What is a specialist? Effects of the male concept of a successful academic person on the performance in a thinking task

Abstract

For many persons the prototypical successful academic person is male. In three studies, the effects of this belief on types of solutions to a thinking task will be shown. In Study 1 (N=144), it is demonstrated that people have major difficulties finding a solution that refutes this belief. In Study 2 (N=21), which uses an equivalent formulation of the thinking problem whose solution corresponds to the male-associated successful academic person, the level of difficulty sinks dramatically. Participants from Study 3 (N=200), who were primed with a story of either an academically successful woman or man before receiving the thinking task were female and male university students from Germany and Israel in a counterbalanced design. The results show that a successful solution requires a restructuring of the thinking task that is based on the availability and accessibility of alternatives. In the discussion, the effects of the male-dominant interpretation of an academically successful person on Israeli and German women and men will be dealt with and starting points for change will be dis-cussed.

Full text

Psychology Science, Volume 46, 2004 (4), p. 514 - 530
What is a Specialist? Effects of the Male Concept of a Successful Academic Person
on Performance in a Thinking Task1
HEIDRUN STOEGER2, ALBERT ZIEGLER3, HANNA DAVID4
Abstract
For many persons the prototypical successful academic person is male. In three studies,
the effects of this belief on types of solutions to a thinking task will be shown. In Study 1
(N=144), it is demonstrated that people have major difficulties finding a solution that refutes
this belief. In Study 2 (N=21), which uses an equivalent formulation of the thinking problem
whose solution corresponds to the male-associated successful academic person, the level of
difficulty sinks dramatically. Participants from Study 3 (N=200), who were primed with a
story of either an academically successful woman or man before receiving the thinking task
were female and male university students from Germany and Israel in a counterbalanced
design. The results show that a successful solution requires a restructuring of the thinking
task that is based on the availability and accessibility of alternatives. In the discussion, the
effects of the male-dominant interpretation of an academically successful person on Israeli
and German women and men will be dealt with and starting points for change will be dis-
cussed.
Key words: thinking task – gender stereotypes – priming – cross-cultural study
1 The authors would like to thank Alice Boetsch, Janet Townsend and Thomas Raul for the translation of
this text, Cornelia Graesel for her help in collecting data and Steffani Knoedler and Julia Opitsch for their
comments on an earlier version.
2 Dr. Heidrun Stoeger, Department of Educational Psychology, University of Ulm, Robert-Koch-Str. 2,
89081 Ulm, Germany; E-mail: heidrun.stoeger @sem-paedagogik.uni-ulm.de
3 Prof. Dr. Dr. Albert Ziegler, Department of Educational Psychology, University of Ulm, Robert-Koch-Str.
2, 89081 Ulm, Germany; E-mail: albert.ziegler@sem-paedagogik.uni-ulm.de
4 Dr. Hanna David, The Katz Institute for Hebrew Literature, Tel Aviv University, Ramat Aviv 69978,
Israel; E-mail: hannadav@post.tau.ac.il

What is a specialist? 515
Introduction
Suppose that a friend tells you the following story:
A father and his son driving together in their car have a terrible car accident. The father
dies upon impact. The son is rushed to the hospital in an ambulance and is immediately
brought to the operating table. The doctor takes a quick look at him and says that a specialist
is needed. The specialist comes, looks at the young man on the operating table and proclaims,
I cannot operate on him, he is my son.
Many listeners would think that this friend is a bad storyteller because the story obviously
doesn't make any sense. However, there is a simple solution: the specialist is female and is
the young man's mother. The difficulty many people have in interpreting this story is, from
many perspectives, interesting. From the psycholinguistic point of view there is a clear indi-
cation that the reception of language is very strongly male-associated (cf. Rackow &
Wackwitz, 1998): even without explicit gender-marking, most people quietly assume that the
doctor and the specialist are male. From a cognitive-psychological viewpoint attention is
drawn to the difficulty many people have in re-interpreting the story (cf. Duncker,
1935/1963), even after the original assumption that the specialist is a man has led to an ab-
surd contradiction. Finally, the story is a further example of the fact, often pointed out in
feminism, that positive roles are frequently represented by masculine prototypes, which
supposedly put women at a clear disadvantage in reaching these positions (Bartholomew &
Schnorr, 1994; McConnell & Fazio, 1996). This could, for example, be demonstrated by the
fact that it is less likely for women to strive for the aforementioned roles or that men who
fulfill the role stereotype more completely are preferred in role assignment (Anderson, 1995;
Basow & Howe, 1980; Forsyth, Heiney, & Wright, 1997).
In the following section, we will first deal with the question of why many people stick to
the male interpretation of the "specialist problem" and why, even after seeing the striking
contradiction at the end of the problem, they still have difficulty in altering their interpreta-
tion. Subsequently, the central cognitive mechanisms that lie behind a successful solution to
the "specialist problem" will be examined from the viewpoint of problem-solving.
Male Language and Partial Matching
A quick glance into the research literature on gender stereotypes shows that the prototype
of a person who either demonstrates leadership qualities or is professionally successful or has
special intellectual abilities is normally male. This is valid for such various prototypes such
as that of an intelligent person (Raty & Snellman, 1997), a successful business manager
(Forsyth et al., 1997; Lord & Maher, 1991), and even for the culturally accepted and personal
concepts of God (Foster & Keating, 1992). Furthermore, in addition to these specific proto-
types, there is a tendency to usually assign the male gender to unspecified groups of people.
Merritt and Kok (1997) argue, on the basis of their empirical findings, that a people=male
bias leads people to attribute the male gender to gender-unspecified persons. For example,
the typical pedestrian is categorized as male (Kowal, O'Connell, & Posner, 1995) or the
gender-neutral pronoun "they" provokes the notion of masculinity to a stronger degree (Hyde,
1984; Switzer, 1990). These research results clearly confirm the presumption that the special-
ist is assigned the masculine gender first. Nevertheless, these findings do not explain why

H. Stoeger, A. Ziegler, H. David 516
people have such great difficulties in correcting gender attribution once it has been estab-
lished. In this context, research findings regarding the so-called Moses illusion provide us
with some first hints.
A typical thinking task of this research paradigm, to which it owes its name, is to answer
the question of how many of each type of animal Moses took with him on the Ark. Typically,
most participants give the same (incorrect) answer two, whereby they overlook the fact that it
was not Moses, but rather Noah, who took the animals onto the Ark. As in the "specialist
problem", most participants working on the Moses illusion tasks form a believable initial
representation of the problem for themselves, which actually makes the task solution more
difficult. In both problems there are immense difficulties in critically questioning and modi-
fying the original representation of the problem.
In a series of tricky studies, some explanations for the Moses illusion could be disproved
(cf. Bredart & Modolo, 1988; Erickson & Mattson, 1981; Kamas, Reder, & Ayers, 1996;
Reder & Kusbit, 1991; Van Oostendorp & de Mul, 1990). The recipient of the question: (1)
does not, according to the cooperation principle, carry out a correction of the question be-
cause he/she presumes to know what the questioner meant, (2) he/she does not miss or ignore
the wrong term or, in other words, make a mistake in encoding, and (3) it is not the case that
he/she is not equipped with the relevant information about whether it was Moses or Noah
who took the animals on the Ark.
In fact, it seems to be that a partial-matching process is responsible for the original ac-
ceptance of the question's contents as well as the participant's sticking to the established
interpretation. According to this explanation, people understand and encode the question
linguistically correct, they have, in principle, the necessary information for the
re-interpretation in their memory, however they overlook an important discrepancy between
the question posed and the information activated in their memory. Thus, the characteristic
that both Moses and Noah were "patriarchs of the old testament" could be enough to verify
the correctness of the first interpretation. It is then decisive in this case that relevant charac-
teristics that would be important for the detection of the contradiction are not activated, and a
partial matching is accepted.
Assuming this explanation can be transferred to the specialist problem, then the difficulty
cannot be due to the facts that: (1) people understand the question wrongly, (2) they overlook
important information when the question is posed or (3) they do not possess the fundamental
knowledge that a specialist can be female. Rather, the information "specialist" triggers a
certain prototype, wherein a characteristic that is important for the solution to the problem,
namely the gender of the specialist, is already considered to be verified, although the special-
ist is only introduced in her gender-neutral professional role.
The Specialist Problem from the Perspective of Problem-Solving Psychology
The actual trick in the solution of the specialist problem seems to be what is often re-
ferred to as a Gestaltwechsel, or a radical restructuring and reinterpretation of the problem
(Fitzek & Salber, 1996). Similar to how when observing a three-dimensional sketch of a cube
the front side depends on which surface one focuses on, the solution of the specialist problem
requires a reinterpretation of the specialist's gender.

What is a specialist? 517
From the perspective of problem-solving psychology, three points can be named which
can positively influence a reinterpretation of the specialist problem: (1) Research has shown
that under certain conditions a problem whose content is familiar can be easier to solve
(Holding, 1985; Ziegler, 1994). This facilitation effect is mainly based on the fact that since a
more economical cognitive representation of the problem's starting condition is in effect,
mental resources are freed that can be used to search the problem space (Ziegler, 2000). (2)
Research in the area of problem-solving shows that in finding solutions, their availability and
accessibility can play an important role. The basic idea of this distinction is very old
(Holzman, Glaser, & Pellegrino, 1976; Tulving & Thomson, 1971), and can be found in
many research fields. A good illustration is, for example, the tip-of-the-tongue-phenomenon
(Brown & McNeill, 1966), whereby a piece of knowledge is in principle available, but cannot
be accessed. In the case of the specialist problem, the availability of a woman who is excep-
tionally skilled in medicine and the ability to have access to this knowledge is of central
importance. Primarily from information gathered in the areas of research on the effect of
beliefs derived from logical reasoning and stereotype research, it is known that expectations
facilitate the access to knowledge-congruent information stored in the knowledge base and
block the access to knowledge-incongruent information (e.g., Lepore & Brown, 1999; Wyer,
1998; Ziegler, 2000). Therefore, persons with a less pronounced male-associated specialist
conception should find it easier to access the example of an exceptionally skilled woman in
medicine in their knowledge base. One must here realize that availability and accessibility do
not necessarily imply that such knowledge units have already been stored in the memory. For
example, availability can also indicate that a specific piece of knowledge can be construed
through inductive or abductive inferences and the result will be accessible for further infer-
ences (Ziegler, 2000). (3) Should the explanation of a partial matching also apply to the spe-
cialist problem, the availability as well as the accessibility of the concept of a female special-
ist would also depend on which attributes are encoded under the concept "specialist". Sup-
pose that a person knows academically successful women, but no successful women in medi-
cine. Should the partial matching rest on the combination of attributes "academically success-
ful" instead of "successful in medicine", the reinterpretation of the specialist as a woman
would probably be carried out more easily.
Overview of the Current Research
In this contribution the results of three studies will be described. In Study 1, women
worked on the specialist problem, providing us with an initial baseline measurement of the
solution rate. Since it is plausible to assume that personality traits of the subjects can have an
influence on the solution, especially when they mediate the effect of pre-existing knowledge,
we considered the personal need for structure as a potential mediator (Neuberg, Judice, &
West, 1997; Thompson, Naccarato, & Parker, 1989). Study 2, which was carried out with
men and women, dealt with the fact that the male interpretation of the specialist problem
causes difficulties in finding a solution. Therefore, the text was reformulated so that the roles
of the mother and the father were exchanged in the problem scenario; the specialist, conse-
quently, was male. Study 3 was carried out with female and male participants as a cross-
cultural study. The countries Germany and Israel differ insofar as the Israeli family concept is
more strongly influenced by patriarchy. While the availability of a female specialist is sup-

H. Stoeger, A. Ziegler, H. David 518
posedly the same in both countries, the accessibility of a professionally successful woman is
expected to be more strongly blocked by the patriarchal family concept, which allows for
differentiated predictions with regards to solution rates. At the same time, the explanatory
power of partial matching was tested through priming. Finally, an investigation was made to
determine whether a further personal trait, the implicit personality theory according to Dweck
(1999), mediates the solution rate.
Study 1
The specialist problem was worked on for the first time in Study 1. Based on the afore-
mentioned factors, it is expected that many subjects will not find the solution. Therefore, the
results will give the first clues concerning the difficulty of the problem. The determination of
whether a specialist is indeed assigned the attribute "male" most of the time was of further
interest. Since in the literature there are many different indications regarding differential
thinking performances or thinking styles of university students majoring in various subjects
(cf. the classical studies by Cox & Griggs, 1982, and Jackson & Griggs, 1988), this variable
was also taken into account. For example, it could easily be the case that language students,
on the grounds of their intense experiences with text analysis would take different paths in
solving the problem than students with qualifications in logic or students of life sciences,
which concentrate on the higher importance of gender variables. Furthermore, a personal trait
of the participants in this study was recorded, which could possibly influence the effect of the
masculine interpretation of the specialist and thus influence the subjects' finding the solution.
In the past few years a series of stable, dispositional motives which influence people's think-
ing have been identified (Gollwitzer & Bargh, 1996; Sorrentino & Higgins, 1996). The per-
sonal need for structure describes a temporally stable preference for cognitive simplicity and
structure (Neuberg et al., 1997; Thompson et al., 1989). People with a high personal need for
structure show, among other things, a stronger tendency to integrate information into existing
knowledge structures (Thompson, Roman, Moskowitz, & Chaiken, 1994), they stereotype
more strongly (Schaller, Boyd, Yohannes, & O'Brien, 1995) and they form simplifying trait
inferences more quickly (Moskowitz, 1993). From this, one could infer that this dispositional
construct of the personal need for structure does indeed influence the ability to find the solu-
tion, in that people with a high need would be expected to show relatively poorer solution
rates because they have more problems freeing themselves from their knowledge and assess-
ing information from another perspective.
Study 1
Method
Participants. Participants in Study 15 were 144 female students of various subjects at the
University of Munich: 51 studied psychology, 21 natural sciences, 26 business, 19 philoso-
5 For each of the studies subjects who were already familiar with specialist problem needed to be disquali-
fied. This figure ranged between 7-18% for these studies.

What is a specialist? 519
phy (which includes two obligatory courses in logic) and 26 German language. They were
randomly approached in their departmental libraries and did not receive any payment for their
participation in the study.
Material. All participants completed a uniform questionnaire in the German language. In
addition to information concerning their age and major, the questionnaire contained a Ger-
man version of the specialist problem printed in its English version above. This problem is
based on a well known logic question6. As a result of a pilot study, the solution was sought in
the following manner: "Can the problem be solved and explained in one sentence? If you are
of this opinion, state your solution in a single sentence. Otherwise, write 'no'." Answers
which stated that the specialist is the mother or a woman were considered to be correct. The
solution was coded as incorrect when, in addition to the statement that the problem is unsolv-
able, answers were given such as the specialist is the boy’s stepfather, a priest or another
member of the clergy, or any sort of supernatural explanations such as reincarnation.
By means of a semantic differential, the participants were examined to determine whether
they imagined a more masculine or feminine person under the concept "specialist". More-
over, four further word pairs were given which were either mentioned in a free-answer de-
scription of a specialist in a pilot study, or were added for our own considerations as in the
case of the age variable. Ratings along a 6-point bipolar scale were made for the following
five pairs: cold-hearted vs. warm-hearted, under 50 years old vs. older than 50 years, male vs.
female, honest vs. dishonest, and a well-rounded scholar vs. a limited expert. The following
statement was printed above the rating scales: "What, in your opinion, is a specialist? Is a
specialist more: ..."
The Personal Need for Structure Scale (Thompson et al., 1989) comprises 12 items along
a 6-point Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree). The scale
satisfies reliability and validity criteria (Thompson et al., 1989; Neuberg et al., 1997) and the
Cronbach's alpha determined in our sample was .85. Sample items are: “I enjoy having a
clear and structured mode of life", and “I hate to be with people who are unpredictable."
Firstly, personal information was asked for in each questionnaire. Then, the semantic dif-
ferential and the specialist problem and, finally, the Personal Need for Structure Scale was
given. It was accepted here that the mention of the possibility that the specialist could be a
woman in the semantic differential might facilitate finding the solution. This effect, in fact,
would be rather conservative, i.e. the expectation that the specialist problem could not be
solved by most of the participants, would therefore be more difficult to confirm.
Results
The average age of the participants was M=23.43 (S=2.89) years. Age had no systematic
relationship to any other variable in this or in either of the subsequent studies, and is there-
fore not further addressed. The results of the semantic differential are shown in Figure 1. It is
of particular interest that a specialist is indeed expected to be male (M=2.76, S= 1.16). Only
22 participants assume that the specialist is more likely to be a woman than a man (scale
6 Here it is important to mention that the term specialist was not translated by the German word Spezialist
which is masculine, but rather by the German word Koryphäe, which is feminine in the German language.

H. Stoeger, A. Ziegler, H. David 520
value > 3). A binomial test shows that except for the well-rounded vs. limited knowledge
(p < 1), all distributions are significantly skewed.
A factor analysis with varimax rotation and an eigenwert > 1 proves two factors, the first
of which explains 31.2% of the variance and the second a further 22.7%. For the first factor,
the tendency is towards cold-hearted (-.64), male gender (-.52), honest (.61) and well-
rounded scholar (.69). For the second factor, there is a tendency towards "older than 50
years" (.81) and male gender (.64), which loaded on both factors. In these results at least a
hint can be found that two different pictures of a specialist exist: a masculine person over 50
years old as well as a female specialist, who is characterized by warm-heartedness, honesty
and well-rounded educational interests. Still, keeping the picture drawn by the means in
mind, one sees that the vision of the male specialist is clearly dominant.
Table 1 shows the solution rate of the specialist problem according to the major subject of
study of the participants. A chi2-test shows no significant differences in their solutions,
chi2(4)=2.79, p < 1. The average solution rate of 32.2% indeed makes clear that the specialist
problem is a difficult task, which could only be solved by one out of three participants.
Coldness Over 50 Male Honesty Broadness of
Education
Figure 1:
Mean ratings of attributes of a specialist
(Min: 1, Max: 6; small values indicate high agreement)
Table 1:
Study 1: Frequency and percent of correct and incorrect answers, classified according to
major subject of study
Major Subject of Study
Answer Psychology Natural
Sciences Business Philosophy German
Language Total
Correct 19 5 7 8 7 46
37,5% 23,8% 26,9% 42,1% 26,9% 32,2%
Wrong 32 16 19 11 19 97
62,7% 76,2% 73,1% 57,9% 73,1% 67,8%

What is a specialist? 521
In the next step, an attempt was made to predict solution behavior with a logistic regres-
sion. The dependent variable was the dichotomous variable solution vs. non-solution of the
specialist problem. Predictors were the personal need for structure as well as the individual
factor scores from the factor analysis of the semantic differential items. None of the predic-
tors proved to be significant and the means for the solvers vs. non-solvers were almost iden-
tical. A further logistic regression in which the gender assigned to the specialist was used as a
predictor led to the same result.
Discussion
The main result of Study 1 is that the specialist problem is indeed a difficult task, which
could only be solved by one third of the participants. The solvability of the task appeared not
to be influenced by a series of variables, namely: major subject of study, whether one of the
two pictures of a specialist that were discovered in the factor analysis was represented, if the
specialist was rated as masculine, and personal need for structure. On the whole, these results
seem more likely to show that the reasons for the low solution rate must be sought from the
aforementioned perspective of problem-solving psychology.
Study 2
The purpose of Study 2 was to show that, above all, it is the Gestaltwechsel which is the
cause for the low solution rate in the specialist problem. It was already argued that, primarily,
the low accessibility of a female specialist is responsible for this. However, according to the
converse conclusion, the degree of difficulty of the specialist problem should sink if accessi-
bility can be ensured. A simple demonstration for this presumption can be reached by means
of a reformulation of the question so that the correct solution can be found through the acces-
sibility of a male specialist, i.e. that the father is the specialist.
Method
Participants. Nine female and 12 male psychology majors in the first semesters of their
studies at the University of Munich took part in this study. The experiment was carried out
during a regular seminar meeting.
Material. All participants worked on a uniform, modified form of the specialist problem,
in which instead of the father, a mother and her son were involved in the car accident. The
gender of the doctor and the specialist remained unspecified in this version.
Results
Of the 21 participants, 17 (7 female, 10 male) solved the modified specialist problem. A
binomial test with the solution rate of 32.2% found in Experiment 1 generates a probability of
p < 0.001 for this result to appear. Equally interesting were the reasons for the wrong solu-

H. Stoeger, A. Ziegler, H. David 522
tions, which all signaled that these subjects did not even realize the existence of a problem! It
was so self-explanatory for them that the specialist was the father, that they did not see a
contradiction.
Discussion
The results of Study 2 prove that the specialist problem no longer presents any difficulty
or even loses its problem nature when it is not a female, but rather a male specialist that has
to be inferred. This finding is a strong indication that in the available knowledge base, the
explanation for the low solution rate in the specialist problem must be sought in its original
format. In order to check this supposition participants with presumably different knowledge
bases and different accessibility to their knowledge bases (women and men, Israeli and Ger-
man university students) were examined in Study 3. Furthermore, a priming was given before
the actual task, which was expected to influence the accessibility of knowledge (as far as it
exists) relevant to the solution.
Study 3
The three factors varied in Study 3 (gender, nationality and priming) should be observed
according to three aspects: (1) How economical is the representation of the problem, i.e. how
much mental capacity is available to the participants for the completion of the problem-
solving operation. (2) Do they have the picture of a woman who is academically successful in
their knowledge base, i.e. availability. (3) Can they get to this knowledge, i.e. is it possible to
access to this knowledge. Through this, specific hypotheses concerning the interplay of these
factors in the solution of the specialist problem will be formed.
It can be assumed that the concept of a male specialist is equally shared by men and
women (Amancio, 1993). However, the participants in Study 3 are university students, which
should lead to an interesting difference in the knowledge bases of the female and male par-
ticipants: Women give more thought to the successful academic careers of women (namely,
their own) and men to the successful academic careers of men (also namely their own). Since
male students are more familiar with successful male scholars, they should be able to repre-
sent male specialists more economically. It follows from this that if women and men repre-
sent a specialist according to the male concept, the latter have more mental resources avail-
able to them for the solution to the problem.
However, women have - as mentioned above - more items of knowledge about successful
female careers in their knowledge bases, due to their concerns with their own careers. On the
other hand, due to the predominant male concept of successful scholars, they probably have
difficulties in activating this knowledge, although it is in principle available, accessing it is
presumably more difficult. A further difficulty is added for Israeli women. On the one hand,
equal access to a career for both sexes formally exists in both countries, even when this
equality of possibilities is, in fact, in no way realized in either country (Bundesministerium
für Familie, Senioren, Frauen und Jugend, 2002; Ma'or, 2002). On the other hand, the family
role concept is still more strongly patriarchal in Israel, i.e. a woman is pushed much more
into the mother/family role than in Germany (Israeli, 1992; Jerby, 1996; Scholz, 1997).

What is a specialist? 523
Therefore, and to a presumably higher degree in Israel than in Germany, there are two di-
verging tendencies open to women in universities, one is that of a woman striving for her
own academic career, the other encompasses a traditional family position. Since the tradi-
tional family role partially rules out that of a successful academic for women, it can be ex-
pected that the accessibility of this knowledge is more difficult for Israeli women.
This argumentation should be quickly summarized and the presumable effects should be
weighed against one another. Men have an advantage in the male-coined representation of the
specialist problem, which provides them with more mental resources for completing prob-
lem-solving operations. As the results of Study 2 have shown, this advantage should be rather
small as the actual difficulty of the specialist problem lies in the Gestaltwechsel. Here,
women should show an advantage due to their concerns with their own career chances, be-
cause more knowledge is available to them. However, because Israeli women belong to a
more strongly patriarchal society, a lower accessibility to their knowledge is expected from
them than from German women. Specifically, an interaction between culture and gender
regarding solution rate is expected.
With the priming used in Study 3, two goals were pursued: Indications for (1) a partial
matching and (2) the necessity to distinguish between availability and accessibility of knowl-
edge. The story used to prime the participants in Study 3 does not mention the attributes
successful/female doctor, but rather the attributes academically successful/female, which only
allows for a partial matching. However, it is important to emphasize here that priming can
generally only be successful when the corresponding knowledge is available. Priming can
therefore only have an effect on the accessibility of knowledge. However, since mainly
women have items of knowledge concerning successful female scholars, an interaction be-
tween priming and gender with regard to the solution rate is expected. Israeli women, in
particular, should also profit from the priming, as their difficulty in solving the problem
exists in the access to their available knowledge.
In addition to the variables discussed above, a further personality scale was used in Study
3. According to Dweck (1999), people can be differentiated as to whether they rate personal-
ity as an unchangeable and rigid characteristic (entity theory) or as an alterable one (incre-
mental theory). She was able to show in various studies that entity theorists have a stronger
tendency to form a belief (Chiu, Hong, & Dweck, 1997). When they have already formed an
opinion, they are additionally less sensitive to new (Erdley & Dweck, 1993; Plaks & Dweck,
1997) and contradictory information and even strive to actively avoid it (Plaks & Dweck,
1997). They are also more likely to retain stereotypes (Dweck, 1999; Levy, Stroessner, &
Dweck, 1998). Should personality factors, in addition to the aforementioned cognitive vari-
ables, play a role in solving the specialist problem, an incremental theory could be a good
candidate for explaining why persons succeed in freeing themselves from existing knowl-
edge.
Method
Participants. Fifty male and fifty female psychology majors in the first semesters of their
studies at the University of Tel Aviv and the University of Munich took part in this study.
They were approached randomly in a computer room (Israel) or in departmental libraries and
received no reimbursement for their participation in the study.

H. Stoeger, A. Ziegler, H. David 524
Material. Study 3 was part of a larger study that was not connected to the questions pur-
sued here, but rather researched the psychological causes of illnesses. In Germany the mate-
rial was made available in a German language format, for Israeli subjects the material was
translated into Hebrew. In order to guarantee that the German and Hebrew questionnaires
contained parallel statements, the German material was translated into Hebrew by one person
and then translated back into German by another person. Besides the specialist problem7
described in Study 1, the questionnaire contained a priming as well as a scale to measure the
implicit theory of personality.
The priming took place by means of a supposed newspaper report about the award of a
scientific prize to a successful female chemist, which is handed over by a female professor
who is also the dean of the recipient's chemistry department. In the control condition, the
chemist and dean of the chemistry department are two male scientists. Because both, the
Israeli and the German languages mark gender, the exact English translation cannot suffi-
ciently convey the gender markings inherent in everyday spoken language. Therefore, in the
following translation of the female priming, the symbol "*" denotes the presence of a female
gender-marking. The wording of the priming was the following: "Yesterday afternoon, Prof.
Dr. Blaschke, director* and chemist* at the Wöhlerinstitut in Munich [Tel Aviv], was pre-
sented the scientist's award of the city of Munich at a small awards ceremony given by Prof.
Dr. Vilser, dean* of the chemistry department. Prof. Dr. Blaschke achieved distinction
through publications in international scientific journals. In the past year, she achieved inter-
national reputation by means of her discovery of a new method of extraction that greatly
simplified gas chromatography. The findings, attained through her new method of extraction,
have helped her to make ground-breaking discoveries, particularly in aroma research. In her
award speech, Dean* Vilser stressed that *Prof. Dr. Blaschke now is counted among the 10
most frequently quoted persons in German [Israeli] chemical research. The cash prize of
$ 20,000 which accompanied the award will be, according to the scientist*, used for the
acquisition of desperately needed research materials."
The participants were asked to evaluate the newspaper report with regard to several di-
mensions concerning journalistic quality. The subjects in the two testing conditions didn´t
differ in the evaluation of these dimensions. Between the priming and the specialist problem,
and in addition to 5 questions belonging to the other study, the implicit theory of personality
was measured by three items in a 6-point Likert-type format ranging from 1 (strongly agree)
to 6 (strongly disagree). Detailed information concerning reliability and validity of the scale
can be found in Dweck (1999). A sample item is "The kind of person someone is, is some-
thing very basic about them and it can't be changed very much." The Cronbach's alpha for the
Israeli sub sample was .84, and for the German subsample it was .81.
7 In Hebrew there are two possible translations of the word specialist (possible translation 1: male form:
mumche, female form: mumchit, possible translation 2: male form: memune, female form: memuna). In
order to avoid an influence by the male/female forms on the investigative participants, the term specialist
was replaced by the word memune/a.

What is a specialist? 525
Results
The data were first analyzed to see if the solution was influenced by age or major subject
of study, which was not the case. Table 2 contains the numbers of correct solutions structured
according to priming, gender and nationality of the participants. The solution rate of the
entire sample was 40%, and was similar to that of Study 1. However, a 2 (priming with a
successful female vs. successful male scientist) X 2 (gender) X 2 (nationality) X 2 (solution
vs. non-solution of the specialist problem) log-linear analysis showed three significant inter-
actions between the solution to the problem and further variables. The 2-way-interaction of
priming and problem solution, chi2(1)=3.24, p < 0.05, was modified by means of the pre-
dicted 3-way-interaction of priming, gender and problem solution, chi2(1)=3.17, p < 0.05, for
this reason only the latter should be interpreted. An inspection of Table 2 shows that in fact,
the priming was only effective with the female participants. Additionally, the predicted
3-way-interaction of gender, nationality and problem solution could be confirmed,
chi2(1)=9.07, p < 0.01. Table 2 illustrates the manifestation of this interaction: In the non
priming condition German women and Israeli men solve the specialist problem somewhat
better. In light of the priming effects on Israeli women, this result allows us to assume that
the Israeli women's low solution rate is not caused by the lacking availability of knowledge,
but rather by their restricted access to this knowledge. This effect was apparently so strong
that they even performed somewhat worse than Israeli men who, by comparison, have a
representation advantage. In contrast, the German women in our study apparently not only
had a higher availability but also a better access to this knowledge than the German men had.
In the last step of the analysis a series of logistic regressions were calculated, wherein the
implicit theory of personality functioned as a predictor, and the solution of the specialist
problem functioned as a dependent variable. Although no alpha-adjustment took place, statis-
tically verifiable possibilities for prediction could not be verified for the whole group nor for
any of the subsamples made up by possible combinations of priming condition, gender and
nationality (each p < 1).
Table 2:
Study 3: Frequency of correct answers, classified according to priming, gender of the
participants and country
Country
Priming Gender Answer Germany Israel Total
Female Scientist Female Correct 17 9 26
Wrong 8 16 24
Male Correct 6 9 15
Wrong 19 16 35
Male Scientist Female Correct 10 4 14
Wrong 15 21 36
Male Correct 6 9 15
Wrong 19 16 35

H. Stoeger, A. Ziegler, H. David 526
Discussion
As in Study 1, no proof for an influence of major subject of study or of a personality trait
- in this case, the implicit theory of personality (Dweck, 1999) - could be found. On the other
hand, the assumptions based on problemsolving psychology could be proven. Even a partial
matching between the specialist problem story of a successful female chemist, which was
placed before the specialist problem, and the successful female doctor was enough to raise, in
women, the accessibility of their knowledge. Furthermore, both of the expected interactions
between priming and gender and between gender and nationality could be statistically con-
firmed. However, the latter interaction deserves more attention because the appearance of the
interaction was somewhat surprising. In fact, Israeli women's low solution rate in comparison
to German women and their somewhat poorer performance in comparison to Israeli men give
evidence for a strong effect of the traditional family concept which obviously inhibits their
access to knowledge of a successful female scholar. It is shown, through the fact that priming
was effective in Israeli women, that at least some of them have this knowledge at their dis-
posal. All in all, the results refer to the importance of an interplay between women's avail-
ability and access to knowledge about the careers of successful academic women, which is
influenced by societal male-dominated concepts. In the following, this will be discussed on
the basis of all three studies.
General Discussion
The specialist problem used in this investigation had an academically successful woman
as its solution, however, this was not discovered by the majority of the participants in our
studies (Studies 1 and 3). In contrast, the participants in Study 2 either had no difficulty in
coming to the equivalent solution of an academically successful man, or did not even per-
ceive a problem in the reformulated version. This result mirrors a reality in which the picture
of an academically successful man is still dominant (David, 2003). At the same time, the
postulated psychological mechanisms offer not only possible explanations for the effects of
the male-coined concept of academically successful people, but can also be understood as
hints as to where a change must be implemented.
The discussion of the results should be opened up by addressing the two potential starting
points for the fight against the effects of a gender-role concept based on a successful male
scholar which found no support in our studies: an improvement in problem solving style and
a change in personality characteristics that cause an orientation on male concepts. The major
subject of study pursued by participants in Studies 1 and 3 had no influence on their ability to
find the solution, which shows that the ways of approaching problems associated with the
different majors cannot clear up the main difficulty in the specialist problem. Included in our
studies were, among others: persons with a training in logic; students of languages, whose
education is supposed to especially sensitize them concerning the use of language; and stu-
dents of life sciences, who supposedly should have been made aware of gender problems
through their studies. The two personality traits that we studied which influence gender-
biased thinking were the personal need for structure (Study 1; cf. Thompson et al., 1989) and
the implicit theory of personality (Study 3; cf. Dweck, 1999). Neither personality traits had
any connection to the solution of the problem. Of course, the negative findings can, neither in

What is a specialist? 527
the case of major subject of study nor in the examination of only two personality traits, give a
final insight into whether influences of problem-solving strategies or personality characteris-
tics actually could be proven. However, the results of our studies suggest that the reasons for
the effect of male-dominated concepts should be sought in other areas.
From the results of Study 3 it can cautiously be concluded that men have a small advan-
tage in the representation of the problem, which is why Israeli men could solve the specialist
problem somewhat better than Israeli women could. The resulting measure to promote the
male-coined gender-role concept of the successful scholar is, however, due to its
socio-political undesirability and moral questionability, unacceptable.
Theoretical analyses, which were carried out under the perspective of problem-solving re-
search, and the results of Study 3 point to the necessity of differentiating between the avail-
ability and accessibility (Holzman et al., 1976; Tulving & Thomson, 1971; Ziegler, 2000) of
the concept of an academically successful woman. Consistent with the assumption of a par-
tial matching, it was shown that even priming of an academically successful woman is
enough to make it possible for many women to come to the assumption of a female medical
specialist. It can plausibly be supposed that a complete matching, i.e. the priming of a female
medical specialist, would have resulted in an even better solution rate. A follow-up of this
question could provide important insights into the effectiveness of role models in career
settings.
The necessity of a differentiation between availability and accessibility could be proven
by means of the effect of priming. This not only made improved performances by German
women possible, but also by Israeli women who, due to the competing concept of their fam-
ily roles, had more difficulty in accessing their knowledge of an academically successful
female. In fact, without priming, Israeli women solved the specialist problem less success-
fully than Israeli men did (16% vs. 36 %), with priming however they did just as well (36 %
vs. 36 %). Yet, German women were especially successful after a priming with a solution
rate of 68%. The corresponding knowledge is not only available, but additionally, their ac-
cess to it is much less hindered by a patriarchal familial concept.
The results of Study 3 allow for differentiated conclusions concerning the presence and
effect of the male concept of academically successful people. The lack of a priming effect on
men suggests that men possess hardly any or at least only insufficient information regarding
academically successful women. However, also women - as it was assumed - have knowl-
edge about this mainly due to their involvement with their own personal professional careers.
Still, there is a lack of corresponding role models, for which the number of female Nobel
Prize winners is provided as a simple indicator. As shown in a study by Cixous (1993),
among 510 winners of the Nobel Prize, only 24 were women. Nevertheless, it seems that the
mere availability of role models is not sufficient. In addition, access to these role models
must also be ensured so that they can exert an effect. In this case, Israeli women are appar-
ently hampered by the still strongly visible contradiction between professional and family
roles. An effective improvement in the situation of women must, therefore, consider at least
two points: (1) The knowledge of the possibility of academically successful women must be
conveyed (Zorman & David, 2000) for which, apart from the presentation of appropriate role
models, intensive individual involvement with this topic is suitable as a first step. This in-
volvement should, however, not be restricted to women, but should also be demanded from
men. For example, empirical studies show that employment decisions can be influenced by
gender-bound expectations that employers have concerning the applicants (Frable, 1989;

H. Stoeger, A. Ziegler, H. David 528
Perry, 1994). (2) Knowledge remains predominantly inert or ineffective when access to this
knowledge cannot be achieved. The results of our studies give evidence that not only the
male concept of an academically successful person but also concepts that are in conflict with
the notion of an academically successful woman are obstructive. Therefore, the desired goal
should not only be to anchor women's and men's knowledge about academically successful
women, but at the same time, to restrict the effectiveness of concepts which conflict with this
goal.
References
1. Amancio, L. (1993). Stereotypes as ideologies: The case of gender categories. Revista de
Psicologia Social, 8, 163-170.
2. Anderson, V. (1995). Identifying special advising needs of women engineering students.
Journal of College Student Development, 36, 322-329.
3. Bartholomew, C.G. & Schnorr, D.L. (1994). Gender equity: Suggestions for broadening
career options of female students. School Counsellor, 41, 245-255.
4. Basow, S.A. & Howe, K.G. (1980). Role-model influence: Effects of sex and sex-role attitude
in college students. Psychology of Women Quarterly, 4, 558-572.
5. Bredart, S. & Modolo, K. (1988). Moses strikes again: Focalisation effect on a semantic
illusion. Acta Psychologica, 67, 135-144.
6. Brown, R. & McNeill, D. (1966). The “tip of the tongue” phenomenon. Journal of Verbal
Learning and Verbal Behaviour, 5, 325-237.
7. Bundesministerium für Familie, Senioren und Frauen (2002). Frauen in der Bundesrepublik
Deutschland. [Women in the Federal Republic of Germany]. Bonn, Germany.
8. Chiu, C., Hong, Y., & Dweck, C. S. (1997). Lay dispositionism and implicit theories of
personality. Journal of Personality and Social Psychology, 73, 19-30.
9. Cixous, H. (1993). We who are free. Are we free? In B. Johnson (Ed.), Freedom and
interpretation. The Oxford Amnesty Lectures 1992 (pp. 17-44). New York, NY: Basic Books.
10. Cox, J. & Griggs, R.A. (1982). The effects of experience on performance in Wason's
selection task. Memory and Cognition 10, 496-502.
11. David, H. (2003).The influence of gender, religion, grade, class-type, and religiosity on
mathematical learning in the Israeli junior high school. Dissertation, LMU München: Fakultät
für Psychologie und Pädagogik.
12. Duncker, K. (1935/1963). Psychologie des produktiven Denkens [The psychology of
productive thinking]. Berlin, Germany: Springer.
13. Dweck, C.S. (1999). Self theories: Their role in motivation, personality, and development.
Philadelphia, PA: Psychology Press.
14. Erdley, C.S. & Dweck, C.S. (1993). Children's implicit theories as predictors of their social
judgements. Child Development, 64, 863-878.
15. Erickson, T.D. & Mattson, M. E. (1981). From words to meaning: A semantic illusion.
Journal of Verbal Learning and Verbal Behavior, 20, 540-551.
16. Fitzek, H. & Salber, W. (1996). Gestaltpsychologie. Geschichte und Praxis
[Gestaltpsychology. History and practice]. Darmstadt, Germany: Wissenschaftliche
Buchgesellschaft.
17. Forsyth, D.R., Heiney, M.M., & Wright, S. S. (1997). Biases in appraisals of women leaders.
Group Dynamics, 98-103.

What is a specialist? 529
18. Foster, R.A. & Keating, J.P. (1992). Measuring androcentrism in the Western god concept.
Journal for the Scientific Study of Religion, 31, 366-375.
19. Frable, D.E. (1989). Sex typing and gender ideology: Two facets of the individual's gender
psychology that go together. Journal of Personality and Social Psychology, 56, 95-108.
20. Gollwitzer, P. & Bargh J.A. (Eds.) (1996). The psychology of action: Linking cognition and
motivation to behavior. New York, NY: The Guilford Press.
21. Holding. D.H. (1985). The psychology of chess skill. Hillsdale, NJ: Erlbaum.
22. Holzman, T.G., Glaser, R., & Pellegrino, J. (1976). Process training derived from a computer
simulation theory. Memory and Cognition, 4, 349-356.
23. Hyde, J.S. ( 1984). Children's understanding of sexist language. Developmental Psychology,
20, 697-706.
24. Izraeli, D. (1992). Women trapped: The situation of the woman in Israel. Tel-Aviv, Israel:
Hakibbutz Hame'uchad (in Hebrew).
25. Jackson, S. & Griggs, R.A. (1988). Education and the selection task. Bulletin of the
Psychonomic Society, 26, 327-330.
26. Jerby, I. (1996). The double price: Women's status in Israeli society, and female military
service at the IDF. Tel Aviv, Israel: Ramot Publishing House.
27. Kamas, E.N., Reder, L.M., & Ayers, M.S. (1996). Partial matching in the Moses illusion:
Response bias not sensitivity. Memory and Cognition, 24, 687-699.
28. Kowal, S., O'Connell D.C., & Posner, R. (1995). Der prototypische Fußgänger: Zum
Menschenbild der amtlichen Verkehrszeichen [The prototypical pedestrian: The picture of
humans in official traffic signs]. Zeitschrift für Semiotik, 17, 151-163.
29. Lepore, L. & Brow, R. (1999). Exploring automatic stereotype activation: A challenge to the
inevitability of prejudice. In D. Abrams & M. A. Hogg (Eds.), Social identity and social
cognition (pp. 141-163). Malden, MA: Blackwell.
30. Levy, S., Stroessner, S., & Dweck, C.S. (1998). Stereotype formation and endorsement: The
role of implicit theories. Journal of Personality and Social Psychology, 74, 1421-1436.
31. Lord, R.G. & Maher, K.J. (1991). Leadership and information processing: Linking
perceptions and performance. London, UK: Unwin Hyman.
32. Ma'or, A. (Ed.) (2002). Women: The rising power. Advancing women at the workplace:
Smashing the "glass ceiling". Tel Aviv, Israel: Sifriyat Po'alim (in Hebrew).
33. McConnell, A.R. & Fazio, R.H. (1996). Women as men and people: Effects of
gender-marked language. Personality and Social Psychology Bulletin, 22, 1004-1013.
34. Merritt, R.D. & Kok, C.J. (1997). Implications of the People=Male Theory for the
interpretation of the Draw-A-Person Test. Journal of Personality Assessment, 68, 211-214.
35. Moskowitz, G.B. (1993). Individual differences in social categorization: The influence of
personal need for structure on spontaneous trait inferences. Journal of Personality and Social
Psychology, 65, 132-142
36. Neuberg, S.L., Judice, T.N., & West, S.G. (1997). What the Need for Closure Scale measures
and what it does not: Toward differentiating among related epistemic motives. Journal of
Personality and Social Psychology, 72, 1396-1410.
37. Perry, E. (1994). A prototype matching approach to understanding the role of applicant
gender and age in the evaluation of job applicants. Journal of Applied Social Psychology, 24,
1433-1473.
38. Plaks, J. & Dweck, C.S. (1997). Implicit person theories and attention to counter-expectant
social information. Paper presented at the Annual Convention of the American Psychological
Society, Washington, DC.

H. Stoeger, A. Ziegler, H. David 530
39. Rakow, L.F. & Wackwitz, L.A. (1998). Communication of sexism. In M.L. Hecht (Ed.),
Communicating prejudice (pp. 99-111). Thousand Oaks, CA: Sage Publications.
40. Raty, H. & Snellman, L. (1997). Children's images of an intelligent person. Journal of Social
Behavior and Personality, 12, 773-784.
41. Reder, L.M. & Kusbit, G.W. (1991). Locus of the Moses Illusion: Imperfect encoding,
retrieval, or match? Journal of Memory and Language, 30, 385-406.
42. Schaller, M., Boyd, C., Yohannes, J., & O'Brien, M. (1995). The prejudiced personality
revisited: Personal need for structure and formation of erroneous group stereotypes. Journal
of Personality and Social Psychology, 68, 544-555.
43. Scholz, R. (1997) Die Verwilderung des Patriarchats in der Postmoderne [The decline of
patriarchy in post modernity]. Psychologie & Gesellschaftskritik, 21, 31-51.
44. Sorrentino, R.M. & Higgings, E.T. (1996). Handbook of motivation and cognition:
Foundation of social behavior (Vol. 3). New York, NY: Guilford Press.
45. Switzer, J.Y. (1990). The impact of generic word choices: An empirical investigation of age-
and sex-related differences. Sex Roles, 22, 69-82.
46. Thompson, M.M., Naccarato, M.E., Parker, K.E. (1989, June). Assessing cognitive need: The
development of the Personal Need for Structure and the Personal Fear of Invalidity Scales.
Paper presented at the annual meeting of the Canadian Psychological Association, Halifax,
Nova Scotia, Canada.
47. Thompson, E.P., Roman, R.J., Moskowitz, G.B., & Chaiken, S. (1994). Accuracy motivation
attenuates covert priming: The systematic reprocessing of social information. Journal of
Personality and Social Psychology, 66, 474-489.
48. Tulving, E. & Thomson, D.M. (1971). Retrieval processes in recognition memory: Effects of
associative context. Journal of Experimental Psychology, 87, 116-124.
49. Van Oostendorp, H. & de Mul, S. (1990). Moses beats Adam: A semantic relatedness effect
on a semantic illusion. Acta Psychologica, 74, 35-46.
50. Wyer, R. S. (ED.) (1998). Stereotype activation and inhibition. Advances in social cognition
(Vol. 11). Mahwah, NJ: Erlbaum.
51. Ziegler, A. (1994). Die Entwicklung schlussfolgernden Denkens [The development of
deductive reasoning]. Frankfurt, Germany: Lang.
52. Ziegler, A. (2000). Der Überzeugungseffekt im logischen Denken Jugendlicher [The belief bias
in the logical reasoning of adolescents]. Regensburg: Roderer.
53. Zorman, R. & David, H. (2000). Female achievement and challenges toward the third
millennium. Jerusalem, Israel: Henrietta Szold Institute and the Ministry of education.