings. Thus, 120 faces gave us more than 95% power to
replicate Oosterhof and Todorov’s results.
Robustness analyses. Oosterhof and Todorov extracted and interpreted components
with an eigenvalue greater than 1.0 using an unrotated PCA. As described above,
we directly replicated their method in our main analyses but acknowledge that this
type of analysis has been criticized.
First, it has been argued that EFA with rotation, rather than an unrotated PCA,
is more appropriate when one intends to measure correlated latent factors, as was
the case in the current study43,44. Second, the extraction rule of eigenvalues greater
than 1.0 has been criticized for not indicating the optimal number of components,
as well as for producing unreliable components45,46.
To address these limitations, we repeated our main analyses using EFA with an
oblimin rotation as the model and a parallel analysis to determine the number of
factors to extract. We also recalculated the congruence coefficient described above
for these EFA results (code 2.2.2).
We used parallel analysis to determine the number of factors to extract
because it has been described as yielding the optimal number of components (or
factors) across the largest array of scenarios43,47,48 (code 2.2.1). In a parallel analysis,
random data matrices are generated such that they have the same number of cases
and variables as the real data. The mean eigenvalue from the components of the
random data is compared with the eigenvalue for each component from the real
data. Components are then retained if their eigenvalues exceed those from the
randomly generated data49.
The purpose of these additional analyses was twofold: (1) to address potential
methodological limitations in the original study; and (2) to ensure that the
results of our replication of Oosterhof and Todorov’s study are robust to the
implementation of those more rigorous analytical techniques. The same criteria for
replicating Oosterhof and Todorov’s model described above were applied to this
analysis (code 2.2.1.3).
Reporting Summary. Further information on research design is available in the
Nature Research Reporting Summary linked to this article.
Data availability
Full data are publicly available at https://osf.io/87rbg/.
Code availability
Full analysis code is publicly available at https://osf.io/87rbg/.
Received: 18 May 2018; Accepted: 23 October 2020;
Published online: 4 January 2021
References
1. Olivola, C. Y. & Todorov, A. Elected in 100 milliseconds: appearance-based
trait inferences and voting. J. Nonverbal Behav. 34, 83–110 (2010).
2. Ritchie, K. L., Palermo, R. & Rhodes, G. Forming impressions of facial
attractiveness is mandatory. Sci. Rep. 7, 469 (2017).
3. Willis, J. & Todorov, A. First impressions: making up your mind aer 100 ms
exposure to a face. Psychol. Sci. 17, 592–598 (2006).
4. Olivola, C. Y., Funk, F. & Todorov, A. Social attributions from faces bias
human choices. Trends Cogn. Sci. 18, 566–570 (2014).
5. Todorov, A., Olivola, C. Y., Dotsch, R. & Mende-Siedlecki, P. Social
attributions from faces: determinants, consequences, accuracy, and functional
signicance. Annu. Rev. Psychol. 66, 519–545 (2015).
6. Van ‘t Wout, M. & Sanfey, A. G. Friend or foe: the eect of implicit
trustworthiness judgments in social decision-making. Cognition 108,
796–803 (2008).
7. Todorov, A., Mandisodza, A. N., Goren, A. & Hall, C. C. Inferences
of competence from faces predict election outcomes. Science 308,
1623–1626 (2005).
8. Langlois, J. H. etal. Maxims or myths of beauty? A meta-analytic and
theoretical review. Psychol. Bull. 126, 390–423 (2000).
9. Wilson, J. P. & Rule, N. O. Facial trustworthiness predicts extreme
criminal-sentencing outcomes. Psychol. Sci. 26, 1325–1331 (2015).
10. Todorov, A., Said, C. P., Engell, A. D. & Oosterhof, N. N. Understanding
evaluation of faces on social dimensions. Trends Cogn. Sci. 12,
455–460 (2008).
11. Jack, R. E. & Schyns, P. G. Toward a social psychophysics of face
communication. Annu. Rev. Psychol. 68, 269–297 (2017).
12. Oosterhof, N. N. & Todorov, A. e functional basis of face evaluation.
Proc. Natl Acad. Sci. USA 105, 11087–11092 (2008).
13. Morrison, D., Wang, H., Hahn, A. C., Jones, B. C. & DeBruine, L. M.
Predicting the reward value of faces and bodies from social perception.
PLoS ONE 12, e0185093 (2017).
14. Wang, H., Hahn, A. C., DeBruine, L. M. & Jones, B. C. e motivational
salience of faces is related to both their valence and dominance. PLoS ONE
11, e0161114 (2016).
15. Henrich, J., Heine, S. & Norenzayan, A. e weirdest people in the world?
Behav. Brain Sci. 33, 61–83 (2010).
16. Kline, M. A., Shamsudheen, R. & Broesch, T. Variation is the universal:
making cultural evolution work in developmental psychology. Phil. Trans. R.
Soc. B 373, 20170059 (2018).
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 165
RegisTeRed RepoRT Nature HumaN BeHaviour
17. Sutherland, C. A. M. etal. Facial rst impressions across culture: data-driven
modeling of Chinese and British perceivers’ unconstrained facial impressions.
Pers. Soc. Psychol. Bull. 44, 521–537 (2018).
18. Wang, H. etal. A data-driven study of Chinese participants’ social judgments
of Chinese faces. PLoS ONE 14, e0210315 (2019).
19. Han, C. etal. Cultural dierences in preferences for facial coloration. Evol.
Hum. Behav. 39, 154–159 (2018).
20. Perrett, D. I. etal. Eects of sexual dimorphism on facial attractiveness.
Nature 394, 884–887 (1998).
21. Xie, S. Y., Flake, J. K. & Hehman, E.Perceiver and target characteristics
contribute to impression formation dierently across race and gender. J. Pers.
Soc. Psychol. 117, 364–385 (2019).
22. Li, N. P., Valentine, K. A. & Patel, L. Mate preferences in the US and
Singapore: a cross-cultural test of the mate preference priority model. Pers.
Individ. Dier. 50, 291–294 (2011).
23. Ting-Toomey, S. in e Challenge of Facework: Cross-Cultural and
Interpersonal Issues (ed. Ting-Toomey, S.) 1–14 (State Univ. New York
Press, 1994).
24. Tan, C. B. Y., Stephen, I. D., Whitehead, R. & Sheppard, E. You look familiar:
how Malaysian Chinese recognize faces. PLoS ONE 7, e29714 (2012).
25. Chartier, C., McCarthy, R. & Urry, H. e Psychological Science Accelerator
(Association for Physical Science, 2018).
26. Chawla, D. S. A new ‘accelerator’ aims to bring big science to psychology.
Science https://doi.org/10.1126/science.aar4464 (2017).
27. Moshontz, H. etal. e Psychological Science Accelerator: advancing
psychology through a distributed collaborative network. Adv. Methods Pract.
Psychol. Sci. 1, 501–515 (2018).
28. Widaman, K. F. On common factor and principal component representations
of data: implications for theory and for conrmatory replications. Struct. Equ.
Modeling 25, 829–847 (2018).
29. Hehman, E., Sutherland, C. A., Flake, J. K. & Slepian, M. L.e unique
contributions of perceiver and target characteristics in person perception.
J. Pers. Soc. Psychol. 113, 513–529 (2017).
30. Sutherland, C. A., Rhodes, G., Burton, N. S. & Young, A. W.Do facial
rst impressions reect a shared social reality? Br. J. Psychol. 111,
215–232 (2020).
31. Oh, D., Dotsch, R., Porter, J. & Todorov, A.Gender biases in impressions from
faces: empirical studies and computational models. J. Exp. Psychol. Gen. 149,
323–342 (2020).
32. Oh, D., Shar, E. & Todorov, A. Economic status cues from clothes aect
perceived competence from faces. Nat. Hum. Behav. 4, 287–293 (2020).
33. Collova, J. R., Sutherland, C. A. & Rhodes, G. Testing the functional basis of
rst impressions: dimensions for children’s faces are not the same as for
adults’ faces. J. Pers. Soc. Psychol. 117, 900–924 (2019).
34. Stolier, R. M., Hehman, E., Keller, M. D., Walker, M. & Freeman, J. B. e
conceptual structure of face impressions. Proc. Natl Acad. Sci. USA 115,
9210–9215 (2018).
35. Stolier, R. M., Hehman, E. & Freeman, J. B. A dynamic structure of social
trait space. Trends Cogn. Sci. 22, 197–200 (2018).
36. Ma, D. S., Correll, J. & Wittenbrink, B. e Chicago Face Database: a
free stimulus set of faces and norming data. Behav. Res. Methods 47,
1122–1135 (2015).
37. Bainbridge, W. A., Isola, P. & Oliva, A. e intrinsic memorability of face
photographs. J. Exp. Psychol. Gen. 142, 1323–1334 (2013).
38. Sutherland, C. A. etal. Social inferences from faces: ambient images generate
a three-dimensional model. Cognition 127, 105–118 (2013).
39. Burt, C. e factorial study of temperament traits. Br. J. Psychol. Stat. Sect. 1,
178–203 (1948).
40. Tucker, L. R. A Method for Synthesis of Factor Analysis Studies Personnel
Research Section Report No. 984 (Department of the Army, 1951).
41. Davenport, E. C. Jr Signicance testing of congruence coecients: a good
idea? Educ. Psychol. Meas. 50, 289–296 (1990).
42. Lorenzo-Seva, U. & ten Berge, J. M. F. Tucker’s congruence coecient as a
meaningful index of factor similarity. Methodology 2, 57–64 (2006).
43. Fabrigar, L. R., Wegener, D. T., MacCallum, R. C. & Strahan, E. J. Evaluating
the use of exploratory factor analysis in psychological research. Psychol.
Methods 4, 272–299 (1999).
44. Park, H. S., Dailey, R. & Lemus, D. e use of exploratory factor analysis and
principal components analysis in communication research. Hum. Commun.
Res. 28, 562–577 (2002).
45. Cli, N. e eigenvalues-greater-than-one rule and the reliability of
components. Psychol. Bull. 103, 276–279 (1988).
46. Zwick, W. R. & Velicer, W. F. Comparison of ve rules for determining the
number of components to retain. Psychol. Bull. 99, 432–442 (1986).
47. O’Connor, B. P. SPSS and SAS programs for determining the number of
components using parallel analysis and Velicer’s MAP test. Behav. Res.
Methods Instrum. Comput. 32, 396–402 (2000).
48. Schmitt, T. A. Current methodological considerations in exploratory and
conrmatory factor analysis. J. Psychoeduc. Assess. 29, 304–321 (2011).
49. Courtney, M. G. R. Determining the number of factors to retain in EFA:
using the SPSS R-Menu v2.0 to make more judicious estimations. Pract.
Assess. Res. Eval. 18, 1–14 (2013).
Acknowledgements
C.L. was supported by the Vienna Science and Technology Fund (WWTF VRG13-007);
L.M.D. was supported by ERC 647910 (KINSHIP); D.I.B. and N.I. received funding from
CONICET, Argentina; L.K., F.K. and Á. Putz were supported by the European Social
Fund (EFOP-3.6.1.-16-2016-00004; ‘C omprehensive Development for Implementing
Smart Specialization Strategies at the University of Pécs’). K.U. and E. Vergauwe were
supported by a grant from the Swiss National Science Foundation (PZ00P1_154911 to E.
Vergauwe). T.G. is supported by the Social Sciences and Humanities Research Council
of Canada (SSHRC). M.A.V. was supported by grants 2016-T1/SOC-1395 (Comunidad
de Madrid) and PSI2017-85159-P (AEI/FEDER UE). K.B. was supported by a grant
from the National Science Centre, Poland (number 2015/19/D/HS6/00641). J. Bonick
and J.W.L. were supported by the Joep Lange Institute. G.B. was supported by the Slovak
Research and Development Agency (APVV-17-0418). H.I.J. and E.S. were supported
by a French National Research Agency ‘Investissements d’Avenir’ programme grant
(ANR-15-IDEX-02). T.D.G. was supported by an Australian Government Research
Training Program Scholarship. The Raipur Group is thankful to: (1) the University
Grants Commission, New Delhi, India for the research grants received through its
SAP-DRS (Phase-III) scheme sanctioned to the School of Studies in Life Science;
and (2) the Center for Translational Chronobiology at the School of Studies in Life
Science, PRSU, Raipur, India for providing logistical support. K. Ask was supported by
a small grant from the Department of Psychology, University of Gothenburg. Y.Q. was
supported by grants from the Beijing Natural Science Foundation (5184035) and CAS
Key Laboratory of Behavioral Science, Institute of Psychology. N.A.C. was supported
by the National Science Foundation Graduate Research Fellowship (R010138018). We
acknowledge the following research assistants: J. Muriithi and J. Ngugi (United States
International University Africa); E. Adamo, D. Cafaro, V. Ciambrone, F. Dolce and E.
Tolomeo (Magna Græcia University of Catanzaro); E. De Stefano (University of Padova);
S. A. Escobar Abadia (University of Lincoln); L. E. Grimstad (Norwegian School of
Economics (NHH)); L. C. Zamora (Franklin and Marshall College); R. E. Liang and R.
C. Lo (Universiti Tunku Abdul Rahman); A. Short and L. Allen (Massey University, New
Zealand), A. Ateş, E. Güneş and S. Can Özdemir (Boğaziçi University); I. Pedersen and T.
Roos (Åbo Akademi University); N. Paetz (Escuela de Comunicación Mónica Herrera);
J. Green (University of Gothenburg); M. Krainz (University of Vienna, Austria); and B.
Todorova (University of Vienna, Austria). The funders had no role in study design, data
collection and analysis, decision to publish or preparation of the manuscript.
Author contributions
Conceptualization: B.C.J., L.M.D., J.K.F., J.P.W., J.B.F., S.Á.-S., H.I., S.M.J.J., H.L.
Data curation: B.C.J., L.M.D., N.C.A., N.G.B., Y.Q., J.W.L., K.G., G.M.M., J.G.L.,
J.B.F., P.C., A.P., N.P., S.P., M.M.S., B.P., M.J.B., V.K., J.P., D.S., S.C.W., J.V.V., P.S.F.,
C.R.C., N.A.C.
Formal analysis: B.C.J., L.M.D., J.K.F., Y.Q., J.B.F.
Funding acquisition: N.C.O., Y.Q., J.W.L., C.C., J. Leongómez, O.R.S., E. Valderrama,
M.V.-A., J.G.L., M.C.P., J.B.F., J.K.O., G.K., H.I., H.D.F., T.J.S.L., E. Vergauwe, K. Ask,
K.W.T., M.I., C.L., P.S.F., C.R.C.
Investigation: B.C.J., L.M.D., M.T.L., J.A., I.L.G.N., N.G.B., S.C.L., F.F., M.L.W.,
C.P.C., M.A.V., S.A.S., N.C.O., D.P.C., A.W., Y.Q., H.M., P. Suavansri, T.R.E., J. Bonick,
J.W.L., C.C., A. Kapucu, A. Karaaslan, J. Leongómez, O.R.S., E. Valderrama, M.V.-A.,
B.A., P. Szecsi, M. Andreychik, E.D.M., C.B., C.-P.H., Q.-L.L., L.A.V., K.B., K.G., I.S., S.S.,
R.A., C.M., W.V., Z.J., Q.W., G.M.M., I.D.S., J.G.L., M.C.P., J.D.A., E.H., S.Y.X., W.J.C.,
M. Seehuus, J.P.W., E.K., M.P.-P., A.E.B.-S., A.d.-G., I.G.-S., H.-H.W., J.B.F., D.W.O., V.S.,
T.E.S., C.A.L., C.L.C., A.K.P., J. Bavolar, P. Kačmár, I. Zakharov, S.Á.-S., E.B., M.T., K.S.,
C.D.C., J.W.S., J.K.O., A.-S.L., T.D.G., J.A.O., B.J.W.D., L.M.S., G.R., M.J.B., B.J., D.R.,
G.K., V.A.F., H.L.U., S.-C.C., G.P., Z.V., D.M.B.-B., H.I., N.V.d.L., C.B.Y.T., V.K., M.F.C.,
H.D.F., D.I.B., G.G., J.P., C.S., K.A.Ś., E.M.O.K., D.S., B.S., M. Sirota, G.V.S., T.J.S.L., K.U.,
E. Vergauwe, J.S., K. Ask, C.J.J.v.Z., A. Körner, S.C.W., J. Boudesseul, F.R.-D., K.L.R.,
N.M.M., K.R.B., D.W., A.R.G.-F., M. Anne, S.M.J.J., K.M.L., T.K.N., C.K.T., J.H.Z.,
A.D.R., L.K., M. Vianello, N.I., A.C., S.L., J. Lutz, M. Adamkovic, P.B., G.B., I.R., V.C.,
K.P., N.K.S., K.W.T., C.A.T., A.M.F., R.M.C.S.H., J.V.V., N.S.C.-F., M.F.-A., J.H., A.M.,
M. Sharifian, B.F., H.L., M.I., C.L., E.P., M. Voracek, J.O., E.M.G., A.A., A.A.Ö., M.T.C.,
B.B.-D., M.A.K., C.O., T.G., J.K.M., Y.D., X.Y., S. Alper, P.S.F., C.R.C., N.A.C.
Methodology: B.C.J., L.M.D., J.K.F., S.C.L., L.A.V., M. Seehuus, S. Azouaghe, A.B.,
J.E., J.P.W., J.B.F., C.A.L., C.D.C., K.H., B.J., J.W., G.K., H.I., T.B., N.V.d.L., H.D.F., J.P.,
F.M.A.W., S.M.J.J., H.L.
Project administration: B.C.J., L.M.D., N.G.B., S.C.L., M.L.W., M.G., A.S., N.C.O.,
A.W., Y.Q., H.M., R.M.S., J. Bonick, J.W.L., C.C., A. Kapucu, A. Karaaslan, J. Leongómez,
O.R.S., E. Valderrama, M.V.-A., B.A., C.B., C.-P.H., L.A.V., K.B., K.G., I.S., S.S., I.D.S.,
M.C.P., S.Y.X., W.J.C., M. Seehuus, A.d.-G., I.G.-S., C.-C.K., J.B.F., D.W.O., C.A.L., J.
Bavolar, P. Kačmár, I. Zakharov, K.S., C.D.C., J.W.S., J.L.B., J.A.O., B.J.W.D., M.J.B., B.J.,
D.R., G.P., Z.V., E.S., N.V.d.L., V.K., M.F.C., H.D.F., J.P., C.S., K.A.S., E.M.O.K., B.S., M.
Sirota, T.J.S.L., K.U., E. Vergauwe, K. Ask, C.J.J.v.Z., S.C.W., J. Boudesseul, F.R.-D., K.L.R.,
D.W., S.M.J.J., C.K.T., J.H.Z., L.K., S.L., V.C., N.K.S., K.W.T., R.M.C.S.H., J.V.V., A.M., M.
Sharifian, B.F., H.L., C.L., E.P., M. Voracek, A.A., A.A.Ö., M.A.K., T.G., X.Y., S. Alper,
P.S.F., C.R.C., N.A.C.
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav
166
RegisTeRed RepoRT
Nature HumaN BeHaviour
Resources: B.C.J., L.M.D., M.T.L., S.C.L., C.P.C., M.A.V., S.A.S., A.W., Y.Q.,
K. Ariyabuddhiphongs, S.J., H.M., P. Suavansri, N.T., R.M.S., C.C., A. Kapucu, J.
Leongómez, M.V.-A., N.H., C.B., L.A.V., K.B., K.G., Z.J., G.M.M., I.D.S., J.G.L., S.Y.X.,
W.J.C., M. Seehuus, S. Azouaghe, A.B., J.E., A.d.-G., C.-C.K., J.B.F., C.A.L., A.K.P., P.
Kačmár, I. Zakharov, E.B., K.S., C.D.C., J.K.O., J.L.B., B.J.W.D., D.R., W.W.A.S., S.-C.C.,
G.P., D.M.B.-B., T.B., C.B.Y.T., V.K., H.D.F., G.G., C.S., K.A.S., E.M.O.K., B.S., M. Sirota,
G.V.S., T.J.S.L., K.U., E. Vergauwe, K.J., K. Ask, J. Boudesseul, F.R.-D., N.M.M., S.M.J.J.,
C.K.T., A.D.R., F.K., Á.P., P.T., M. Vianello, A.C., S.L., J. Lutz, M. Adamkovic, P.B., V.C.,
A.M.F., R.M.C.S.H., J.V.V., N.S.C.-F., M.F.-A., A.M., M. Sharifian, H.L., C.L., M. Voracek,
E.M.G., A.A.Ö., M.A.K., C.O., X.Y., S. Alper, P.S.F., C.R.C. Software: B.C.J., L.M.D.,
J.K.F., G.M.M., I.D.S., N.P., B.P., C.D.C., H.D.F., C.S., K.R.B., R.M.C.S.H., C.R.C., N.A.C.
Supervision: B.C.J., L.M.D., J.K.F., M.T.L., S.C.L., M.L.W., N.C.O., A.W., H.M., J.W.L.,
C.C., A. Kapucu, J. Leongómez, O.R.S., E. Valderrama, M.V.-A., M. Andreychik, E.D.M.,
C.B., L.A.V., K.B., I.D.S., M.C.P., E.H., W.J.C., M. Seehuus, C.-C.K., J.B.F., C.A.L., P.
Kačmár, I. Zakharov, K.S., C.D.C., J.W.S., J.K.O., A.-S.L., J.L.B., J.A.O., B.J.W.D., M.J.B.,
H.I., V.K., M.F.C., H.D.F., J.P., C.S., E.M.O.K., D.S., B.S., M. Sirota, T.J.S.L., K.U., E.
Vergauwe, K. Ask, C.J.J.v.Z., D.W., S.M.J.J., A.C., S.L., K.P., N.K.S., K.W.T., A.M.F., J.V.V.,
M. Sharifian, M.I., C.L., M. Voracek, A.A., A.A.Ö., M.A.K., S. Alper, P.S.F., C.R.C., N.A.C.
Validation: B.C.J., L.M.D., J.K.F., C.C., Q.W., S.Y.X., M. Seehuus, C.L.C., A.K.P., I.
Zakharov, J.W.S., E.S., V.K., H.D.F., J.P., M. Sirota, E. Vergauwe, C.J.J.v.Z., P.T., J.H., M.
Voracek, M.A.K.
Visualization: B.C.J., L.M.D., J.K.F., H.D.F., M.A.K., P.S.F. Writing (original draft):
B.C.J., L.M.D., J.K.F., F.F., Y.Q., C.B., I.G.-S., J.B.F., K.S., B.J.W.D., G.K., H.L.U., H.I., H.D.F.,
D.I.B., J.P., C.S., D.S., K.L.R., S.M.J.J., A.D.R., N.K.S., J.O., A.A.Ö., M.A.K., P.S.F., N.A.C.
Writing (review & editing): B.C.J., L.M.D., J.K.F., M.T.L., J.A., I.L.G.N., S.C.L.,
F.F., M.L.W., M.A.V., A.S., D.P.C., A.W., Y.Q., K. Ariyabuddhiphongs, H.M., T.R.E., J.
Bonick, J.W.L., C.C., J. Leongómez, B.A., N.H., P. Szecsi, M. Andreychik, E.D.M., C.B.,
N.L., L.A.V., K.B., I.S., S.S., Z.J., I.D.S., M.C.P., J.D.A., E.H., S.Y.X., W.J.C., M. Seehuus,
S. Azouaghe, A.B., J.E., J.P.W., E.K., M.P.-P., A.E.B.-S., A.d.-G., J.B.F., V.S., T.E.S., C.A.L.,
C.L.C., P.C., P. Kujur, A.P., N.P., A.K.P., S.P., M.M.S., B.P., P. Kačmár, I. Zakharov, S.Á.-S.,
E.B., M.T., K.S., C.D.C., J.W.S., J.K.O., A.-S.L., J.L.B., T.D.G., J.A.O., B.J.W.D., G.R., M.J.B.,
K.H., B.J., G.K., V.A.F., H.L.U., G.P., Z.V., H.I., T.B., N.V.d.L., C.B.Y.T., V.K., M.F.C.,
H.D.F., D.I.B., G.G., C.S., E.M.O.K., D.S., B.S., M. Sirota, T.J.S.L., K.U., E. Vergauwe,
J.S., K. Ask, C.J.J.v.Z., A. Körner, K.L.R., K.R.B., D.W., A.R.G.-F., S.M.J.J., T.K.N., C.K.T.,
J.H.Z., M. Vianello, N.I., M. Adamkovic, G.B., I.R., V.C., K.P., N.K.S., K.W.T., C.A.T.,
A.M.F., R.M.C.S.H., J.V.V., B.F., H.L., C.L., E.P., M. Voracek, J.O., E.M.G., A.A., A.A.Ö.,
B.B.-D., M.A.K., T.G., J.K.M., Y.D., P.S.F., C.R.C., N.A.C.
The following people did not indicate specific contributions: A.F.D., A.C.H.,
A.D.L.R.-G., D.R.F., D.T., E.T., E.G.-S., H.I.J., I. Zettler, I.R.P., J.A.M.-R., J.D.L., L.N.,
L.F.A., M.A.C.V., M.M.A., M.L.B.-G., M.H.S., N.O.R., P.P., P.F., R.J.M., S.G., S.J.C., T.H.,
V.K.M.S., W.-J.Y.
Competing interests
The authors declare no competing interests.
Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/
s41562-020-01007-2.
Correspondence and requests for materials should be addressed to B.C.J.
Peer review information Primary Handling Editor: Stavroula Kousta.
Reprints and permissions information is available at www.nature.com/reprints.
© The Author(s), under exclusive licence to Springer Nature Limited 2021
Benedict C. Jones 1,156 ✉ , Lisa M. DeBruine 2,156, Jessica K. Flake3,156, Marco Tullio Liuzza 4,
Jan Antfolk 5, Nwadiogo C. Arinze 6, Izuchukwu L. G. Ndukaihe6, Nicholas G. Bloxsom7,
Savannah C. Lewis 7, Francesco Foroni 8, Megan L. Willis 8, Carmelo P. Cubillas9,
Miguel A. Vadillo 9, Enrique Turiegano 10, Michael Gilead11, Almog Simchon 11, S. Adil Saribay 12,
Nicholas C. Owsley13, Chaning Jang 13, Georgina Mburu13, Dustin P. Calvillo14, Anna Wlodarczyk 15,
Yue Qi16, Kris Ariyabuddhiphongs 17, Somboon Jarukasemthawee17, Harry Manley 17,
Panita Suavansri 17, Nattasuda Taephant17, Ryan M. Stolier 18, Thomas R. Evans 19,
Judson Bonick 20, Jan W. Lindemans 20, Logan F. Ashworth21, Amanda C. Hahn 21,
Coralie Chevallier 22, Aycan Kapucu 23, Aslan Karaaslan 23, Juan David Leongómez 24,
Oscar R. Sánchez 24, Eugenio Valderrama 24, Milena Vásquez-Amézquita 24, Nandor Hajdu 25,26,
Balazs Aczel 26, Peter Szecsi 26, Michael Andreychik 27, Erica D. Musser 28, Carlota Batres 29,
Chuan-Peng Hu 30, Qing-Lan Liu31, Nicole Legate 32, Leigh Ann Vaughn 33,
Krystian Barzykowski 34, Karolina Golik 34, Irina Schmid 35, Stefan Stieger 35, Richard Artner 36,
Chiel Mues 36, Wolf Vanpaemel 37, Zhongqing Jiang 38, Qi Wu38, Gabriela M. Marcu 39,
Ian D. Stephen 40, Jackson G. Lu 41, Michael C. Philipp 42, Jack D. Arnal 43, Eric Hehman3,
Sally Y. Xie3, William J. Chopik 44, Martin Seehuus45, Soufian Azouaghe 46,47, Abdelkarim Belhaj46,
Jamal Elouafa46, John P. Wilson 48, Elliott Kruse49, Marietta Papadatou-Pastou 50,
Anabel De La Rosa-Gómez 51, Alan E. Barba-Sánchez 51, Isaac González-Santoyo 52,
Tsuyueh Hsu 53, Chun-Chia Kung 53, Hsiao-Hsin Wang53, Jonathan B. Freeman 54,
Dong Won Oh 55, Vidar Schei 56, Therese E. Sverdrup 56, Carmel A. Levitan 57, Corey L. Cook58,
Priyanka Chandel 59, Pratibha Kujur 59, Arti Parganiha 59, Noorshama Parveen 59,
Atanu Kumar Pati 59, Sraddha Pradhan 59, Margaret M. Singh59, Babita Pande 60, Jozef Bavolar 61,
Pavol Kačmár 61, Ilya Zakharov 62, Sara Álvarez-Solas 63, Ernest Baskin 64, Martin Thirkettle 65,
Kathleen Schmidt 66, Cody D. Christopherson 67, Trinity Leonis67, Jordan W. Suchow68,
Jonas K. Olofsson 69, Teodor Jernsäther 69, Ai-Suan Lee 70, Jennifer L. Beaudry 71,
Taylor D. Gogan 71, Julian A. Oldmeadow 71, Benjamin Balas72, Laura M. Stevens73,
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 167
RegisTeRed RepoRT Nature HumaN BeHaviour
Melissa F. Colloff 73, Heather D. Flowe 73, Sami Gülgöz 74, Mark J. Brandt 75, Karlijn Hoyer 75,
Bastian Jaeger 75, Dongning Ren 75, Willem W. A. Sleegers75, Joeri Wissink75,
Gwenaël Kaminski 76, Victoria A. Floerke 77, Heather L. Urry 77, Sau-Chin Chen 78, Gerit Pfuhl 79,
Zahir Vally 80, Dana M. Basnight-Brown81, Hans I. Jzerman 47, Elisa Sarda47, Lison Neyroud82,
Touhami Badidi83, Nicolas Vander Linden 84, Chrystalle B. Y. Tan 85, Vanja Kovic86,
Waldir Sampaio 87, Paulo Ferreira 88, Diana Santos 88, Debora I. Burin 89, Gwendolyn Gardiner90,
John Protzko 91, Christoph Schild92, Karolina A. Ścigała92, Ingo Zettler 92, Erin M. O’Mara Kunz 93,
Daniel Storage 94, Fieke M. A. Wagemans95, Blair Saunders96, Miroslav Sirota97, Guyan V. Sloane97,
Tiago J. S. Lima 98, Kim Uittenhove 99, Evie Vergauwe99, Katarzyna Jaworska 2, Julia Stern 100,
Karl Ask 101, Casper J. J. van Zyl102, Anita Körner 103, Sophia C. Weissgerber 103,
Jordane Boudesseul 104, Fernando Ruiz-Dodobara104, Kay L. Ritchie105, Nicholas M. Michalak 106,
Khandis R. Blake 107,108, David White107, Alasdair R. Gordon-Finlayson 109, Michele Anne 110,
Steve M. J. Janssen 110, Kean Mun Lee 110, Tonje K. Nielsen111, Christian K. Tamnes 111,
Janis H. Zickfeld 112, Anna Dalla Rosa 113, Michelangelo Vianello 113, Ferenc Kocsor114,
Luca Kozma 114, Ádám Putz114, Patrizio Tressoldi 115, Natalia Irrazabal 116, Armand Chatard 117,
Samuel Lins 118, Isabel R. Pinto118, Johannes Lutz119, Matus Adamkovic 120, Peter Babincak 120,
Gabriel Baník 120, Ivan Ropovik 121,122, Vinet Coetzee 123, Barnaby J. W. Dixson 124,
Gianni Ribeiro 124, Kim Peters 124, Niklas K. Steffens 124, Kok Wei Tan125,
Christopher A. Thorstenson126, Ana Maria Fernandez 127, Rafael M. C. S. Hsu128,
Jaroslava V. Valentova 128, Marco A. C. Varella 128, Nadia S. Corral-Frías 129,
Martha Frías-Armenta 129, Javad Hatami130, Arash Monajem 130, MohammadHasan Sharifian130,
Brooke Frohlich131, Hause Lin 132, Michael Inzlicht 132, Ravin Alaei132, Nicholas O. Rule132,
Claus Lamm 133, Ekaterina Pronizius 133, Martin Voracek 133, Jerome Olsen 134,
Erik Mac Giolla 135, Aysegul Akgoz136, Asil A. Özdoğru 136, Matthew T. Crawford137,
Brooke Bennett-Day 138, Monica A. Koehn 139, Ceylan Okan140, Tripat Gill 141, Jeremy K. Miller 142,
Yarrow Dunham 143, Xin Yang 143, Sinan Alper 144, Martha Lucia Borras-Guevara145, Sun Jun Cai146,
Dong Tiantian 146, Alexander F. Danvers147, David R. Feinberg 148, Marie M. Armstrong 148,
Eva Gilboa-Schechtman149, Randy J. McCarthy150, Jose Antonio Muñoz-Reyes151, Pablo Polo151,
Victor K. M. Shiramazu 152, Wen-Jing Yan153, Lilian Carvalho 154, Patrick S. Forscher 82,
Christopher R. Chartier 7 and Nicholas A. Coles155
1School of Psychological Sciences and Health, University of Strathclyde, Glasgow, UK. 2Institute of Neuroscience and Psychology, University of Glasgow,
Glasgow, UK. 3Department of Psychology, McGill University, Montreal, Québec, Canada. 4Department of Medical and Surgical Sciences, Magna Græcia
University of Catanzaro, Catanzaro, Italy. 5Faculty of Arts, Psychology and Theology, Åbo Akademi University, Turku, Finland. 6Department of Psychology,
Alex Ekwueme Federal University Ndufu Alike, Ikwo, Nigeria. 7Department of Psychology, Ashland University, Danville, CA, USA. 8School of Behavioural
and Health Sciences, Australian Catholic University, Sydney, New South Wales, Australia. 9Department of Basic Psychology, Autonomous University of
Madrid, Madrid, Spain. 10Department of Biology, Autonomous University of Madrid, Madrid, Spain. 11Department of Psychology, Ben-Gurion University of
the Negev, Beersheba, Israel. 12Department of Psychology, Boğaziçi University, Beşiktaş, Turkey. 13Busara Center for Behavioral Economics, Nairobi, Kenya.
14Psychology Department, California State University San Marcos, San Marcos, CA, USA. 15School of Psychology, Catholic University of the North,
Antofagasta, Chile. 16Department of Psychology, Renmin University of China, Beijing, China. 17Faculty of Psychology, Chulalongkorn University, Bangkok,
Thailand. 18Department of Psychology, Columbia University, New York, NY, USA. 19School of Psychological, Social and Behavioural Sciences, Coventry
University, Coventry, UK. 20Center for Advanced Hindsight, Duke University, Durham, NC, USA. 21Department of Psychology, Humboldt State University,
Arcata, CA, USA. 22Laboratoire de Neurosciences Cognitives et Computationnelles, Département d’Études Cognitives, INSERM U960, École Normale
Supérieure, Paris, France. 23Psychology Department, Ege University, İzmir, Turkey. 24 Faculty of Psychology, Universidad El Bosque, Bogotá, Colombia.
25Doctoral School of Psychology, ELTE Eötvös Loránd University, Budapest, Hungary. 26 Institute of Psychology, ELTE Eötvös Loránd University, Budapest,
Hungary. 27Department of Psychology, Fairfield University, Fairfield, CT, USA. 28Department of Psychology, Florida International University, Miami, FL, USA.
29Department of Psychology, Franklin and Marshall College, Lancaster, PA, USA. 30Leibniz Institute for Resilience Research, Mainz, Germany. 31Department
of Psychology, Hubei University, Wuhan, China. 32Department of Psychology, Illinois Institute of Technology, Chicago, IL, USA. 33Department of Psychology,
Ithaca College, Ithaca, NY, USA. 34Institute of Psychology, Jagiellonian University, Kraków, Poland. 35Department of Psychology and Psychodynamics, Karl
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav
168
RegisTeRed RepoRT
Nature HumaN BeHaviour
Landsteiner University of Health Sciences, Krems an der Donau, Austria. 36Research Group of Quantitative Psychology and Individual Differences,
Katholieke Universiteit Leuven, Leuven, Belgium. 37Faculty of Psychology and Educational Sciences, Katholieke Universiteit Leuven, Leuven, Belgium.
38Department of Psychology, Liaoning Normal University, Dalian, China. 39Department of Psychology, Lucian Blaga University of Sibiu, Sibiu, Romania.
40Department of Psychology, Macquarie University, Sydney, New South Wales, Australia. 41Sloan School of Management, Massachusetts Institute of
Technology, Cambridge, MA, USA. 42School of Psychology, Massey University, Palmerston North, New Zealand. 43Psychology Department, McDaniel
College, Westminster, CO, USA. 44Department of Psychology, Michigan State University, East Lansing, MI, USA. 45Department of Psychology, Middlebury
College, Middlebury, VT, USA. 46Department of Psychology, Mohammed V University in Rabat, Rabat, Morocco. 47LIP/PC2S, Université Grenoble Alpes,
Grenoble, France. 48Psychology Department, Montclair State University, Montclair, NJ, USA. 49EGADE Business School, Monterrey Institute of Technology
and Higher Education, Monterrey, Mexico. 50School of Education, National and Kapodistrian University of Athens, Athens, Greece. 51School of Higher
Studies Iztacala, National Autonomous University of Mexico, Mexico City, Mexico. 52Department of Psychology, National Autonomous University of
Mexico, Mexico City, Mexico. 53Department of Psychology, National Cheng Kung University, Tainan City, Taiwan. 54Department of Psychology and Center
for Neural Science, New York University, New York, NY, USA. 55Department of Psychology, New York University, New York, NY, USA. 56Department of
Strategy and Management, Norwegian School of Economics (NHH), Bergen, Norway. 57Department of Cognitive Science, Occidental College, Los Angeles,
CA, USA. 58Department of Psychology, Pacific Lutheran University, Tacoma, WA, USA. 59School of Studies in Life Science, Pandit Ravishankar Shukla
University, Raipur, India. 60Center for Basic Sciences, Pandit Ravishankar Shukla University, Raipur, India. 61Department of Psychology, Pavol Jozef Šafárik
University in Košice, Košice, Slovakia. 62Developmental Behavioral Genetics Lab, Psychological Institute of Russian Academy of Education, Moscow, Russia.
63Facultad de Ciencias dela Vida, Universidad Regional Amazónica Ikiam, Guayaquil, Ecuador. 64Department of Food Marketing, Saint Joseph’s University,
Philadelphia, PA, USA. 65Centre for Behavioural Science and Applied Psychology, Sheffield Hallam University, Sheffield, UK. 66School of Psychological and
Behavioral Sciences, Southern Illinois University, Carbondale, IL, USA. 67 Psychology Department, Southern Oregon University, Ashland, OR, USA. 68School
of Business, Stevens Institute of Technology, Hoboken, NJ, USA. 69Department of Psychology, Stockholm University, Stockholm, Sweden. 70Department of
Psychology, Sunway University, Subang Jaya, Malaysia. 71Department of Psychological Sciences, Swinburne University of Technology, Melbourne, Victoria,
Australia. 72Department of Psychology, North Dakota State University, Fargo, ND, USA. 73School of Psychology, University of Birmingham, Birmingham, UK.
74Koç University, Istanbul, Turkey. 75Department of Social Psychology, Tilburg University, Tilburg, the Netherlands. 76CLLE, Toulouse University, Toulouse,
France. 77Department of Psychology, Tufts University, Medford, MA, USA. 78Department of Human Development and Psychology, Tzu-Chi University,
Hualien, Taiwan. 79Department of Psychology, UiT The Arctic University of Norway, Tromsø, Norway. 80Department of Psychology and Counseling, United
Arab Emirates University, Abu Dhabi, United Arab Emirates. 81United States International University Africa, Nairobi, Kenya. 82Department of Psychology,
Université Grenoble Alpes, Saint-Martin-d’Hères, France. 83Department of Psychology, Université Ibn Tofail, Kénitra, Morocco. 84Center for Social and
Cultural Psychology, Université Libre de Bruxelles, Brussels, Belgium. 85Department of Community and Family Medicine, Universiti Malaysia Sabah, Kota
Kinabalu, Malaysia. 86Department of Psychology, Faculty of Philosophy, University of Belgrade, Belgrade, Serbia. 87Universidade Federal de São Carlos, São
Paulo, Brazil. 88Universidade Federal da Grande Dourados, Dourados, Brazil. 89Instituto de Investigaciones, Facultad de Psicologia, Universidad de Buenos
Aires, Buenos Aires, Argentina. 90Department of Psychology, University of California, Riverside, Riverside, CA, USA. 91Department of Psychological and
Brain Sciences, University of California, Santa Barbara, Santa Barbara, CA, USA. 92Department of Psychology, University of Copenhagen, Copenhagen,
Denmark. 93Department of Psychology, University of Dayton, Dayton, OH, USA. 94Department of Psychology, University of Denver, Denver, CO, USA.
95Institute for Socio-Economics, University of Duisburg-Essen, Essen, Germany. 96School of Social Sciences, University of Dundee, Dundee, UK.
97Department of Psychology, University of Essex, Colchester, UK. 98Department of Social and Work Psychology, University of Brasília, Brasília, Brazil.
99Faculty of Psychology and Educational Sciences, University of Geneva, Geneva, Switzerland. 100Department of Psychology, University of Goettingen,
Goettingen, Germany. 101Department of Psychology, University of Gothenburg, Gothenburg, Sweden. 102Department of Psychology, University of
Johannesburg, Johannesburg, South Africa. 103Department of Psychology, University of Kassel, Kassel, Germany. 104Institute of Scientific Research, Faculty
of Psychology, University of Lima, Lima, Peru. 105School of Psychology, University of Lincoln, Lincoln, UK. 106Department of Psychology, University of
Michigan, Ann Arbor, MI, USA. 107Evolution and Ecology Research Centre, University of New South Wales Sydney, Sydney, New South Wales, Australia.
108Melbourne School of Psychological Sciences, University of Melbourne, Melbourne, Victoria, Australia. 109Faculty of Health, Education and Society,
University of Northampton, Northampton, UK. 110School of Psychology, University of Nottingham Malaysia, Semenyih, Malaysia. 111Department of
Psychology, University of Oslo, Oslo, Norway. 112Department of Management, Aarhus University, Aarhus, Denmark. 113Department of Philosophy, Sociology,
Education and Applied Psychology, University of Padova, Padova, Italy. 114Institute of Psychology, University of Pécs, Pécs, Hungary. 115Department of
General Psychology, University of Padova, Padova, Italy. 116Faculty of Social Sciences, University of Palermo, Buenos Aires, Argentina. 117Psychology
Department, University of Poitiers, Poitiers, France. 118Department of Psychology, University of Porto, Porto, Portugal. 119Department of Psychology,
University of Potsdam, Potsdam, Germany. 120Institute of Psychology, Faculty of Arts, University of Presov, Presov, Slovakia. 121Faculty of Education,
University of Presov, Presov, Slovakia. 122Institute for Research and Development of Education, Faculty of Education, Charles University, Prague, Czechia.
123Department of Biochemistry, Genetics and Microbiology, University of Pretoria, Pretoria, South Africa. 124School of Psychology, University of Queensland,
Brisbane, Queensland, Australia. 125School of Psychology and Clinical Language Sciences, University of Reading Malaysia, Johor, Malaysia. 126Department of
Clinical and Social Sciences in Psychology, University of Rochester, Rochester, NY, USA. 127School of Psychology, University of Santiago, Chile, Santiago,
Chile. 128Institute of Psychology, Department of Experimental Psychology, University of São Paulo, São Paulo, Brazil. 129Department of Psychology,
University of Sonora, Hermosillo, Mexico. 130Department of Psychology, University of Tehran, Tehran, Iran. 131Department of Psychology, University of
Tennessee, Knoxville, Knoxville, TN, USA. 132Department of Psychology, University of Toronto, Toronto, Ontario, Canada. 133Department of Cognition,
Emotion, and Methods in Psychology, Faculty of Psychology, University of Vienna, Vienna, Austria. 134Department of Applied Psychology: Work, Education
and Economy, Faculty of Psychology, University of Vienna, Vienna, Austria. 135Department of Behavioral Sciences, University West, Trollhättan, Sweden.
136Department of Psychology, Üsküdar University, Istanbul, Turkey. 137School of Psychology, Victoria University of Wellington, Wellington, New Zealand.
138Department of Psychology, Wesleyan College, Middletown, CT, USA. 139Discipline of Psychology, Faculty of Health, University of Canberra, Canberra,
Australian Capital Territory, Australia. 140School of Social Science and Psychology, Western Sydney University, Sydney, New South Wales, Australia.
141Lazaridis School of Business and Economics, Wilfrid Laurier University, Waterloo, Ontario, Canada. 142Department of Psychology, Willamette University,
Salem, OR, USA. 143Department of Psychology, Yale University, New Haven, CT, USA. 144Department of Psychology, Yasar University, Izmir, Turkey.
145University of St. Andrews, St. Andrews, UK. 146Qufu Normal University, Jining, China. 147University of Oklahoma, Norman, OK, USA. 148Department of
Psychology, Neuroscience, and Behaviour, McMaster University, Hamilton, Ontario, Canada. 149Bar-Ilan University, Tel Aviv, Israel. 150Northern Illinois
University, DeKalb, IL, USA. 151Playa Ancha University of Educational Sciences, Valparaiso, Chile. 152Federal University of Rio Grande do Norte, Rio Grande
do Norte, Brazil. 153Wenzhou University, Wenzhou, China. 154FGV/EAESP, Sao Paulo, Brazil. 155Harvard Kennedy School, Harvard University, Cambridge, MA,
USA. 156These authors contributed equally: Benedict C. Jones, Lisa M. DeBruine, Jessica K. Flake. e-mail: psysciacc.001@gmail.com
NATURE HUMAN BEHAVIOUR | VOL 5 | JANUARY 2021 | 159–169 | www.nature.com/nathumbehav 169
1
nature research | reporting summary April 2020
Corresponding author(s): Benedict Jones
Last updated by author(s): 2020 -05-29
Reporting Summary
Nature Research wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Research policies, see our Editorial Policies and the Editorial Policy Checklist.
Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.
A description of all covariates tested
A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient)
AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)
For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.
For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Our web collection on statistics for biologists contains articles on many of the points above.
Software and code
Policy information about availability of computer code
Data collection DeBruine. L.M. (2019, April 9). Experimentum: Beta release 1 (Version v.0.1). Zenodo. http://doi.org/10.5281/zenodo.2634356
Data analysis https://osf.io/87rbg/
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and
reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Research guidelines for submitting code & software for further information.
Data
Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A list of figures that have associated raw data
- A description of any restrictions on data availability
https://osf.io/87rbg/
2
nature research | reporting summary April 2020
Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf
Behavioural & social sciences study design
All studies must disclose on these points even when the disclosure is negative.
Study description Quantitative analysis of ratings data
Research sample Student-type samples collected in eleven different world regions
Sampling strategy Opportunistic sampling by individual research groups
Data collection Rating of faces on one of thirteen randomly determined traits
Timing Throughout 2019
Data exclusions All data exclusions were described in the stage one protocol and in the analysis code
Non-participation These data are given in the analysis code and output
Randomization Participants were randonly allocated to rate one triat
Reporting for specific materials, systems and methods
We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.
Materials & experimental systems
n/a Involved in the study
Antibodies
Eukaryotic cell lines
Palaeontology and archaeology
Animals and other organisms
Human research participants
Clinical data
Dual use research of concern
Methods
n/a Involved in the study
ChIP-seq
Flow cytometry
MRI-based neuroimaging
Human research participants
Policy information about studies involving human research participants
Population characteristics Raters from a range of geographic regions and countries. Region and country information are reported.
Recruitment Opportunistic recruitment by individual labs.
Ethics oversight main ethics approval was from University of Glasgow, although some individual groups also obtained their own thicas
approvals.
Note that full information on the approval of the study protocol must also be provided in the manuscript.
... However, other work suggests that the expectation that synthetic voices possess robotic qualities can also cause listeners to have difficulty when they are asked to classify voices as natural or artificial 12 . In light of these points, Study 1 first investigated whether trait-ratings of the synthetic voices used by conversational agents are underpinned by valence and dominance dimensions similar to those observed for natural human stimuli in previous work 4,5 . ...
... Consistent with previous research using natural human voices and faces as stimuli 4,5,7 , the first PC was highly correlated with pro-social traits, such as trustworthiness, competence, responsibility, emotional stability, and sociableness, but weakly correlated with dominance and aggressiveness, and the second PC was highly correlated with dominance and aggressiveness, but weakly correlated with trustworthiness, competence, responsibility, emotional stability, and sociableness. Following previous research showing this pattern of results, we labelled these PCs Valence and Dominance, respectively. ...
... The second component, which explained substantially less of the variance in ratings, was highly correlated with dominance and aggressiveness ratings and weakly correlated with trustworthiness, competence, responsibility, emotional stability, and sociable ratings. This pattern of results is extremely similar to those obtained when ratings of natural human faces and voices were subject to PCA in previous studies [4][5][6][7] , suggesting that social perceptions of synthetic voices are underpinned by valence and dominance dimensions similar to those previously found to underpin social perceptions of natural human stimuli. www.nature.com/scientificreports/ ...
Article
Full-text available
There is growing concern that artificial intelligence conversational agents (e.g., Siri, Alexa) reinforce voice-based social stereotypes. Because little is known about social perceptions of conversational agents’ voices, we investigated (1) the dimensions that underpin perceptions of these synthetic voices and (2) the role that acoustic parameters play in these perceptions. Study 1 (N = 504) found that perceptions of synthetic voices are underpinned by Valence and Dominance components similar to those previously reported for natural human stimuli and that the Dominance component was strongly and negatively related to voice pitch. Study 2 (N = 160) found that experimentally manipulating pitch in synthetic voices directly influenced dominance-related, but not valence-related, perceptions. Collectively, these results suggest that greater consideration of the role that voice pitch plays in dominance-related perceptions when designing conversational agents may be an effective method for controlling stereotypic perceptions of their voices and the downstream consequences of those perceptions.
... For example, a multi-site study from the Psychological Science Accelerator (PSA) initiative 43 was generally able to replicate the original findings of Oosterhof and Todorov 4 across 11 world regions and 41 countries in ethnically diverse stimuli-including the central role of valence/trustworthiness in social face evaluations. However, model fit for the valence-dominance model differed significantly across world regions, with diminished fit in Asian countries, alluding to the possibility of systematic ethnicity-based perceptual differences 44 . Other studies further lend support to this assumption. ...
... Translations. In line with past translational procedures from large cross-cultural, multi-lab projects 44 , the following translational steps were performed. ...
... Examining RQ2, we found that the ethnic ingroup-outgroup effect,-that is, preferential treatment of ingroup members-was detected across all stimulus types, and ethnicities, evidenced by raters judging ingroup members as more trustworthy than outgroup members. Taken together, we interpret this to support the position (e.g., 40,44,50 ) that ethnicity-based ingroup preference on the basis of trustworthiness judgments is a widespread social phenomenon, evidenced across cultures, and relevant to face perception. Relatedly, the findings of our exploratory analyses suggest ingroup-outgroup effects beyond ethnicity. ...
Article
Full-text available
Social face evaluation is a common and consequential element of everyday life based on the judgement of trustworthiness. However, the particular facial regions that guide such trustworthiness judgements are largely unknown. It is also unclear whether different facial regions are consistently utilized to guide judgments for different ethnic groups, and whether previous exposure to specific ethnicities in one’s social environment has an influence on trustworthiness judgements made from faces or facial regions. This registered report addressed these questions through a global online survey study that recruited Asian, Black, Latino, and White raters (N = 4580). Raters were shown full faces and specific parts of the face for an ethnically diverse, sex-balanced set of 32 targets and rated targets’ trustworthiness. Multilevel modelling showed that in forming trustworthiness judgements, raters relied most strongly on the eyes (with no substantial information loss vis-à-vis full faces). Corroborating ingroup–outgroup effects, raters rated faces and facial parts of targets with whom they shared their ethnicity, sex, or eye color as significantly more trustworthy. Exposure to ethnic groups in raters’ social environment predicted trustworthiness ratings of other ethnic groups in nuanced ways. That is, raters from the ambient ethnic majority provided slightly higher trustworthiness ratings for stimuli of their own ethnicity compared to minority ethnicities. In contrast, raters from an ambient ethnic minority (e.g., immigrants) provided substantially lower trustworthiness ratings for stimuli of the ethnic majority. Taken together, the current study provides a new window into the psychological processes underlying social face evaluation and its cultural generalizability. Protocol registration The stage 1 protocol for this Registered Report was accepted in principle on 7 January 2022. The protocol, as accepted by the journal, can be found at: https://doi.org/10.6084/m9.figshare.18319244.
... Data was collected across 45 countries, divided into 11 world regions (see Table 1). The decision to not combine the UK with Western Europe was made before data collection started since the Psychological Science Accelerator (Moshontz et al., 2018) network could get a minimum number of participants in the UK (for details see Jones et al., 2021a). ...
... Each participant completed the ratings twice and the ratings from the first and second blocks were averaged for all the analyses. Participants took the study in labs or online and data from those who did not rate all 120 faces in the first block, who provided the same rating for 75% or more of the faces, or who did not specify their region were excluded from the analyses (for details see Jones et al., 2021a). ...
Article
Research has found that attractiveness has a positive “halo effect”, where people tend to attribute socially desirable personality traits to physically attractive individuals. Several studies have documented this “attractiveness halo effect”, with most research using western samples. This study sought to examine the “attractiveness halo effect” across 45 countries in 11 world regions. Data was collected through the Psychological Science Accelerator and participants were asked to rate 120 faces on one of several traits. Results showed that attractiveness correlated positively with most of the socially desirable personality traits. More specifically, across all 11 world regions, male and female faces rated as more attractive were rated as more confident, emotionally stable, intelligent, responsible, sociable, and trustworthy. These findings, thus, provide evidence that the “attractiveness halo effect” can be found cross-culturally.
... When is reasonably small and experiments are inexpensive to run, it may be possible to exhaustively explore the space by conducting every experiment in a full factorial design. For example, when = 8, there are 256 experiments in the design space, a number that is beyond the scale of most studies in the social and behavioral sciences but is potentially achievable with recent innovations in crowdsourcing and other "high-throughput" methods, especially if distributed among a consortium of labs (Byers-Heinlein et al., 2020;Jones et al., 2021). Moreover, running all possible experiments may not be necessary: If the goal is to estimate the impact that each variable has, together with their interactions, a random (or more efficient) sample of the experiments can be run (Auspurg & Hinz, 2014). ...
Article
The dominant paradigm of experiments in the social and behavioral sciences views an experiment as a test of a theory, where the theory is assumed to generalize beyond the experiment's specific conditions. According to this view, which Alan Newell once characterized as “playing twenty questions with nature,” theory is advanced one experiment at a time, and the integration of disparate findings is assumed to happen via the scientific publishing process. In this article, we argue that the process of integration is at best inefficient, and at worst it does not, in fact, occur. We further show that the challenge of integration cannot be adequately addressed by recently proposed reforms that focus on the reliability and replicability of individual findings, nor simply by conducting more or larger experiments. Rather, the problem arises from the imprecise nature of social and behavioral theories and, consequently, a lack of commensurability across experiments conducted under different conditions. Therefore, researchers must fundamentally rethink how they design experiments and how the experiments relate to theory. We specifically describe an alternative framework, integrative experiment design, which intrinsically promotes commensurability and continuous integration of knowledge. In this paradigm, researchers explicitly map the design space of possible experiments associated with a given research question, embracing many potentially relevant theories rather than focusing on just one. The researchers then iteratively generate theories and test them with experiments explicitly sampled from the design space, allowing results to be integrated across experiments. Given recent methodological and technological developments, we conclude that this approach is feasible and would generate more-reliable, more-cumulative empirical and theoretical knowledge than the current paradigm—and with far greater efficiency.
... Consistent with a learning account, there are systematic cultural differences in first impressions (Chen, Jing, Lee, & Bai, 2016;Jones et al., 2021;Lakshmi, Wittenbrink, Correll, & Ma, 2021;Over, Eggleston, & Cook, 2020a;Sofer et al., 2017;Sutherland et al., 2018;Walker, Jiang, Vetter, & Sczesny, 2011;Zebrowitz et al., 2012). For example, in so-called WEIRD cultures (Western, Educated, Industrialized, Rich, Democratic), straight white teeth are associated with attractiveness, social status, and a host of other positive characteristics (Dion et al., 1972;Eagly et al., 1991). ...
Article
Full-text available
When we encounter a stranger for the first time, we spontaneously attribute to them a wide variety of character traits based on their facial appearance. There is increasing consensus that learning plays a key role in these first impressions. According to the Trait Inference Mapping (TIM) model, first impressions are the products of mappings between ‘face space’ and ‘trait space’ acquired through domain-general associative processes. Drawing on the associative learning literature, TIM predicts that first-learned associations between facial appearance and character will be particularly influential: they will be difficult to unlearn and will be more likely to generalise to novel contexts than appearance-trait associations acquired subsequently. The study of face-trait learning de novo is complicated by the fact that participants, even young children, already have extensive experience with faces before they enter the lab. This renders the study of first-learned associations from faces intractable. Here, we overcome this problem by using Greebles – a class of novel synthetic objects about which participants had no previous knowledge or preconceptions – as a proxy for faces. In four experiments (total N = 640) with adult participants we adapt classic AB-A and AB-C renewal paradigms to study appearance-trait learning. Our results indicate that appearance-trait associations are subject to contextual control, and are resistant to counter-stereotypical experience.
Article
Full-text available
Social dominance is an important feature of social life. Dominance has been proposed to be one of two trait dimensions underpinning social judgments of human faces. Yet, the neural bases of the ability to identify different dominance levels in others based on intrinsically facial cues remains poorly understood. Here, we used event-related potentials to determine the temporal dynamics of facial dominance evaluation based on facial features signaling physical strength/weakness in humans. Twenty-seven participants performed a dominance perception task where they passively viewed faces with different dominance levels. Dominance levels did not modulate an early component of face processing, known as the N170 component, but did modulate the late positive potential (LPP) component. These findings indicate that participants inferred dominance levels at a late stage of face evaluation. Furthermore, the highest level of dominant faces and the lowest level of submissive faces both elicited higher LPP amplitudes than faces with a neutral dominance level. Taken together, the present study provides new insights regarding the dynamics of the neurocognitive processes underlying facial dominance evaluation.
Article
Full-text available
People differ in their tendency to infer others' personalities and abilities from their faces. An extreme form of such face-based trait inference (FBTI) is problematic because of its unwarranted impact on real-world decision making. Evolutionary perspectives on FBTI suggest that its inter-individual variation would be trait-specific : e.g. those who make extreme face-based inferences about trustworthiness may not necessarily do so about dominance. However, there are several psychological variables that can increase the FBTI extremity across traits. Here, we show that there is a generalized individual tendency to make extreme FBTI across traits, in support of the latter view. We found that the degrees of extremity of face-based inferences about seven traits had high cross-trait correlations, constituting a general factor. This generalized FBTI extremity had good test–retest reliability and was neither an artefact of extreme nor socially desirable response biases. Moreover, it was positively associated with facial emotion recognition ability and tendencies to believe physiognomy and endorse stereotypes. Our results demonstrate that there are individuals who have a temporally stable disposition to draw extreme conclusions about various traits of others from facial appearance as well as their psychological characteristics.
Article
Faces vary from image to image, eliciting different judgments of traits and often different judgments of identity. Knowledge that two face images belong to the same person facilitates the processing of identity information across images, but it is unclear if this also applies to trait judgments. In this preregistered study, participants (N = 100) rated the same 340 face images on perceived trustworthiness, dominance, or attractiveness presented in randomised order and again later presented in sets consisting of the same identity. We also explored the role of implicit person theory beliefs in the variability of social judgements across images. We found that judgements of trustworthiness varied less when images were presented in sets consisting of the same identity than in randomised order and were more consistent for images presented later in a set than those presented earlier. However, knowledge of identity had little effect on perceptions of dominance and attractiveness. Finally, implicit person theory beliefs were not associated with variability in social judgements and did not account for effects of knowledge of identity. Our findings suggest that knowledge of identity and perceptual familiarity stabilises judgements of trustworthiness, but not perceptions of dominance and attractiveness.
Article
Remembering of episodic memory is characterized by autonoetic consciousness, which enables us to mentally re-experience the past events. It means that the system of episodic memory enables us to mentally travel into the temporally passed event. The orientation of mental time travel is not only for the past events, also for the future or counterfactual events. And then, the memory system could be interpreted as a system to recombine episodic details and construct events in various time frames. We introduce some research which is related to this memory system; episodic future thinking, details in autobiographical memory, cognitive offloading, intentional cognitive control and forgetting, and computational study on subjective meta-memory. Finally, we discuss the recent perspective of episodic memory or episodic sciences and future research directions. Key words:
Article
Full-text available
First impressions of a person, including social judgements, are often based on appearance. The widely accepted valence-dominance model of face perception (Oosterhof and Todorov 2008 Proc. Natl Acad. Sci. USA 105 , 11 087–11 092 ( doi:10.1073/pnas.0805664105 )) posits that social judgements of faces fall along two orthogonal dimensions: trustworthiness (valence) and dominance. The current study aimed to establish the principal components of social judgements based on the perception of bodies, hypothesizing that these would follow the same dimensions as face perception. Stimuli were black and white photographs showing bodies dressed in grey clothing, standing in their natural posture, in left profile. Raters ( N = 237) judged the stimuli on the 14 traits used in Oosterhof and Todorov's original study (Oosterhof and Todorov 2008 Proc. Natl Acad. Sci. USA 105 , 11 087–11 092 ( doi:10.1073/pnas.0805664105 )). Data were analysed using principal component analysis (PCA), as in the original study, with an additional exploratory factor analysis (EFA) using oblique rotation. While PCA produced a third dimension in line with several replications of the original study, results from the EFA produced two dimensions, representing trustworthiness and dominance, providing support for the hypothesis that social perceptions of bodies can be summarized using the valence-dominance model. These two factors could represent universal perceptions we have about people. Future research could explore social judgements of humans based on other stimuli, such as voices or body odour, to evaluate whether the trustworthiness and dominance dimensions are consistent across modalities.
Article
Full-text available
Despite warnings not to “judge a book by its cover”, people rapidly form facial impressions. In Oosterhof and Todorov’s (2008) two-dimensional model of facial impressions, trustworthiness and dominance underlie impressions and primarily function to signal the potential threat of others. Here, we test a key assumption of these models, namely that these dimensions are functional, by evaluating whether the adult-face dimensions apply to young children’s faces. Although it may be functional for adults to judge adult faces on dimensions that signal threat, adults associate different social goals with children, and these goals are likely to impact the impressions adults make of such faces. Thus, a functional approach would predict that the dimensions for children’s faces are not threat focused. In Studies 1-2, we build a data-driven model of Caucasian adults’ impressions of Caucasian children’s faces, finding evidence for two dimensions. The first dimension, niceness, is similar (although not identical) to the adult dimension of trustworthiness. However, we find a second dimension, shyness, that is clearly dissociable from dominance (Study 3), and critically, is not focused on threat. We demonstrate that adults are sensitive to subtle facial manipulations of these dimensions (Studies 4-5) and that these impressions impact adults’ behavioral expectations of children (Study 6). Finally, we show that niceness and shyness dimensions generalize to an independent sample of ambient images, demonstrating their robustness (Study 7). Our results suggest that social goals have the power to drive functional impressions and highlight the flexibility of our visual system when forming such inferences.
Article
Full-text available
Social judgments of faces made by Western participants are thought to be underpinned by two dimensions: valence and dominance. Because some research suggests that Western and Eastern participants process faces differently, the two-dimensional model of face evaluation may not necessarily apply to judgments of faces by Eastern participants. Here we used a data-driven approach to investigate the components underlying social judgments of Chinese faces by Chinese participants. Analyses showed that social judgments of Chinese faces by Chinese participants are partly underpinned by a general approachability dimension similar to the valence dimension previously found to underpin Western participants’ evaluations of White faces. However, we found that a general capability dimension, rather than a dominance dimension, contributed to Chinese participants’ evaluations of Chinese faces. Thus, our findings present evidence for both cultural similarities and cultural differences in social evaluations of faces. Importantly, the dimension that explained most of the variance in Chinese participants’ social judgments of faces was strikingly similar to the valence dimension previously reported for Western participants.
Article
Full-text available
Concerns about the veracity of psychological research have been growing. Many findings in psychological science are based on studies with insufficient statistical power and nonrepresentative samples, or may otherwise be limited to specific, ungeneralizable settings or populations. Crowdsourced research, a type of large-scale collaboration in which one or more research projects are conducted across multiple lab sites, offers a pragmatic solution to these and other current methodological challenges. The Psychological Science Accelerator (PSA) is a distributed network of laboratories designed to enable and support crowdsourced research projects. These projects can focus on novel research questions or replicate prior research in large, diverse samples. The PSA’s mission is to accelerate the accumulation of reliable and generalizable evidence in psychological science. Here, we describe the background, structure, principles, procedures, benefits, and challenges of the PSA. In contrast to other crowdsourced research networks, the PSA is ongoing (as opposed to time limited), efficient (in that structures and principles are reused for different projects), decentralized, diverse (in both subjects and researchers), and inclusive (of proposals, contributions, and other relevant input from anyone inside or outside the network). The PSA and other approaches to crowdsourced psychological science will advance understanding of mental processes and behaviors by enabling rigorous research and systematic examination of its generalizability.
Article
Full-text available
Impressions of competence from faces predict important real-world outcomes, including electoral success and CEO selection. Presumed competence is associated with social status. Here, we show that subtle economic status cues in clothes affect perceived competence from faces. In nine studies, people rated the competence of faces presented in frontal headshots. Faces were shown with different upper-body clothing rated by independent judges as looking “richer” or “poorer”, although not notably perceived as such when explicitly described. The same face when seen with “richer” clothes was judged significantly more competent than with “poorer” clothes. The effect persists even when perceivers are exposed to the stimuli briefly (129 ms), warned that clothing cues are non-informative, and instructed to ignore the clothes (in one study, with considerable incentives). These findings demonstrate the uncontrollable effect of economic status cues on person perception. They add yet another hurdle to the challenges faced by low-status individuals.
Article
Full-text available
Culture is a human universal, yet it is a source of variation in human psychology, behaviour and development. Developmental researchers are now expanding the geographical scope of research to include populations beyond relatively wealthy Western communities. However, culture and context still play a secondary role in the theoretical grounding of developmental psychology research, far too often. In this paper, we highlight four false assumptions that are common in psychology, and that detract from the quality of both standard and cross-cultural research in development. These assumptions are: (i) the universality assumption , that empirical uniformity is evidence for universality, while any variation is evidence for culturally derived variation; (ii) the Western centrality assumption , that Western populations represent a normal and/or healthy standard against which development in all societies can be compared; (iii) the deficit assumption , that population-level differences in developmental timing or outcomes are necessarily due to something lacking among non-Western populations; and (iv) the equivalency assumption , that using identical research methods will necessarily produce equivalent and externally valid data, across disparate cultural contexts. For each assumption, we draw on cultural evolutionary theory to critique and replace the assumption with a theoretically grounded approach to culture in development. We support these suggestions with positive examples drawn from research in development. Finally, we conclude with a call for researchers to take reasonable steps towards more fully incorporating culture and context into studies of development, by expanding their participant pools in strategic ways. This will lead to a more inclusive and therefore more accurate description of human development. This article is part of the theme issue ‘Bridging cultural gaps: interdisciplinary studies in human cultural evolution’.
Article
Influential facial impression models have repeatedly shown that trustworthiness, youthful–attractiveness, and dominance dimensions subserve a wide variety of first impressions formed from strangers’ faces, suggestive of a shared social reality. However, these models are built from impressions aggregated across observers. Critically, recent work has now shown substantial inter‐observer differences in facial impressions, raising the important question of whether these dimensional models based on aggregated group data are meaningful at the individual observer level. We addressed this question with a novel case series approach, using factor analyses of ratings of twelve different traits to build individual models of facial impressions for different observers. Strikingly, three dimensions of trustworthiness, youthful/attractiveness, and competence/dominance appeared across the majority of these individual observer models, demonstrating that the dimensional approach is indeed meaningful at the individual level. Nonetheless, we also found differences in the stability of the competence/dominance dimension across observers. Taken together, results suggest that individual differences in impressions arise in the context of a largely common structure that supports a shared social reality.
Article
Social impressions arise from characteristics of both perceivers and targets. However, empirical research in the domain of impression formation has yet to quantify the extent to which perceiver and target characteristics uniquely contribute to impressions across group boundaries (e.g., race, gender). To what extent does an impression arise from “our mind” versus “a target’s face”, and does this process differ for impressions across race and gender? We explored this question by estimating intraclass correlation coefficients (ICCs) from cross-classified multilevel models of 188,472 face ratings from 2,230 participants (Study 1) and 219,658 ratings from 2,984 participants (Study 2). We partitioned the total variance in ratings on a trait dimension (trustworthiness, dominance, youthful/attractiveness) into variance explained by perceivers versus targets, and compared these ICCs among different groups (e.g., ratings of own- vs. other-group targets). Overarching results reveal (a) target appearance matters more for women than men, (b) target appearance matters more for impressions on youthful/attractiveness than trustworthiness or dominance dimensions, (c) differences in perceiver/target influences across race did not consistently replicate, and (d) these differences are absent in minimal groups, supporting the role of racial and gender stereotypes in driving these effects. Overall, perceiver characteristics contribute more to impressions than target appearance. Our findings disentangle the contributions of perceiver and targets to impressions and illustrate that the process of impression formation is not equal across various group boundaries.
Article
Significance Current theory of face-based trait impressions focuses on their foundation in facial morphology, from which emerges a correlation structure of face impressions due to shared feature dependence, “face trait space.” Here, we proposed that perceivers’ lay conceptual beliefs about how personality traits correlate structure their face impressions. We demonstrate that “conceptual trait space” explains a substantial portion of variance in face trait space. Further, we find that perceivers who believe any set of personality traits (e.g., trustworthiness, intelligence) is more correlated in others use more similar facial features when making impressions of those traits. These findings suggest lay conceptual beliefs about personality play a crucial role in face-based trait impressions and may underlie both their similarities and differences across perceivers.
Article
Common factor analysis (FA) and principal component analysis (PCA) are commonly used to obtain lower-dimensional representations of matrices of correlations among manifest variables. Whereas some experts argue that differences in results from use of FA and PCA are small and relatively unimportant in empirical studies, the fundamental rationales for the two methods are very different. Here, FA and PCA are contrasted on four key issues: the range of possible dimensional loadings, the range of potential correlations among dimensions, the structure of residual covariances and correlations, and the relation between population parameters and the correlational structures with which they are associated. For decades, experts have emphasized indeterminacies of the FA model, particularly indeterminacy of common factor scores. Determinate in most respects, a heretofore unacknowledged, pernicious indeterminacy of PCA is demonstrated: the indeterminacy between PCA structural representations and the correlational structures from which they are derived. Researchers are often advised to use either FA or PCA in exploratory rounds of data analysis to understand and refine the dimensional structure of a domain before moving to Structural Equation Modeling in later theory-testing, confirmatory, replication studies. Results from the current study suggest that PCA is an unreliable method to use for such purposes and may lead to serious misrepresentation of the structure of a domain. Hence, PCA should never be used if the goal is to understand and represent the latent structure of a domain; only FA techniques should be used for this purpose, as only FA provides reliable structural representations as the basis for confirmatory tests in future studies.
Article
Facial appearance evokes robust impressions of other people's personality traits. Recent research suggests that the trait space arising from face-based impressions shifts due to context and social cognitive factors. We suggest a novel framework in which multiple bottom-up and top-down processes mutually determine a dynamic rather than fixed trait space.