You are on page 1of 5

BR A IN RE S E A RCH 1 3 11 ( 20 1 0 ) 8 1 8 5

available at

Research Report

Perception of intentions and actions: Gender

stereotype susceptibility
Marina A. Pavlova a,b,, Matthias Wecker a,b , Kerstin Krombholz a,b , Arseny A. Sokolov c

Department of Pediatric Neurology and Child Development, Children's Hospital, Eberhard Karls University of Tbingen, Tbingen, Germany
Institute of Medical Psychology and Behavioral Neurobiology, MEG Centre, Eberhard Karls University of Tbingen, Tbingen, Germany
Department of Neurosurgery, Eberhard Karls University of Tbingen, Tbingen, Germany



Article history:

Gender differences are evident in the comprehension of social signals, but the underlying

Accepted 19 November 2009

basis for these differences is unclear. There is some indication that gender effects have

Available online 27 November 2009

neurobiological sources. Here we manipulated stereotype messages about gender

differences in a social cognition task, on which no gender gap has previously been


documented. The outcome indicates that manipulation of stereotype messages elicits

Perception of intentions and actions

gender effects. A positive message enhances performance, whereas a negative message


diminishes it. Furthermore, this effect is more pronounced in females, with a greater force of


a negative stereotype message. The study provides novel insights into the possible sources

Visual social cognition

of gender related fluctuations in social cognition. The findings are discussed in terms of

Brain mechanisms

behavioral components and brain mechanisms underpinning gender effects in social

2009 Elsevier B.V. All rights reserved.



Gender differences are well known in performance on a

variety of cognitive tasks. In general, males outperform on
tasks involving spatial and mathematical reasoning, whereas
females excel in language and episodic memory (Andreano
and Cahill, 2009; Hamilton, 2008). Recent data indicate that not
just visual spatial abilities of navigation and mental rotation
are impacted by gender, but also some aspects of social
perception are gender dependent. Perception of intentions
and actions is of extreme importance for a variety of daily life
situations and adaptive social behavior. This ability constitutes a central component of social competence. In accordance
with wide spread wisdom, women are reported to be more
proficient in recognition of facial emotions (Montagne et al.,
2005) and intentional expressions (discrimination between

friendliness and sexual interest) in still photographs (Farris

et al., 2008). As a rule, however, facial expressions can only
signal emotional states and affect, but do not provide
information on how to deal with it (De Gelder, 2006).
Expressions of the whole body, body language, gestures and
actions of others are much richer source of information for
social interaction. Gender influence on social perception of
actions and intentions is, however, largely unknown.
There is some evidence that gender effects have neurobiological sources (e.g., Bourne, 2005; Cahill, 2006). On the other
hand, gender differences in performance can be elicited by
experiential factors and stereotypes (Wraga et al., 2006). A
positive (though false) stereotype message enhances performance of women on the mental rotation task, on which
females are consistently reported to be least successful than
males. However, performance of women on mathematical

Corresponding author. Social and Cognitive Developmental Neuroscience Unit, Department of Pediatric Neurology and Child
Development, Children's Hospital, Hoppe-Seyler-Str. 1, 72076 Tbingen, Germany. Fax: +49 7071 295253.
E-mail address: (M.A. Pavlova).
0006-8993/$ see front matter 2009 Elsevier B.V. All rights reserved.


BR A IN RE S EA RCH 1 3 11 ( 20 1 0 ) 8 1 85

tests is hindered when they are reminded of gender differences in mathematical abilities (Cadinu et al., 2003, 2005).
Furthermore, gender-specific stereotype messages may not
only enhance or diminish performance level, but can also
substantially modulate activation of the underpinning brain
networks (Wraga et al., 2007).
Here we address the issue of whether, and, if so, how
manipulation of information about gender differences in
social cognition affects performance. The event arrangement
(EA) task was administered to female and male participants.
For this task, participants have to organize a set of cards
depicting an event as a series of snapshots in a comic-strip
fashion. It is assumed that good performance on such a
task requires understanding the characters' mental states
(Baron-Cohen et al., 1986; Pavlova et al., 2008; Vllm et al.,
2006). For successful performance, participants need to reflect
the core of the story, which is often based on veridical
perception of intentions and dispositions of the characters
involved in this particular event. Previous work indicated that
there is the lack of gender differences in performance on this
task (Pavlova, 2009). Here we manipulated gender stereotype
messages, informing one group of participants prior to testing
that usually females over-perform on the EA task, whereas the
other group has been provided with information that males
have an edge in performance. We asked whether and, if so,
how these gender stereotype messages affect performance.



Individual scores on the task were submitted to a 2 3 ANOVA

with factors Gender (female/male) and Information (neutral/
positive for females/positive for males). This analysis reveals a
highly significant interaction between the factors (F(2,80) =
7.89, p < 0.001). As expected from earlier work (Pavlova, 2009),
no gender differences in performance of the group without
any gender-specific information have been revealed by post
hoc pair-wise comparisons (t(25) = 0.33, two-tailed, p = 0.74, n.
s.; average 9.44 2.46 and 10.14 3.3, for females and males,
respectively). The information that females are usually better
on the task leads to significant differences in performance
within the group, enhancing performance of females and
diminishing performance of males (t(24) = 2.27, one-tailed,
p < 0.03; average 10.7 1.07 and 9.58 1.73, for females and
males, respectively). In turn, the information that males are
usually better leads to highly significant differences in
performance, enhancing performance of males and substantially deteriorating performance of females (t(28) = 4.82, onetailed, p < 0.0001; average 7.18 2.09 and 10.68 2.19, for
females and males, respectively).
Fig. 1 shows mean scores for females and males of each
group. Inspection of this figure suggests that gender related
information more strongly affects performance of females. To
prove this assumption, a 1 3 ANOVA with factor Information
(neutral/positive for females/positive for males) was separately conducted for females and males. For females, the effect of
information was highly significant (F(2,39) = 13.1, p < 0.0001),
whereas for males it was not (F(2,38) = 0.65, p = 0.53). In other
words, females demonstrate substantial fluctuations in performance depending on the gender stereotype message.

Fig. 1 Mean scores on the EA task. Female (circles) and male

(triangles) participants in the groups with different
information prior to testing: (A) neutral information, (B)
explicit positive gender stereotype information for females
(INFO 1: Females are usually better on this task) and (C)
explicit positive gender stereotype information for males
(INFO 2: Males are usually better on this task). Vertical bars
represent SE.

Inspection of Fig. 1 raises the possibility that positive and

negative stereotype information has different impact on
performance of females and males. Moreover, in females,
the impact of positive and negative stereotype messages
substantially differs in its magnitude. A 2 2 ANOVA with
factors Gender (female/male) and Information (positive/negative) revealed significant main effects of Gender (F(1;54) = 5.7,
p < 0.02) as well as Information (F(1,54) = 21,6, p < 0.0001). The
interaction of factors was also significant (F(1,54) = 5.9, p < 0.02).
Under the influence of an explicit positive stereotype message,
there was no difference in performance of females and males
(t(27) = 0.71, one-tailed, p = 0.48, n.s.), whereas an implicit
negative information causes a dramatic decrease in performance of women as compared with men (t(25) = 3.59, onetailed, p < 0.001).



The outcome of the study indicates that gender stereotype

messages affects perception and understanding of intentions
and actions. A positive message enhances performance on the
event arrangement task, whereas a negative message causes a
decrease in performance. Furthermore, the impact of stereotype messages about gender differences on social perception is
gender dependent. The effect is more pronounced in females
with a greater impact of a negative stereotype message.
To the best of our knowledge, for the first time, we
demonstrated gender-specific effects of stereotype message
in social cognition. Furthermore, as compared with previous
work (e.g., Cadinu et al., 2003, 2005; Wraga et al., 2006, 2007),
the effect of stereotype message occurred on the task on
which no gender differences have been documented in the
absence of any gender stereotype information. Therefore, we
demonstrated pure effects of both positive and negative
gender stereotype information on performance.

BR A IN RE S E A RCH 1 3 11 ( 20 1 0 ) 8 1 8 5

The sharp decrease in performance of women under the

influence of negative stereotype information may be explained
by stereotype threat, the fear of confirming a negative
stereotype about a group to which one belongs, that deteriorates performance on cognitive tests (Cadinu et al., 2003, 2005;
Kit et al., 2008; Schmader et al., 2008; Steele, 1997). Stereotype
threat is most robust and pronounced in minority groups or in
the elderly. However, this effect also occurs in Caucasian men
when they are informed, for instance, of Asian's superiority in
math or of under-performance of males on affective tasks.
Furthermore, the stereotype threat effect might also occur in
patients with neurological injury (Kit et al., 2008; Suhr and
Gunstad, 2005). Individuals who are highly identified with their
minority groups are most vulnerable to the stereotype threat. In
other words, women who possess high identification with their
gender are stronger affected by negative gender stereotype.
The present data indicate that a negative stereotype
message leads to a sharp decrease in performance of women,
even if this stereotype is implicit: females were informed that
males are usually better, not that women are usually worse.
The observed effects appear to be even more impressive because they occur in a group of medical students, i.e. in young
adults with a relatively high educational level, socio-economic
status, and high motivation for achievements, ambitions and
expectations for future professional career. The findings
suggest that even in this population, in females, it is easier to
deteriorate cognitive performance via experiential and situational factors than to improve it. We assume, therefore, that
susceptibility of females to a negative stereotype message may
be, at least, partly, responsible for under-representation of
females in leading positions, especially, in domains where
negative gender stereotypes are widely spread. It is argued that
stereotype threat disrupts performance via different interrelated mechanisms such as a physiological stress response that
impairs prefrontal processing and is mediated largely by interaction with sex hormones, a tendency to actively monitor performance and impaired divided attention, efforts to suppress
negative thoughts and emotions that can lead to reduced working memory and other cognitive capacities, and lower expectations and motivation (Kit et al., 2008; Schmader et al., 2008).
The present study addresses behavioral component of the
gender stereotype impact on visual social cognition. Clarification of the brain mechanisms underpinning effects of
gender stereotype messages requires further examination. In
patients with ventromedial prefrontal cortical lesions, for
example, the impact of implicit stereotype messages is less
strong than in healthy controls (Milner and Grafman, 2001).
Recent functional magnetic resonance imaging (fMRI) data
shows that in females, gender stereotype messages may not
only affect performance on a mental rotation of the self task,
but also can modulate activation of the engaged brain network
(Wraga et al., 2007). Under the negative stereotype message,
activation increases in brain areas associated with emotional
load such as the amygdala. Improved performance in the
positive stereotype group is associated with increased activation in task-specific brain areas, which are involved in visual
processing and working memory.
It remains unclear whether stereotype messages affect
social perception of actions represented in dynamic displays.
Neural mechanisms of event reconstruction from static


images might differ from the visual perception of real or

filmed actions (Kits et al., 2003). However, the neural
representations of actual body movements in the temporal
lobe, which is considered of substantial importance for social
perception, also extend to body actions implied from still
photographs (Jellema and Perrett, 2003; Barraclough et al.,
2006; Lorteije et al., 2006). It is assumed that dynamic displays
possess greater ecological validity as compared with static
images, and, therefore, provide more direct and reliable
information for social perception (Kits et al., 2003; Pavlova et
al., 2005; Pelphrey et al., 2003, 2007).
Future research should shed light on gender related
vulnerability to neuropsychiatric disorders such as autistic
spectrum disorders (ASD) related to impairments in nonverbal social cognition. Males are more commonly affected by
ASD than females with a ratio of about 4:1 (Newschaffer et al.,
2007). Females, however, are affected much more severely,
and therefore in high-functioning autistic individuals, this
ratio is even much higher. The stereotype threat might affect
neuropsychological test performance in neurological and
psychiatric population (Corrigan and Holzman, 2001; Kit et
al. (2008); Suhr and Gunstad, 2005).


Experimental procedures



Eighty-three adults, students of the University of Tbingen

Medical School (aged 2036 years), were enrolled in the study
for a course credit. They were assigned to one of the three
experimental groups. One group consisted of 27 participants
(13 female, age range 2134 years, mean = 22.7 years, SD = 3.7,
and 14 male, age range 2127 years, mean = 21.7 years, SD = 1.7).
Second group included 26 participants (14 female, age range
2134 years, mean = 25.5 years, SD = 3.6, and 12 male, age range
2236 years, mean = 23.8 years, SD = 1.9). Third group involved
30 participants (15 female, age range 2027, mean = 22.1 years,
SD = 1.7, and 15 males, age range 2128, mean = 23.2 years,
SD = 2.1). All participants had normal or corrected-to-normal
vision. None had a history of neurological or psychiatric
disorders including ASD. None had previous experience with
such tasks. They were run individually.


Task and procedure

We used the EA task included in the Wechsler Intelligenztest

fr Erwachsene (WIE), a battery based on the Wechsler Adult
Intelligence Scale (WAIS-III) by David Wechsler adapted to the
German population (Von Aster et al., 2006; Fig. 2). Being a part
of the Wechsler Intelligence test battery, the EA task is a wellestablished tool for psychological assessment, psychometrically standardized, and provides with normative scores
obtained in a large healthy population. For this task, several
sets of cards are presented to participants. These sets portray
human characters, their actions, intentions, and dispositions.
The sets of cards differ in their complexity ranging in number
of cards from 3 to 6. Each set is presented in a predetermined
scrambled (false) order, which is the same for all participants.
The participant has to rearrange cards into a predetermined


BR A IN RE S EA RCH 1 3 11 ( 20 1 0 ) 8 1 85

Fig. 2 Examples of the visual EA tasks administered to participants. (A) Task number 5 and (B) task number 8. First string
represents scrambled (false) order of cards as presented to participants, whereas second string is the correct order of cards
representing an event in a comic-strip fashion.

correct sequence depicting an event in a comic-strip fashion,

thereby showing understanding of the event represented in
the pictures (Fig. 2).
Participants were presented with 11 sets of black-and-white
cards. Both accuracy (correct order of cards in a sequence) and
time needed for an event arrangement (as a specific time limit
for each set ranging, according to the event complexity, from
30 to 120 s) are taken into account when assessing performance
on the task. Each participant was required to start immediately
once a set of cards was presented. They were also told that
each set has a specific time limit for its rearrangement. For
each set, the number of errors corresponds to specific raw
scores given in the WIE (von Aster et al., 2006). According to
tables found in the WIE Manual that take into account age of
participant, raw values (sum of scores with a maximum of 22,
resulting from summing up the highest possible scores for
each sequence, namely, 2 11) are then transformed into the
standardized normative scores ranging from 1 (floor performance) through 10 (normal performance for this age) to 19
(extraordinary or ceiling performance for this age).
Participants were not informed that the task was aimed at
investigation of social cognition. The first group of participants
did not receive any information concerning gender influence

on performance on the EA task. The second group was told that

females usually perform better on this task (an explicit positive
stereotype message for females, and an implicit negative
stereotype message for males), and the third group was
provided with information that males usually perform better
on the EA task (an explicit positive stereotype message for
males, and an implicit negative message for females).

We are grateful to Lea-Sophie Gartlgruber, Daniela Maurer,
Thomas Savigny and Matthias S. Zerer for assistance with
testing, Boris Kotchoubey and Alexander N. Sokolov for fruitful
discussions and valuable comments. Paper preparation was
supported by the Else Krner-Fresenius Foundation (Grant
P63/2008 to MP).


Andreano, J.M., Cahill, L., 2009. Sex influences on the neurobiology

of learning and memory. Lern. Mem. 16, 248266.

BR A IN RE S E A RCH 1 3 11 ( 20 1 0 ) 8 1 8 5

Baron-Cohen, S., Leslie, A.M., Frith, U., 1986. Mechanical,

behavioral and intentional understanding of picture stories in
autistic children. Br. J. Dev. Psychol. 4, 113115.
Barraclough, N.E., Xiao, D., Oram, M.W., Perrett, D.I., 2006. The
sensitivity of primate STS neurons to walking sequences and to
the degree of articulation in static images. Prog. Brain Res. 154,
Bourne, V.J., 2005. Lateralised processing of positive facial
emotion: sex differences in strength of hemispheric
dominance. Neuropsychologia 43, 953956.
Cadinu, M., Maas, A., Frigerio, S., Impagliazo, L., Latinotti, S., 2003.
Stereotype threat: the effect of expectancy on performance.
Eur. J. Soc. Psychol. 33, 267285.
Cadinu, M., Maas, A., Rosabianca, A., Kiessner, J., 2005. Why do
women underperform under stereotype threat? Evidence for
the role of negative thinking. Psychol. Sci. 16, 572578.
Cahill, L., 2006. Why sex matters for neuroscience? Nat. Rev.
Neurosci. 7, 477484.
Corrigan, P.W., Holzman, K.L., 2001. Do stereotype threats
influence social cognitive deficits in schizophrenia? In:
Corrigan, P.W., Penn, D.L. (Eds.), Social Cognition and
Schizophrenia. American Psychological Association,
Washington, DC, US, pp. 175192.
De Gelder, B., 2006. Towards the neurobiology of emotional body
language. Nat. Rev. Neurosci. 7, 242249.
Farris, C., Treat, T.A., Vilken, R.J., McFall, R.M., 2008. Perceptual
mechanisms that characterize gender differences in decoding
women's sexual intent. Psychol. Sci. 19, 348354.
Hamilton, C., 2008. Cognition and Sex Differences. Palgrave
Macmillan, New York.
Jellema, T., Perrett, D.I., 2003. Cells in monkey STS responsive to
articulated body motions and consequent static posture:
a case of implied motion? Neuropsychologia 41, 17281737.
Kit, K.A., Tuokko, H.A., Mateer, C.A., 2008. A review of the
stereotype threat literature and its application in neurological
patients. Neuropsychol. Rev. 18, 132148.
Kits, C.D., Egan, G., Gideon, D.A., Ely, T.D., Hoffman, J.M., 2003.
Dissociable neural pathways are involved in the recognition of
emotion in static and dynamic facial expressions. NeuroImage
18, 156168.
Lorteije, J.A., Kenemans, J.L., Jellema, T., van der Lubbe, R.H.,
de Heer, F., van Wezel, R.J., 2006. Delayed response to animate
implied motion in human motion processing areas. J. Cogn.
Neurosci. 18, 158168.
Milne, E., Grafman, J., 2001. Ventromedial prefrontal cortex lesions
in humans eliminate implicit gender stereotyping. J. Neurosci.
21 RC 150 (1-6).


Montagne, B., Kessels, R.P., Frigerio, E., de Haan, E.H., Perrett, D.I.,
2005. Sex differences in perception of affective facial expressions: do men really lack emotional sensitivity? Cogn. Process.
6, 136141.
Newschaffer, C.J., Croen, L.A., Daniels, J., Giarelli, E., Grether, J.K.,
Levy, S.E., Mandell, D.S., Miller, L.A., Pinto-Martin, J., Reaven, J.,
Reynolds, A.M., Rice, C.E., Schendel, D., Windham, G.C., 2007.
The epidemiology of autism spectrum disorders. Annu. Rev.
Public Health 28, 235258.
Pavlova, M., 2009. Perception and understanding of intentions and
actions: does gender matter? Neurosci. Lett. 499, 133136.
Pavlova, M., Sokolov, A.A., Sokolov, A., 2005. Perceived dynamics of
static images enables emotional attribution. Perception 34,
Pavlova, M., Sokolov, A.N., Birbaumer, N., Krgeloh-Mann, I., 2008.
Perception and understanding of others' actions and brain
connectivity. J. Cogn. Neurosci. 20, 494504.
Pelphrey, K.A., Mitchell, T.V., McKeown, M.J., Goldstein, J., Allison,
T., McCarthy, G., 2003. Brain activity evoked by the perception
of human walking: controlling for meaningful coherent
motion. J. Cogn. Neurosci. 23, 68196825.
Pelphrey, K.A., Morris, J.P., McCarthy, G., Labar, K.S., 2007.
Perception of dynamic changes in facial affect and identity in
autism. Soc. Cogn. Affect. Neurosci. 2, 140149.
Schmader, T., Johns, M., Forbes, C., 2008. An integrated process
model of stereotype threat effects on performance. Psychol.
Rev. 115, 336356.
Steele, C.M., 1997. A threat in the air: how stereotypes shape
intellectual identity and performance. Am. Psychol. 52,
Suhr, J.A., Gunstad, J., 2005. Further exploration of the effect of
diagnosis threat on cognitive performance in individuals
with mild head injury. J. Int. Neuropsychol. Soc. 11, 2329.
Vllm, B.A., Taylor, A.N.W., Richardson, P., Corcoran, R., Stirling, J.,
McKie, S., Deakin, J.F.W., Elliot, R., 2006. Neuronal correlates of
theory of mind and empathy: a functional magnetic resonance
imaging study in a nonverbal task. NeuroImage 29, 9098.
Von Aster, M., Neubauer, A., Horn, R., 2006.
Wechsler-Intelligenztest fr Erwachsene WIE. Manual.
bersetzung und Adaptation der WAIS-III von David Wechsler.
The Psychological Corporation, USA.
Wraga, M., Duncan, L., Jacobs, E.C., Helt, M., Church, J., 2006.
Stereotype susceptibility narrows the gender gap in imagined
self-rotation performance. Psychon. Bull. Rev. 13, 813819.
Wraga, M., Helt, M., Jacobs, E., Sullivan, K., 2007. Neural basis of
stereotype-induced shifts in women's mental rotation
performance. Soc. Cogn. Affect. Neurosci. 2, 1219.