See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/246468067

Influence of Personality on Artistic and Scientific Creativity

Article · January 1999

CITATIONS READS
546 4,382

1 author:

Gregory J. Feist
San Jose State University
60 PUBLICATIONS 5,068 CITATIONS

SEE PROFILE

All content following this page was uploaded by Gregory J. Feist on 04 April 2016.

The user has requested enhancement of the downloaded file.

Review of General Psychology Copyright 2006 by the American Psychological Association
2006, Vol. 10, No. 2, 163–182 1089-2680/06/$12.00 DOI: 10.1037/1089-2680.10.2.163

How Development and Personality Influence Scientific Thought,

Interest, and Achievement
Gregory J. Feist
University of California, Davis

In the present article, I review and summarize two subdisciplines of the psychology of
science, namely development and personality. In the first section concerning develop-
mental psychology of science, I review three major developmental topics: 1) the
literature on the developmental and familial influences behind scientific interest and
scientific talent (e.g., birth-order and theory acceptance, immigrant status and scientific
talent); 2) gender and scientific interest and talent; and lastly, 3) age and scientific
interest and productivity. In the second section concerning personality psychology of
science, I organize the review around four major topics: 1) which traits make scientific
interest in general more likely; 2) which traits make interest in specific domains of
science more likely (especially social and physical science); 3) which traits make
different theoretical orientations more likely; and finally, 4) which traits make scientific
achievement and creativity more likely. From the empirical evidence reviewed, it is
quite clear that developmental and personality factors impact directly and indirectly
scientific thought, interest, and achievement.

Keywords: psychology of science, development, personality, mathematics, talent

To think of a scientist who appears full- many important implications and applications
blown as a scientist without a developmental to a well-formed psychology of science. Jean
path behind him or is to think the unimaginable. Piaget, for instance, was a key figure in inves-
Likewise, imagining a scientist without a tigating and conceptualizing how cognitive pro-
unique style of behaving and thinking is nearly cesses develop, change, and maintain over the
impossible. Scientific interest and achievement lifespan (Inhelder & Piaget, 1958; Piaget, 1952,
have fascinating and complex developmental 1972). As important of a figure as Piaget has
paths and are more likely to come from people been to the field of cognitive development, there
with particular kinds of personalities and traits have been important developments since Piaget,
than with other kinds of personalities. In this some of which are consistent with Piaget’s work
paper, I provide an overview and summary of and some of which are not. In this section, I
the current empirical literature on two subdisci- review three major topics in the field of devel-
plines in the psychology of science, namely opmental psychology of science: birth-order
development and personality psychologies of and theory acceptance, gender and development
science.1 of scientific interest, and developmental paths to
scientific eminence (i.e., age and scientific in-
Developmental Psychology of Science terest and productivity).
How do talented children become scientists
Developmental psychology of science is one of the first order? Family environments, which
of the most vibrant and active disciplines in the can either facilitate or hinder development of
psychology of science. For example, the study
and theorizing of cognitive development have
1
For sake of space, one developmental topic I leave for
another publication is the evidence for and theory on how
children use first principles in distinct domains of thought
Correspondence concerning this article should be ad- and can be viewed very much as implicit (folk) psycholo-
dressed to Gregory J. Feist, Department of Psychology, San gists, physicists, biologists, and mathematicians (see Feist,
Jose State University, San Jose, CA 95192. E-mail: 2006; Feist & Gorman, 1998; Klahr, 2000; Zimmerman,
[email protected] 2000).

163
164 FEIST

scientific interest and talent, clearly play an sun-centered theory. On the other hand, first-
important role in shaping and guiding such in- borns were more likely than laterborns to en-
terest and talent. There are numerous influences dorse and support conservative scientific theo-
from the early home environment that can have ries, such as vitalism or eugenics. Perhaps an
these effects, but the four that I review are even more telling finding is the fact that cre-
birth-order and theory acceptance, immigrant- ative-revolutionary thinkers themselves, at least
status, gender, and age and productivity. in the ideological revolutions of Copernicus and
Darwin, are more likely to be laterborns than
Birth-Order and Theory Acceptance firstborns. No such effect held for technical
revolutions (e.g., Newton, Einstein, Quantum
In his book Born to Rebel, Frank Sulloway theory), or conservative theories (e.g., vitalism,
(1996) makes a persuasive case that birth-order idealist taxonomy, or eugenics).
is a fundamental influence on an individual’s
disposition to accept or reject authority, whether Immigrant Status and Scientific Interest
it be familial, educational, political, social, or and Talent
scientific. The fundamental finding, one that he
puts in the context of evolutionary theory of One of the more interesting findings predict-
sibling rivalry and competition for resources, is ing scientific interest and talent has been immi-
that firstborns are disposed toward accepting the grant status, specifically being from a family
power structure they are born into, because they that is within two generations of immigrating to
are the oldest, strongest, and most identified the U.S. A disproportionate number of science
with the authority of their parents. Due to their majors, scientists, and elite scientists had at
temporary only-born status, they once garnered least one parent who was new to this country.
all the parental resources of attention and care, By disproportionate I mean upwards of 40%
so when siblings come they are then thrust into when only about 12% of the U.S. population in
positions of responsibility and power. Laterborn 2005 is first generation (foreign-born) Ameri-
children, on the other hand, are inherently dis- can (Berger, 1994; Camarota, 2005; Feist, in
posed toward questioning and challenging the press; Helson & Crutchfield, 1970; Portes &
innate power structure of the family, given their Rumbaut, 2001; Simonton, 1988a). For in-
built-in inferior status within the family. stance, in a study of elite scientists, Feist (in
What makes Sulloway’s argument persuasive press) recently reported that 20 out of 55 (36%)
is his extensive historical documentation and Westinghouse Science Fair finalists had a father
systematic testing of the basic hypothesis that and 22 out of 55 (40%) had a mother either born
firstborn individuals are more likely to accept elsewhere or were first generation Americans.
intellectually and politically conservative theo- The Westinghouse competition, as it was
ries and/or revolutions, whereas laterborn indi- known until 1998 (now it is the Intel competi-
viduals are more likely to support liberal theo- tion), is the oldest and most prestigious science
ries and/or revolutions. From the perspective of competition for high school students. Similarly,
the psychology of science, most relevant is Sul- in a sample of members of the National Acad-
loway’s detailed analysis of revolutionary the- emy of Sciences, 28 of 85 (33%) had fathers
ory acceptance in the history of science, mostly and 29 (34%) had mothers who were immi-
focusing on Charles Darwin’s theory of evolu- grants or first generation Americans (Feist, in
tion by natural selection, but also on Coperni- press). Membership in the National Academy
cus’s heliocentric theory as well as dozens of ranks second only to the Nobel Prize in prestige
other radical, technical, or conservative theories (Cole & Cole, 1973; Feist, 1997). What makes
in the history of science. To give but a few these figures all the more remarkable is that, in
examples of many: laterborns were 4.6 times general, the foreign-born population in the U.S.
more likely than firstborns to accept Darwin’s is poorer and less-well educated than the native
theory of natural selection in the sixteen years population (Camarota, 2005). One inference,
after it was first published than firstborns. They therefore, is that a particular subset of immi-
were also almost ten times as likely to accept grants or immigrant children appears to use
evolutionary theory prior to Darwin and more math, science, and technology careers as a way
than five times as likely to accept Copernicus’ out of poverty.
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 165

Families who recently come to the shores of think more scientifically (Despete, Roeyers, &
the U.S. may well foster a particular set of Buysse, 2001; Georghiades, 2000; Glynn &
values that encourages and maybe even de- Muth, 1994; White & Frederiksen, 1998).
mands high level achievement, whether it be in Scientific reasoning and hypothesis testing
science, medicine, or business. As suggested by require just such a distancing between one’s
classic work in the sociology of science, an thoughts and the evidence for them. Kuhn and
interesting speculation on this phenomenon is her colleagues have argued that the coordina-
that science may be more meritocratic than most tion of theory and evidence is the sine qua non
other career paths and therefore talent and of scientific reasoning (Kuhn, 1989, 1993;
achievement in and of itself is more likely to be Kuhn & Pearsall, 2000; Kuhn, Amstel,
recognized and rewarded (see Cole & Cole, O’Loughlin, 1988): “Accordingly, the develop-
1973; Merton, 1973). A significant scientific ment in scientific thinking believed to occur
finding is perhaps more likely to be evaluated across the childhood and adolescent years might
on its own merits than novel business or polit- be characterized as the achievement of increas-
ical ideas. Immigrant families may realize this, ing cognitive control over the coordination of
and, given that fluency in the native language theory and evidence. This achievement, note, is
may not be as critical as it is in other careers, metacognitive in nature because it entails men-
parents may therefore encourage their children tal operations on entities that are themselves
to go into math, science, or engineering careers. mental operations” (Kuhn & Pearsall, 2000, p.
Simonton (1988b) offers another possible ex- 115).
planation: “Individuals raised in one culture, but In short, being first or second generation can
living in another are blessed with a heteroge- for some make quite clear what their assump-
neous array of mental elements, permitting tions are and, by so doing, make reflective and
combinatory variations unavailable to those metacognitive thinking more likely. These cog-
who reside solely in on cultural world” (p. 126). nitive abilities, in turn, facilitate scientific inter-
Having been an exchange student in high school est and reasoning. So by being bicultural, a
myself, I can personally attest to the power of person grows up with quite a cognitive advan-
simultaneously having two cultural lenses tage, one that in the end makes him or her more
through which to compare experiences. By be- likely than others to be interested in and have
ing exposed to a different way of doing things talent for science.
and a different way of thinking, one’s own
implicit assumptions are more obvious and one Gender and Development of Scientific
takes less for granted what one believes, that is, Interest
without reflection.
And reflection and explicit thought is, as the One of the more entrenched influences on the
developmental psychologist Annette Karmiloff- development of scientific interest appears to be
Smith (1992) has made clear, a fundamental gender. As Evelyn Fox Keller (1985), among
feature of cognitive development. With devel- others, has pointed out, the history of science is
opment, thinking becomes more and more ex- replete with associations, both implicit and ex-
plicit and therefore more flexible and manipu- plicit, between science and men; male scientists
latable. Moreover, this is precisely what meta- historically have tried to “tame” or “control” the
cognition is— being aware of and being able to feminine “Mother Nature” (Fox Keller, 1985;
think about one’s thinking (Flavell, 1979; Sper- cf. Nosek, Banaji, & Greenwald, 2002).
ber, 1994; Sternberg, 1985). Highly intelligent There is empirical evidence that supports
and gifted students do tend to have higher meta- some gender differences in interest in science or
cognitive skills than less intelligent and gifted math, whether it comes in the form of explicit
students (Chan, 1996; Schwanenflugel, Stevens, attitudes (Eccles, 1987; Hyde, Fennema, Ryan,
& Carr, 1997; Shore & Dover, 1987). And yet Frost, & Hopp, 1990), implicit attitudes (Nosek
metacognition can be learned. A significant et al., 2002), performance on aptitude tests
body of literature now exists demonstrating the (Benbow & Stanley, 1983; Benbow & Lubinski,
effectiveness of teaching metacognitive skills in 1993; Benbow, Lubinski, Shea, & Eftekhari-
helping students to better understand mathemat- Sanjani, 2000; Geary, 1998; Halpern, 2000) or
ical and scientific concepts and, therefore, to actual graduation and career data (Cole, 1987;
166 FEIST

Cole & Zuckerman, 1987; Farmer, Wardrop, & doctoral degrees in science and engineering be-
Rotella, 1999; Jacobwitz, 1983; National Sci- ing awarded to women. And finally in terms of
ence Foundation [NSF], 1999; O’Brien, Mar- career, the disparity widens even more, with
tinez-Pons, Kopala, 1999; Reis & Park, 2001; only 4 to 6 percent of the full professors in
Subotnik, Duschl, & Selmon, 1993). The gen- science and math being women (NSF, 1999).
eral conclusion from this body of research is The most extreme gender difference is seen at
that men are more likely than women to view the most elite level (Long, 2001). On average,
science positively and be more interested in only two percent of the members of the National
science and math as a career. This is true when Academy of Sciences are female. Similarly,
both men and women view themselves as “the Feist (in press) found that although, compared
science type,” but even more so when they do to norms, male and female Westinghouse final-
not (Feist, Paletz, and Weitzer, 2005). More- ists were much more likely to earn PhDs or
over, although there is no overall gender differ- MDs (upwards of 80%), female finalists were
ence in intelligence, there does appear to be more likely than the male finalists to move away
some systematic differences in the mathemati- from science in terms of training and career.
cal domain (males being both higher and lower This finding is the same that Subotnik reported
than females) and in the verbal domain (females with the 1983 Westinghouse semifinalists (Sub-
being higher; Benbow & Stanley, 1983; Geary, otnik & Steiner, 1994), as well as with the study
1998; Halpern, 2000; Kimura, 1999; Stumpf & of mathematically precocious youth (Webb, Lu-
Stanley, 2002). binski, & Benbow, 2002). It is important to
There are, however, at least two important note, as Webb and colleagues (2002) have re-
qualifications to these generalizations. First, cently, that just because women may opt out of
gender differences are less apparent in child- science-oriented careers does not mean they opt
hood and adolescence than adulthood, which out of productive and achieving careers.
has been referred to as an “inverted funnel”
The second qualification concerning gender
effect; second, gender differences are less ap-
and science is that not all fields of science are
parent in the social sciences than in the physical
equally gender biased in their distributions. Un-
sciences, with biological sciences being in the
equal distributions are most striking in the phys-
middle.
Regarding the inverted funnel phenomenon, ical sciences, less striking in the biological sci-
in terms of courses taken, the “gender gap” in ences, and least striking in the social sciences.
science is not evident at the high school or Only 17 percent of the engineering degrees
undergraduate level or in the social sciences. and 35 percent of both the mathematics and
High school male and female students were physical-earth science degrees were awarded to
equally likely to take advanced math courses women, whereas nearly 50 percent of the bio-
(trigonometry and calculus) and almost as likely logical and social science degrees, and 73 per-
to take advanced science courses (biology, cent of the psychology degrees were awarded to
chemistry, and physics). In advanced science women in 1995 (NSF, 1999). Similarly, Long’s
courses, there were a slightly higher percentage (2001) analysis of trends in national samples
of females taking biology and chemistry and a showed that from 1973 to 1995 women went
slighter higher percentage of males taking phys- from being 2 percent to 12 percent of the engi-
ics. As students progress through their academic neering PhD graduates and from 21 percent
careers, however, there is an increasing gender to 51 percent of the social-behavioral science
disparity in interest in science and math (Long, PhD graduates. All other scientific fields were
2001; NSF, 1999; Rosser, 1988; Subotnik et al., in between these two ends of the continuum. In
1993). At the undergraduate level, the percent- addition, in a sample of mathematically preco-
age of women who earned science or engineer- cious students who immediately after high
ing degrees in 1995 was 46 percent (after being school said they intended to major in math or
about 38 percent ten years earlier). At the grad- science, five years later men were more likely to
uate level, a more obvious gender gap exists, have received engineering and physical science
even in the biological and social sciences, degrees and women more likely to have re-
with 39 percent of the masters degrees in sci- ceived biological science and medical science
ence and engineering and 33 percent of the degrees (Webb et al., 2002).
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 167

The gender gap, however, does seem to be likely to engage the child’s interest and curios-
narrowing somewhat compared to 30 years ago. ity. These results suggest that already at a young
For instance, the most exhaustive and extensive age, parents may be treating boys and girls
study of PhD scientists over a twenty-two year differently in science.
period (1973 to 1995) by the National Research Another set of possible explanations were
Council has documented progress but not yet proposed in 2005 by Harvard President Law-
equality for women in science (Long, 2001). rence Summers in his now infamous talk at a
When the appropriate controls (such as rank, conference on women in science. In order of
field, and institution) are made, the gender dis- priority, he hypothesized that high-powered
parity is not so extreme, but it still exists. For competitive jobs that require 80 hours a week
instance, men hold a fourteen percentage-point are less suited to women’s lifestyles, that “in-
advantage in holding tenure-track positions, but trinsic aptitude” differences might exist be-
this difference approaches zero once career age tween women and men, and differences in so-
is held constant. This suggests that the gender cialization and discrimination discourage
disparity in tenure-track science positions women from pursuing careers in science. The
should continue to decline as more and more one that started the firestorm was the “intrinsic
women become eligible. Also, salary differ- aptitude” hypothesis. In making his argument,
ences diminish once rank is controlled for, but Summers primarily focused on the differences
they do not disappear completely, suggesting in standard deviations between men and
that men do get paid a bit more for the same women, with men being disproportionately rep-
position. Similarly, Feist (in press) reported a resented at the low and high ends of many
significant increase in the percentage of female aptitude tests, especially in math and science.
Westinghouse finalists from 1965 to 1995. Here is a crucial excerpt from his speech:
In addition, marriage and family does affect
So my best guess, to provoke you, of what’s behind
men and women rank and productivity differ- all of this is that the largest phenomenon, by far, is the
ently, but not necessarily in the manner one general clash between people’s legitimate family de-
might expect. Long (2001), for instance, re- sires and employers’ current desire for high power and
ported that women who interrupted their careers high intensity, that in the special case of science and
engineering, there are issues of intrinsic aptitude, and
for marriage and family in 1979 were less likely particularly of the variability of aptitude, and that those
to obtain a tenure-track position, but there was considerations are reinforced by what are in fact lesser
no effect in 1995. For men, on the other hand, factors involving socialization and continuing discrim-
the effect of getting married and having children ination. I would like nothing better than to be proved
had a positive effect on productivity and this wrong, because I would like nothing better than for
these problems to be addressable simply by everybody
effect increased between 1979 and 1995. understanding what they are, and working very hard to
If the fact that some gender differences do address them (Summers, 2005).
exist (at least later in life and in the physical
sciences) is relatively agreed upon, the expla- In March of 2005, hundreds of mostly unfavor-
nation and cause of these differences remain able and critical responses to Summer’s com-
unclear and controversial. Insight into the ori- ments were published in newspapers and mag-
gins of the gender difference in scientific inter- azines around the country. One of the more
est comes from a study by Crowley, Callanan, sympathetic ones was written by Joan Ryan of
Tenenbaum, and Allen (2001). In samples of the San Francisco Chronicle, and it focused on
children ages 1 to 8, they found that there was a an interview Ryan had with University of Cal-
gender difference in frequency with which par- ifornia at San Francisco psychiatrist Louann
ents provided explanations (causal, correla- Brizendine, who is finishing up a book entitled
tional, or analogical) versus mere descriptions The Female Brain. Here is an excerpt:
of the exhibits on a visit to a local science I wanted to know if [Brizendine] bought into
museum. Parents were more likely to provide the argument that “intrinsic aptitude”—Summers’
explanations to boys and descriptions to girls. phrase— keeps women out of the top tiers of science
The explanation rates were about 29% for boys and engineering. Yes, she said, but not in the simple
way Summers suggests. It has nothing to do with the
and about 9% for girls. Explanations, of course, aptitude of men and women. It’s all about the aptitude
provide more complex reasoning about how of boys and girls, she said. The difference is in the
things work and why and therefore are more circuitry and the time line on which it develops. Dif-
168 FEIST

ferent abilities emerge at different ages for boys and to be a scientist by age 14, 50% knowing by
girls. Girls, Brizendine explained, develop language age 18, and 75% knowing by age 20. In terms of
skills earlier than boys do; boys develop visual and
spatial skills earlier than girls. By 2 1/2, many girls are first realizing they had talent for science, 25%
actively choosing not to play with boys, not for any of the NAS members realized their talent by
cultural or sociological reason but because boys have age 13, 50% by age 16, and 75% by age 21. The
not yet grasped the concept of verbal give-and-take. range was ages 5 to 33. Finally, NAS members
Boys, with their faster-developing spatial skills, are
more likely to gravitate to building blocks and train
began doing science early, with 75% having
sets and physical activities that require minimal verbal participated in formal research by age 21 (mean
interaction. (And they are more likely to find them- age ⫽ 19.2; median age ⫽ 20.0), and no gender
selves in altercations because they have poor external difference between men and women. Half of the
language to hammer out a solution and poorer internal NAS members had published a scientific article
language to mediate their impulses.) (Ryan, 2005).
by age 23, and at least one member had pub-
Brizendine then went on to discuss how by lished by age 16. Those who go on to have real
high school the brain differences have for the talent in science as adults seem to realize at a
most part dissipated and “their brains catch up young age that they want to become a scientist
to each other.” But due to their histories up to (on average by age 12 for Westinghouse final-
that point of developing different aptitudes, ists and by age 18 for NAS members). Given
there is an educational inertia where boys tend that most college students change their major
to stay in the spatial-mechanical domains and many times during their four years as an under-
girls more in the verbal-linguistic domains. In graduate, to have a group know by the end of
short, there are perhaps some “intrinsic differ- high school or before that a particular career is
ences” that biologically speaking narrow but right for them is quite remarkable. Early and
sociologically speaking widen with development. clear insight into one’s career calling is often an
indicator that one knows where one’s talents
Developmental Paths to Scientific Interest lies, and indeed such “crystallizing experi-
and Achievement ences” are frequently seen in adolescents who
go on to be our most creative adults (Cameron,
For those who go on to become scientists, Mills, & Heinzen, 1995; Freeman, 1999; Gard-
when do they know that science is the career for ner, 1993).
them? For those who do not go into science, Age and scientific productivity. The oldest
was there a time when science was interesting to and most established question concerning the
them? Finally, if a child expresses prodigious development of scientific productivity is of
talent for science before their teen years, are growth curves and publication rates, that is, how
they likely to actually become a scientist and, if productivity (i.e., publication rates) changes
so, are they likely to make significant contribu- with age. The findings have converged on the
tions to the discipline? Moreover, how to what conclusion that the relationship between age
extent does publishing early foreshadow a life- and productivity in science (and other profes-
time of scientific productivity? These are the sions) is an inverted-U (Bayer & Dutton, 1977;
central questions involving the development of Cole, 1979; Dennis, 1956; Diamond, 1986;
scientific interest and talent. Horner, Rushton, & Vernon, 1986; Lehman,
Age and scientific interest and talent. A few 1953, 1966; Over, 1982, 1989; Simonton,
researchers have addressed the question of who 1988a, 1988b, 1991; Zuckerman, 1996). Fur-
is likely to make significant contributions to ther, once controls are made for different ways
science and if so did they show early signs of of operationalizing output, the curve peaks
their talent. Feist (in press), for instance, exam- around 20 years into one’s career, usually in
ined the development of scientific interest in one’s early 40s. To graphically model this rela-
Westinghouse finalists and members of the tionship, Simonton has developed one of his
NAS. For Westinghouse finalists, both men and better-known differential equations, with the
women knew science was for them at an equally peak occurring roughly 20 years into one’s ca-
young age (males: M age ⫽ 11.12; females: M reer and thereafter slowly declining (Simonton,
age ⫽ 12.20). Members of the National Acad- 1988b). However, it does peak somewhat dif-
emy also developed an interest in science at a ferently for various disciplines (earlier in math
very early age, with 25% knowing they wanted and physics, later in biology and geology).
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 169

More recently, in a study of National Acad- ton’s curvilinear model where the peak tends to
emy of Science members, Feist (in press) re- occur approximately 20 years into one’s career,
ported three unconditional growth curve models with a gradual decline thereafter (Simonton,
that were constructed to test Simonton’s curvi- 1988a, 1988b, 1991). The current results, in
linear model of age and productivity (Simonton, fact, support another phenomenon reported by
1988b), namely a linear model, a quadratic Simonton, namely an end of life “swan-song”
model, and a cubic model. Each model provided effect (Simonton, 1990). In the sample of NAS
a close fit to the data, suggesting that publica- scientists, a cubic model with a second peak
tion rates increase over time. Out of the three toward the end of one’s career did a better job of
models, however, the curvilinear model pro- explaining the data than did the single peak
vided a better fit than the linear model, and the (curvilinear) model.
cubic model a better fit than the curvilinear Simonton has developed a complex theoreti-
model. In other words, a model with two peaks cal model that attempts to predict and explain
(approximately 20 years into one’s career and this age-productivity relationship by focusing
then again at the very end) was the best model on intrinsic factors, namely cognitive compo-
of age and productivity (see Figure 1). A con- nents (Simonton, 1988a, 1988b, 1991). His the-
ditional model was then tested, in which age of ory is based on his notion of “chance-configu-
first publication was used to predict the inter- ration” and consists of a few key assumptions:
cepts and growth curve trajectories of each First, each creator starts off with a set amount of
model. Age of first publication did predict the creative potential (number of contributions
midpoints (intercepts) in each model, but not made over a normal, unrestricted life span);
the trajectories. Such findings suggest that those second, the actualization of creative potential
who start publishing earlier compared with later can be broken down into two components, ide-
do have higher publication means at the mid- ation and elaboration. Ideation is the rate at
points in their career (time ⫽ 0), but do not have which potential ideas are expressed, whereas
different, linear, quadratic, or cubic trajectories. elaboration is the rate at which ideas are put into
These results are quite consistent with Simon- concrete, public form. So, as each creator pro-

Figure 1. Growth Curve Models of Scientific Productivity. In Feist (in press). The devel-
opment of scientific talent in Westinghouse finalists and members of the National Academy
of Sciences. Journal of Adult Development, Fig. 1. Copyright Springer Publishing. Reprinted
with kind permission of Springer Science and Business Media.
170 FEIST

duces a new work she or he “uses up” some eral on the personality traits and dispositions
creative potential. The rate at which a creator that predict scientific interest and scientific tal-
actualizes potential and produces works is a ent and how these traits continue to exert their
direct function of the two cognitive transforma- influence over the course of the lifespan. There
tions, ideation and elaboration. are at least four topics that have garnered
Related to age and productivity, there is the enough empirical support to review and sum-
question of whether early recognition of talent marize: which traits make scientific interest in
and producing works early in life predicts later general more likely; which traits make interest
levels of productivity. The empirical consensus in specific domains of science more likely (es-
is that early levels of high productivity do reg- pecially social and physical science); which
ularly predict continued levels of high produc- traits make different theoretical orientations
tivity across one’s lifetime (Cole, 1979; Dennis, more likely; and, finally, which traits make sci-
1966; Helson & Crutchfield, 1970; Horner et entific achievement and creativity more likely.
al., 1986; Lehman, 1953; Over, 1982; Reskin, Before getting into a review of these literatures,
1977; Roe, 1965; Simonton, 1988a, 1991). I first need to set the stage with a brief overview
Those who are prolific early in their careers tend of the function of traits. The functional-interac-
to continue to be productive for the longest tionist view lays out the mechanism by which
periods of time. personality is related to behavior, in this case,
Lastly, there is also the question of whether scientific behavior, interest, and achievement.
the age at which one’s talent for science is first
expressed predicts lifetime achievement in sci- Functional-Interactionist View of Traits
ence. Regarding age of recognized talent, Feist
(in press) predicted that in a sample of NAS To cite Gordon Allport’s famous phrase:
members, age of talent should predict age of “Personality is something and does something.”
publishing and obtaining the PhD, which in turn (Allport, 1937, p. 48). What it is and does is
should predict productivity and impact. Results directly affect behavior. In response to the in-
showed that the four precocity variables were famous “person-situation debate” (Block, 1977;
modestly positively correlated with age of first Epstein, 1979; Kenrick & Funder, 1988; Mis-
publication, which is an intermediate variable chel, 1968; Nisbett & Ross, 1980), which con-
between precocity and achievement. In other trasted personality and situational forces as
words, the younger NAS members were when competing explanations of behavior, many per-
they and others recognized their scientific tal- sonality psychologists have recently developed
ent, when they wanted to be a scientist, and a functional-interactivist theory of traits (Ey-
when they first conducted scientific research, senck, 1990; Feist, 1999; Funder, 1991; Mischel
the younger they were when they published & Shoda, 1999; Rosenberg, 1998). The func-
their first paper. Age of first publication in turn tional perspective maintains that traits function
predicted total publication rate over the lifetime, to lower behavioral thresholds, that is, make
meaning that the earlier one publishes, the more particular behaviors more likely in given situa-
productive one will be. This pattern of relation- tions; in short, they raise conditional probabili-
ships—from precocity to age of first publication ties (Mischel & Shoda, 1999). The primary
to lifetime productivity—implies an indirect function of traits, therefore, is to lower thresh-
connection between precocity and publication olds for trait congruent behavior (Brody & Ehr-
rate. The only precocity variable that reached lichman, 1998; Ekman, 1984; Eysenck, Mogg,
the .05 level of significance with lifetime pro- May, Richards, & Mathews, 1991; Rosenberg,
ductivity was age that one first conducted for- 1998). For instance, if a person has the traits of
mal research. “warm and friendly,” this means that in any
given situation she is more likely to act in a
Personality Psychology of Science warm and friendly manner than someone who
does not possess that trait. Her threshold for
In addition to developmental psychology, an- behaving in a friendly manner is lower than if
other field of psychology of science that is mod- she did not have that trait. Moreover, there are
erately active is personality psychology. Per- particular situations, such as on meeting a new
sonality psychology of science focuses in gen- person or being in a group of people, where
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 171

behaving this way is most likely. That is, in mensions of personality? Fortunately, the field
certain situations, traited behaviors are most of personality has recently witnessed a rela-
likely for particular people. The function of tively well agreed upon standardization of the
traits is they raise (or lower) the conditional basic dimensions of personality and these have
probability of a given behavior in a given been labeled the “Five Factor Model” (FFM) or
situation. the “Big Five.” The FFM is based on factor-
As social-cognitive theorists, such as Ban- analytic studies of personality structure that
dura and Mischel, have argued, to understand consistently extract five major factors of person-
how a person behaves in a particular situation ality (Costa & McCrae, 1995; Digman, 1990;
requires an interactionist approach (Bandura, Goldberg & Rosolack, 1994; John, 1990; Mc-
1986; Mischel & Shoda, 1999). Specifically, Crae & John, 1992). The five factors have var-
there are three main components that exert bi- ious labels, depending on the specific re-
directional influence on each other: Person, Be- searcher, but one of the more common labeling
havior, and Environment (Bandura, 1986; Ey- systems, and the one adapted here, is the fol-
senck, 1990; Mischel & Shoda, 1999). Bandura lowing: Extraversion (E), Agreeableness (A),
has presented perhaps the clearest model of how Conscientiousness (C), Neuroticism (N), and
these three components all are both causes and Openness (O) (Costa & McCrae, 1995). For
effects of each other. Behavior results from both current purposes, I used empirical findings from
environmental and personal characteristics, but the literature to classify a trait term or scale onto
environments are created by people acting in one of the FFM dimensions (Gerbing & Tuley,
particular ways. Furthermore, personal charac- 1991; Gough & Bradley, 1995; McCrae, 1991;
teristics, whether they are attitudes such as self- McCrae & Costa, 1985; McCrae, Costa, &
efficacy or traits, result from a person behaving Busch, 1986; McCrae, Costa & Piedmont,
consistently in a particular situation. 1993; Piedmont, McCrae, & Costa, 1991).
The functional-interactionist model of per- The two strongest effect sizes (medium in
sonality is quite useful in interpreting and ex- magnitude) were for the positive and negative
plaining the personality findings on scientific poles of conscientiousness (C; see Table 1).
behavior, interest, and talent. That is, certain Being high in conscientiousness (C⫹) consists
personality traits do make interest in and talent of scales and items such as careful, cautious,
for science more likely. What are they? conscientious, fastidious, and self-controlled,
whereas being low in conscientiousness (C⫺)
Personality and Scientific Interest
The first step toward being a scientist is sim-
ply having an interest in one form of science or Table 1
Personality Traits That Increase Interest in. . .
another. As it turns out, personality dispositions
have something to do with whether one be- Science
comes interested in science as a career choice or Conscientiousness
not. In 1998, I published a quantitative review Confidence
Dominance
of the literature on personality and scientific Introversion
interest and creativity (Feist, 1998). In this Physical science
meta-analytic review of which personality traits “Thing” vocational orientation
make interest and creativity in science more Asperger’s syndrome
likely, I found every published (and some un- (Extreme) Introversion
published studies) that examined the role in Social science
“People” vocational orientation
personality in scientific interest or scientific cre-
Psychological mindedness
ativity from 1950 to 1998. There were 26 stud- Extraversion
ies that reported quantitative effects of person- Subjective-internal theoretical orientation
ality in scientists compared to non-scientists. Empathy
One problem immediately arises when at- Flexibility
tempting to summarize on the same metric myr- Dominance
iad personality findings using different scales Intuitive
Intellectual efficiency
and items: How does one standardize the di-
172 FEIST

consists only of two scales/items, namely, direct autism than social scientists. In other words,
expression of needs and psychopathic deviance. physical scientists often have temperaments that
Although the C⫺ dimension comprised only orient them away from the social and toward the
five comparisons, it is clear that relative to inanimate—their interest and ability in science
non-scientists, scientists are roughly a half a is then just one expression of this orientation.
standard deviation higher on conscientiousness Such an orientation in one sense is an extreme
and controlling of impulses. In addition, low form of introversion, that is, it involves a lack of
openness to experience had a median d of .30, social interest and a not well-developed sense of
whereas introversion had a median effect size of theory of mind. Of course, autism and Asperg-
.26. Low openness consists of scales such as er’s syndrome are not simply extreme forms of
conventional, rigid, and socialized, whereas in- introversion, but rather their own category of
troversion consisted of terms such as deferent, social disorder. Nevertheless, there are impor-
reserved, introverted, and dependent. Finally, tant parallels between Asperger’s and introver-
examining the effect sizes of the two subcom- sion that warrant them being conceptualized as
ponents of extraversion separately (confidence aspects of a less social personality orientation.
and sociability), the confidence component had Moreover, autistic children are more than twice
a small positive effect and the sociability com- as likely as non-autistic children to have a father
ponent a near zero negative effect. In short, the or grandfather who was an engineer (Baron-
FFM dimensions of openness, confidence/dom- Cohen et al., 1998, Baron-Cohen, Wheelwright,
inance (E), introversion, and conscientiousness Skinner, Martin, & Clubley, 2001; Baron-Co-
and discipline appear to be the personality fac- hen, Wheelwright, Stone, & Rutherford, 1999;
tors that make scientific interest most likely. Baron-Cohen, Wheelwright, Stott, Bolton, &
Personality and interest in social versus Goodyer, 1997).
physical science. More specific than general One problem with the research on personality
interest in science, I also contend that one’s and scientific interest or achievement is that it is
preference and orientation toward people or not specific to any specific domain of science,
things plays a crucial role in the kind of science but rather covers scientists in general. Very
that one becomes interested in, especially phys- little if any research has compared the person-
ical or social science (see Table 1). As dis- ality dispositions of physical, biological, and
cussed already, the foundation for the People- social scientists to examine whether the social
Thing orientation comes from the vocational scientists have more sociable and extraverted
interest literature. Dale Prediger was the first to personalities compared to their physical scien-
modify John Holland’s hexagonal model of vo- tist peers. Of most interest would be develop-
cational interests onto two basic dimensions: mental research that examined whether a pref-
People-Things and Data-Ideas. The “People” erence for things is evident early in life for
end of the dimension is mapped onto Holland’s future physical scientists and, likewise, whether
“Social” career types, whereas the “Thing” end a preference for people is evident early in life
of the dimension is mapped onto “Realistic” for future social scientists. Similarly, cross-cul-
career types. According to Holland, the social tural work showing the same association be-
career type prefers occupations that involve in- tween thing-orientation and physical science
forming, training, enlightening other people. and social-orientation and social science the
The realistic career type, on the other hand, world over would be quite valuable. Therefore,
prefers careers that involve manipulating things, the next line of research for the personality
machines, objects, tools, and animals (Holland, psychology of science is to explore differences
1992; Lippa, 1998; Prediger, 1982). in personality between physical, biological, and
Supporting this domain-specific view of sci- social scientists. Based on the evidence just
entific interest, Simon Baron-Cohen and his col- cited, my prediction is that the physical scien-
leagues have found that engineers, mathemati- tists as a group will be more introverted and
cians, and physical scientists score much higher thing-oriented (that is, have more developed
on measures of high functioning autism (As- implicit physical intelligence) than the biologi-
perger’s syndrome) than non-scientists, and that cal scientists, who in turn will be less sociable
physical scientists, mathematicians, and engi- and extraverted than social scientists (i.e., have
neers are higher on non-clinical measure of more developed implicit social intelligence).
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 173

Personality and Theoretical Predilection In addition, the work on personality can also
shed light on theory acceptance and even theory
In addition to interest in science in general or creation. Or stated as a question: Does having a
the physical or social science in particular, there certain personality style predispose a scientist to
is the question of whether distinct personalities create, accept, and/or reject certain kinds of
are attracted to different theoretical perspectives theories? The first work on this question was in
in science. As we saw earlier from the meta- the mid 1970s by George Atwood and Silvan
analytic review, the answer seems to be yes, Tomkins (1976), who showed through case
certain personality traits do predict interest in studies how the personality of the theorist in-
science and research, even within the branch of fluenced his or her theory of personality. More
social science of psychology. Clinical psychol- systematic empirical investigations have ex-
ogy, for example, is an ideal domain in which to panded this work and have demonstrated that
address this question, because it emphasizes personality influences not only theories of per-
two different and distinct sets of skills in grad- sonality, but also how quantitatively or qualita-
uate education, namely applied-clinical skills tively oriented and how productive psycholo-
and research and scientific skills. What is gists are (Arthur, 2001; Atwood & Tomkins,
known as the “Boulder Model” was imple- 1976; Conway, 1988; Costa, McCrae, & Hol-
mented in the late 1940s, and it places equal land, 1984; Hart, 1982; Johnson, Germer, Efran,
emphasis on training in both research and clin- & Overton, 1988; Simonton, 2000). One gen-
ical practice. And yet in reality, a major concern eral finding from these studies is that psychol-
for PhD programs in clinical psychology is the ogists who have more objective and mechanis-
high rate of students who are not interested in tic theoretical orientations are more rational and
science and research. Clinicians do tend to be extraverted than those who have more subjec-
more people-oriented than investigative and re- tive and humanistic orientations. For instance,
search-oriented (Malinckrodt, Gelso, & Roy- Johnson and colleagues collected personality
alty, 1990; Zachar & Leong, 2000). data on four groups of psychologists (evolution-
An important question, therefore, has be- ary-sociobiologists, behaviorists, personality
come, “What predicts interest in science and psychologists, and developmental psycholo-
gists) and found that distinct personality profiles
research in these students and can this interest
were evident in the different theoretical groups.
be increased by particular kinds of training en-
That is, scientists who were more holistic, pur-
vironments?” The general conclusion from the
posive, and constructivist in orientation were
studies on these questions is that one of the
higher on the Empathy, Dominance, Intellectual
strongest predictors in interest in research (or Efficiency, and Flexibility scales of the Califor-
lack thereof) is personality-vocational interest nia Psychological Inventory and the Intuition
and that training environment plays but a mod- scale of the Myers-Briggs Type Indicator
est role in increasing interest in research (Kahn (MBTI). However, most of these studies have
& Scott, 1997; Mallinckrodt et al., 1990; Roy- been with psychologists, so answering the ques-
alty & Magoon, 1985; Zachar & Leong 1992, tion of whether these results generalize to the
1997). For example, a study by Mallinckrodt biological and natural sciences remains a task
and colleagues (1990) examined the impact of for future psychologists of science.
training environment, personality-vocational in-
terest, and the interaction between the two on Personality and Scientific Achievement
increasing research interest and found that per- and Creativity
sonality-vocational interest was a stronger pre-
dictor than research environment in increasing Not only do certain traits lower thresholds for
interest in research over the course of graduate scientific interest and theoretical orientation, but
training (see Table 1). I should point out, how- a somewhat different pattern of traits also low-
ever, that not everyone agrees with conceptual- ers thresholds for scientific creativity and emi-
izing vocational interests and personality to- nence. The meta-analysis conducted by Feist
gether. Waller, Lykken and Tellegen (1995), for (1998) also addressed the question of which
instance, argue they are separate dimensions, traits make creativity and eminence in science
even if vocational interests can be traits. more likely and what their magnitude of effect
174 FEIST

was. The traits can be arranged into three psy- and Intellect. Specifically, two measures of per-
chologically meaningful categories: cognitive, sonality—California Psychological Inventory
motivational, and social. scales of Tolerance (To) and Psychological
Cognitive traits that make scientific creativity Mindedness (Py)–resulted in the 20% increase
and eminence more likely. A consistent find- in variance explained (20%) over and above
ing in the personality and creativity in science potential and intellect. The more tolerant and
literature has been that creative and eminent psychologically minded the student was, the
scientists tend to be more open to experience more likely he was to make creative achieve-
and more flexible in thought than less creative ments over his lifetime. Together, the four pre-
and eminent scientists (see Table 2). Many of dictors (Potential, Intellect, Tolerance, and Psy-
these findings stem from data on the flexibility chological Mindedness) explained a little more
(Fe) and tolerance (To) scales of the California than a third of the variance in lifetime creative
Psychological Inventory (Feist & Barron, 2003; achievement. I should point out that these find-
Garwood, 1964; Gough, 1961; Helson, 1971; ings on To and Py mirror very closely those
Helson & Crutchfield, 1970; Parloff & Datta, reported by Helson and Pals (2000) in a longi-
1965). The Fe scale, for instance, taps into tudinal study of women from age 21 to 52.
flexibility and adaptability of thought and be- Motivational traits that make scientific cre-
havior as well as the preference for change and ativity and eminence more likely. The most
novelty (Gough, 1987). The few studies that eminent and creative scientists also tend to be
have reported either no effect or a negative more driven, ambitious, and achievement-ori-
effect of flexibility in scientific creativity have ented than their less eminent peers (see Table
been with student samples (Davids, 1968; 2). Busse and Mansfield (1984), for example,
Smithers & Batcock, 1970). studied the personality characteristics of 196
For instance, Feist and Barron (2003) exam- biologists, 201 chemists, and 171 physicists,
ined personality, intellect, potential, and cre-
and commitment to work (i.e., “need to concen-
ative achievement in a 44-year longitudinal
trate intensively over long periods of time on
study. More specifically, they predicted that
one’s work”) was the strongest predictor of
personality would explain unique variance in
productivity (i.e., publication quantity) even
creativity over and above that already explained
by intellect and potential. Results showed that when holding age and professional age con-
observer-rated Potential and Intellect at age 27 stant. Helmreich, Spence, Beane, Lucker, and
predicted Lifetime Creativity at age 72, and yet Matthews (1980) studied a group of 196 aca-
personality variables (such as Tolerance and demic psychologists and found that different
Psychological Mindedness) explained up to components of achievement and drive had dif-
20% of the variance over and above Potential ferent relationships with objective measures of
attainment (i.e., publications and citations).
With a self-report measure, they assessed three
Table 2 different aspects of achievement: “mastery”
Personality Traits That Make Scientific Creativity preferring challenging and difficult tasks;
More Likely. . . “work” enjoying working hard; and “competi-
tiveness” liking interpersonal competition and
Cognitive traits
Openness to experience
bettering others. According to Amabile’s (1996)
Tolerance well-known typology, the first two measures
Flexibility could be classified as “intrinsic motives” and
Social traits the last measure could be an “extrinsic motive.”
Dominance Helmreich and his colleagues found that mas-
Arrogance
tery and work were positively related to both
Hostility
Introversion publication and citation totals, whereas compet-
Self-confidence itiveness was positively related to publications
Motivational traits but negatively related to citations. Being intrin-
Driven sically motivated (mastery and work) appears to
Ambitious increase one’s productivity and positive evalu-
Intrinsically motivated
ation by peers (citations), whereas wanting to be
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 175

superior to peers leads to an increased produc- ligent, emotionally invested in intellectual

tivity, and yet a lower positive evaluation by work, and relatively happy. Similarly, Cham-
peers. The inference here is that being driven by bers (1964) reported that creative psychologists
the need for superiority may backfire in terms of and chemists were markedly more dominant,
having an impact on the field. Indeed, in a ambitious, self-sufficient, and had more initia-
further analysis of the male psychologists in tive compared to less creative peers. Helson
their 1980 data set, Helmreich and colleagues (1971) compared creative female mathemati-
(Helmreich, Spence, & Pred, 1988) factor ana- cians with less creative female mathematicians,
lyzed the Jenkins Activity Survey and extracted matched on IQ. Observers blindly rated the
an achievement striving factor and an impa- former as having more “unconventional thought
tience/irritability factor. Achievement striving processes,” as being more “rebellious and non-
was positively related to both citation and pub- conforming,” and as being less likely to judge
lication counts, whereas impatience/irritability “self and others in conventional terms.” More
was related to neither publications nor citations. recently, Rushton, Murray, and Paunonen
Social traits that make scientific creativity (1987) conducted factor analyses of the person-
and eminence more likely. In the highly com- ality traits most strongly loading on the “re-
petitive world of science, especially big science, search” factor (in contrast to a “teaching” fac-
where the most productive and influential con- tor) in two separate samples of academic psy-
tinue to be rewarded with more and more of the chologists. Among other results, they found that
resources, success is more likely for those who “independence” tended to load on the research
thrive in competitive environments, that is for factor, whereas “extraversion” tended to load on
the dominant, arrogant, hostile and self-confi- the teaching factor.
dent (see Table 2). For example, Van Zelst and To summarize the distinguishing traits of
Kerr (1954) collected personality self-descrip- creative scientists: they are generally more
open and flexible, driven and ambitious, and
tions on 514 technical and scientific personnel
although they tend to be relatively asocial,
from a research foundation and a university.
when they do interact with others, they tend to
Holding age constant, they reported significant
be somewhat prone to arrogance, self-confi-
partial correlations between productivity and
dence, and hostility.
describing oneself as “argumentative,” “as-
sertive,” and “self-confident.” In one of the
few studies to examine female scientists,
Bachtold and Werner (1972) administered Summary and Conclusions
Cattell’s 16 Personality Factor to 146 women In this paper, I have summarized and inte-
scientists and found that they were signifi- grated two of the five or six major psychologies
cantly different from women in general on of science, namely developmental and person-
nine of the sixteen scales, including domi- ality. Developmental psychology of science has
nance (Factor E) and self-confidence (Factor revolved around at least four major topics, three
O). Similarly, Feist (1993) reported a struc- of which I review here: a) familial influences on
tural equation model of scientific eminence in scientific interest and talent (i.e., birth-order and
which the path between observer-rated hostil- immigrant status); b) gender and science; and c)
ity and eminence was direct and the path age and scientific interest, talent, and produc-
between arrogant working style and eminence tivity. A fourth topic— how children are im-
was indirect but significant (see Figure 2). plicit psychologists, physicists, biologists, and
The scientific elite also tend to be more aloof, mathematicians—is summarized and reviewed
asocial, and introverted than their less creative elsewhere (Feist, 2006). Some of the highlights
peers. In a classic study concerning the creative of these findings are:
person in science, Roe (1952, 1953) found that • Latter-born scientists are more likely to
creative scientists were more achievement-ori- accept novel, revolutionary scientific theo-
ented and less affiliative than less creative sci- ries than first-born scientists.
entists. In another seminal study of the scientific • The scientific elite is about three times
personality, Eiduson (1962) found that scien- more likely to be offspring of foreign-born
tists were independent, curious, sensitive, intel- parents than the population in general.
 176
FEIST

Figure 2. Structural Model of Scientific Eminence. In Feist (1993). Model of scientific eminence. Psychological
Science, 4, 366-371. Copyright Blackwell Publishing. Reprinted with permission.
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 177

• There are some real gender differences in References

scientific interest and likelihood of remain-
ing in science, but these differences exist Allport, G. (1937). Personality: A psychological in-
mostly in the physical sciences and mostly terpretation. New York: Holt, Rinehart, &
Winston.
after adolescence. Causes and explanations Amabile, T. (1996). Creativity in context. New York:
for these differences remain unclear and Westview.
controversial. Arthur, A. R. (2001). Personality, epistemology and
• Those who go on to be our scientific elite psychotherapists’ choice of theoretical model: A
are very likely to have crystallized their review and analysis. European Journal of Psycho-
interest in and shown talent for science at therapy, Counseling and Health, 4, 45– 64.
an early age (before age 12) Atwood, G. E., & Tomkins, S. S. (1976). On subjec-
• When plotted against career age, publica- tivity of personality theory. Journal of the History
tion rates peak at about 20 years into one’s of the Behavioral Sciences, 12, 166 –177.
career. The growth rate before this peak is Bachtold, L. M., & Werner, E. E. (1972). Personality
characteristics of women scientists. Psychological
steeper than the decline rate after it.
Reports, 31, 391–396.
• Conscientiousness, self-confidence, domi- Bandura, A. (1986). Social foundations of thought
nance, and openness are the traits most and action: A social cognitive theory. Englewood
likely to lower the thresholds for scientific Cliffs, NJ: Prentice-Hall.
interest. Baron-Cohen, S. Bolton, P., Wheelwright, S., Short,
• Being thing-oriented and introverted low- L., Mead, G., Smith, A., & Scahill, V. (1998).
ers one’s threshold for interest in and talent Autism occurs more often in families of physicists,
for physical science, whereas being peo- engineers, and mathematicians. Autism, 2, 296 –
ple-oriented, psychologically minded, and 301.
extraverted do the same for the social sci- Baron-Cohen, S., Wheelwright, S., Skinner, R., Mar-
tin, J., & Clubley, E. (2001). The Autism-Spec-
ence interest and talent.
trum Quotient (AQ): Evidence from Asperger syn-
• Cognitive traits (e.g., tolerance and flexi- drome/high-functioning autism, males and fe-
bility), social traits (e.g., dominance, arro- males, scientists and mathematicians. Journal of
gance, hostility), and motivational traits Autism & Developmental Disorders, 31, 5–17.
(e.g., driven, intrinsically motivated) each Baron-Cohen, S., Wheelwright, S., Stone, V., &
lower thresholds for scientific creativity Rutherford, M. (1999). A mathematician, a phys-
and achievement. icist, and a computer scientist with Asperger syn-
As the results of this paper should make drome: Performance on folk psychology and folk
clear, scientists are people with particular de- physics tests. Neurocase, 5, 475– 483.
velopmental histories and personality traits. Baron-Cohen, S., Wheelwright, S., Stott, C., Bolton,
P., & Goodyer, I. (1997). Is there a link between
Gradually but steadily, what psychologists of
engineering and autism? Autism, 1, 101–109.
science are learning about the nature of scien- Bayer, A. E., & Dutton, J. E. (1977). Career age and
tific interest, motivation, talent, and achieve- research–professional activities of academic scien-
ment is reaching a “point of no-return”—that is, tists: Tests of alternative non-linear models and
if we want to understand what science is at its some implications for higher education faculty
best and worst, and who does it, then we must policies. Journal of Higher Education, 48, 259 –
include the psychological perspective. No 822.
longer can other disciplines that study science Benbow, C. P., & Lubinski, D. (1993). Psychological
ignore the insights into scientific thought and profiles of the mathematically talented. Some sex
behavior that psychologists are uncovering. differences and evidence supporting their biologi-
More to the point for psychology, no longer can cal basis. In G. R. Bock and K. Ackrill (Eds.). The
origins and development of high ability (pp. 44 –
psychologists who study any form of scientific
66). Chichester, UK: Wiley and Sons.
thought or behavior–from the neuroscience of Benbow, C. P., Lubinski, D., Shea, D. L., &
mathematical problem solving to group influ- Eftekhari-Sanjani, H. E. (2000). Sex differences in
ence in the scientific laboratory— continue to mathematical reasoning ability at age 13: Their
do so blind to the fact that they are doing status 20 years later. Psychological Science, 11,
“psychology of science.” 474 – 480.
178 FEIST

Benbow, C. P., & Stanley, J. C. (1983). Sex differ- Davids, A. (1968). Psychological characteristics of
ences in mathematical ability: More facts. Science, high school male and female potential scientists in
222, 1029 –1031. comparison with academic underachievers. Psy-
Berger, J. (1994). The young scientists: America’s chology in the Schools, 3, 79 – 87.
future and the winning of the Westinghouse. Read- Dennis, W. (1956). Age and productivity among sci-
ing, MA: Addison-Wesley. entists. Science, 123, 724 –725.
Block, J. (1977). Recognizing the coherence in per- Dennis, W. (1966). Creative productivity between
sonality. In D. Magnusson & N. D. Endler (Eds.). the ages of 20 and 80 years. Journal of Gerontol-
Personality at the crossroads: Current issues in ogy, 21, 1– 8.
interactional psychology. Hillsdale, NJ.: Erlbaum Despete, A., Roeyers, H., & Buysse, A. (2001).
& Associates. Metacognition and mathematical problem solving
Brody, N., & Ehrlichman, H. (1998). Personality in Grade 3. Journal of Learning Disabilities, 34,
psychology: The science of individuality. Upper 435– 449.
Saddle River, NJ: Prentice Hall. Diamond, A. M. (1986). The life-cycle research pro-
Busse, T. V.& Mansfield, R. S. (1984). Selected ductivity of mathematicians and scientists. Journal
personality traits and achievement in male scien- of Gerontology, 41, 520 –525.
tists. The Journal of Psychology, 116, 117–131. Digman, J. M. (1990). Personality structure: Emer-
Camarota, D. (2005). Immigrants at mid-decade: A gence of the five-factor model. Annual Review of
snapshot of America’s foreign-born population in Psychology, 41, 417– 440.
2005. Retrieved on March 23, 2006, from http:// Eccles, J. (1987). Gender roles and women’s achieve-
www.cis.org/articles/2005/back1405.pdf ment-related decisions. Psychology of Women
Cameron, P. A., Mills, C. J., & Heinzen, T. E. (1995). Quarterly, 11, 135–172.
The social context and developmental patterns of Eiduson, B. T. (1962). Scientists: Their psychologi-
crystallizing experiences among academically tal- cal world. New York: Basic Books.
ented youth. Roeper Review, 17, 197–200. Ekman, P. (1984). Expression and the nature of emo-
tion. In K. R. Scherer & P. Ekman (Eds.). Ap-
Chambers, J. A. (1964). Relating personality and
proaches to emotion, (pp. 319 –344). Hillsdale, NJ:
biographical factors to scientific creativity. Psy-
Erlbaum.
chological Monographs: General and Applied, 78,
Epstein, S. (1979). The stability of behavior: I. On
1–20.
predicting most of the people much of the time.
Chan, L. K. S. (1996). Motivational orientations and
Journal of Personality and Social Psychology, 37,
metacognitive abilities of intellectually gifted stu-
1097–1126.
dents. Gifted Child Quarterly, 40, 184 –193.
Eysenck, H. J. (1990). Biological dimensions of per-
Cole, J. R. (1987). Women in science. In D. Jackson sonality. In L. A. Pervin (Ed.), Handbook of per-
& P. J. Rushton (Eds.), Scientific excellence (pp. sonality theory and research (pp. 244 –276). New
359 –375). Beverly Hills, CA: Sage. York: Guilford.
Cole, J. R., & Cole, S. (1973). Social stratification in Eysenck, M. W., Mogg, K., May, J., Richards, A., &
science. Chicago: University of Chicago Press. Mathews, A. (1991). Bias in interpretation of am-
Cole, J. R., & Zuckerman, H. (1987). Marriage, biguous sentences related to threat in anxiety.
motherhood, and research performance in science. Journal of Abnormal Psychology, 100, 144 –150.
Scientific American, 256, 119 –125. Farmer, H. S., Wardrop, J. L., & Rotella, S. C.
Cole, S. (1979). Age and scientific performance. (1999). Antecedent factors differentiating women
American Journal of Sociology, 84, 958 –977. and men in science/nonscience careers. Psychol-
Conway, J. B. (1988). Differences among clinical ogy of Women Quarterly, 23, 763–780.
psychologists: Scientists, practitioners, and sci- Feist, G. J. (1993). A structural model of scientific
ence-practitioners. Professional Psychology: Re- eminence. Psychological Science, 4, 366 –371.
search and Practice, 19, 642– 655. Feist, G. J. (1997). Quantity, impact, and depth of
Costa, P., & McCrae, R. R. (1995). Solid ground in research as influences on scientific eminence: Is
the wetlands of personality: A reply to Block. quantity most important? Creativity Research
Psychological Bulletin, 117, 216 –220. Journal, 10, 325–335.
Costa, P. T., McCrae, R. R., & Holland, J. L. (1984). Feist, G. J. (1998). A meta-analysis of the impact of
Personality and vocational interests in an adult personality on scientific and artistic creativity.
sample. Journal of Applied Psychology, 69, 390 – Personality and Social Psychological Review, 2,
400. 290 –309.
Crowley, K., Callanan, M. A., Tenenbaum, H. R., & Feist, G. J. (1999). Personality in scientific and artis-
Allen, E. (2001). Parents explain more often to tic creativity. In R. J. Sternberg (Ed.). Handbook of
boys than to girls during shared scientific thinking. Human Creativity. (pp 273–296). Cambridge, En-
Psychological Science, 12, 258 –261. gland: Cambridge University Press.
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 179

Feist, G. J. (2006). The psychology of science and the Gough, H. G. (1961, February). A personality sketch
origins of the scientific mind. New Haven, CT: of the creative research scientist. Paper presented
Yale University Press. at 5th Annual Conference on Personnel and Indus-
Feist, G. J. (in press). The development of scientific trial Relations Research, UCLA, Los Angeles, CA.
talent in Westinghouse finalists and members of Gough, H. G. (1987). California Psychological In-
the National Academy of Sciences. Journal of ventory: Administrators guide. Palo Alto, CA:
Adult Development. Consulting Psychologists Press.
Feist, G. J., & Barron, F. X. (2003). Predicting cre- Gough, H. G., & Bradley, P. (1995). [ACL, CPI, and
ativity from early to late adulthood: Intellect, po- the Big Five Dimensions]. Unpublished raw data.
tential and personality. Journal of Research in Halpern, D. F. (2000). Sex differences in cognitive
Personality, 37, 62– 88. abilities (3rd ed.). Mahwah, NJ: Lawrence Erl-
Feist, G. J., & Gorman, M. E. (1998). Psychology of baum Associates.
science: Review and integration of a nascent dis- Hart, J. J. (1982). Psychology of the scientists: XLVI:
cipline. Review of General Psychology, 2, 3– 47. Correlation between theoretical orientation in psy-
Feist, G. J., Paletz, S., & Weitzer, W. (2005). Pre- chology and personality type. Psychological Re-
dicting scientific interest in college students: The ports, 50, 795– 801.
effects of quantitative skills, gender, self-image, Helmreich, R. L., Spence, J. T., Beane, W. E.,
and personality. Manuscript in preparation. Lucker, G. W., & Matthews, K. A. (1980). Making
Flavell, J. H. (1979). Metacognition and cognitive it in academic psychology: Demographic and per-
monitoring: A new area of cognitive-developmen- sonality correlates of attainment. Journal of Per-
tal inquiry. American Psychologist, 34, 906 –911. sonality and Social Psychology, 39, 896 –908.
Fox Keller, E. (1985). Reflections on gender and Helmreich, R. L., Spence, J. T., & Pred, R. S. (1988).
science. New Haven, CT: Yale University Press. Making it without losing it: Type A, achievement
Freeman, C. (1999). The crystallizing experience: A motivation and scientific attainment revisited. Per-
study in musical precocity. Gifted Child Quar- sonality and Social Psychology Bulletin, 14, 495–
504.
terly, 43, 75– 85.
Helson, R. (1971). Women mathematicians and the
Funder, D. C. (1991). Global traits: A neo-Allportian
creative personality. Journal of Consulting and
approach to personality. Psychological Science, 2,
Clinical Psychology, 36, 210 –220.
31–39.
Helson, R., & Crutchfield, R. (1970). Mathemati-
Gardner, H. (1983). Frames of mind: The theory of
cians: The creative researcher and the average
multiple intelligences. NewYork: Basic Books.
PhD. Journal of Consulting and Clinical Psychol-
Garwood, D. S. (1964). Personality factors related to
ogy, 34, 250 –257.
creativity in young scientists. Journal of Abnormal Helson, R., & Pals, J. (2000). Creative potential,
and Social Psychology, 68, 413– 419. creative achievement, and personal growth. Jour-
Geary, D. C. (1998). Male, female: The evolution of nal of Personality, 68, 1–27.
human sex differences. Washington, DC: Ameri- Holland, J. L. (1992). Making vocational choices
can Psychological Association. (2nd edition). Odessa, FL: Psychological Assess-
Georghiades, P. (2000). Beyond conceptual change ment Resources.
learning in science education: Focusing on trans- Horner, K. L., Rushton, J. P., & Vernon, P. A.
fer, durability and metacognition. Educational Re- (1986). Relation between aging and research pro-
search, 42, 119 –139. ductivity of academic psychologists. Psychology
Gerbing, D. W., & Tuley, M. R. (1991). The 16PF and Aging, 4, 319 –324.
related to the Five-Factor Model in personality: Hyde, J. S., Fennema, E., Ryan, M., Frost, L. A., &
Multiple-indicator measurement versus the a priori Hopp, C. (1990). Gender comparisons of mathe-
scales. Multivariate Behavioral Research, 26, matics attitudes and affect: A meta-analysis. Psy-
271–289. chology of Women Quarterly, 14, 299 –324.
Glynn, S. M., & Muth, K. D. (1994). Reading and Inhelder, B., & Piaget, J. (1958). The growth of
writing to learn science: Achieving scientific liter- logical thinking from childhood to adolescence.
acy. Journal of Research in Science Teaching, 31, (A. Parsons & S. Milgram, Trans.). New York:
1057–1073. Basic Books.
Goldberg, L. R., & Rosolack, T. K. (1994). The Big Jacobwitz, T. (1983). Relationship of sex, achieve-
Five factor structure as an integrative framework: ment, and science self-concept to the science ca-
An empirical comparison with Eysenck’s P-E-N reer preferences of Black students. Journal of Re-
model. In C. F. Halverson Jr., G. A. Kohnstamm, search in Science Teaching, 20, 621– 628.
& R. P. Martion (Eds.). The developing structure John, O. P. (1990). The “Big-Five” factor taxonomy:
of temperament and personality from infancy to Dimensions of personality in the natural language
adulthood (pp. 7–35). Hillsdale, NJ: Erlbaum. and in questionnaires. In L. A. Pervin (Ed.), Hand-
180 FEIST

book of personality theory and research (pp. 66 – McCrae, R. R., Costa, P. T., & Busch, C. M. (1986).
100). New York: Guilford. Evaluating comprehensiveness in personality sys-
Johnson, J. A., Germer, C. K., Efran, J. S., & Over- tems: The California Q-sets and the Five-Factor
ton, W. F. (1988). Personality as the basis for Model. Journal of Personality, 54, 430 – 446.
theoretical predilections. Journal of Personality McCrae, R. R., Costa, P. T., & Piedmont, R. L.
and Social Psychology, 55, 824 – 835. (1993). Folk concepts, natural language, and psy-
Kahn, J. H., & Scott, N. A. (1997). Predictors of chological constructs: The California Psychologi-
research productivity and science-related career cal Inventory and the Five-Factor Model. Journal
goals among counseling psychology doctoral stu- of Personality, 61, 1–26.
dents. Counseling Psychologist, 25, 38 – 67. McCrae, R. R., & John, O. P. (1992). An introduction
Karmiloff-Smith, A. (1992). Beyond modularity: A to the Five-Factor Model and its applications.
developmental perspective on cognitive science. Journal of Personality, 60, 175–215.
Cambridge: MIT Press. Merton, R. K. (1973). The sociology of science:
Kenrick, D. T., & Funder, D. C. (1988). Profiting Theoretical and empirical investigations. Chicago:
from controversy: Lessons from the person-situa- Chicago University Press.
tion debate. American Psychologist, 43, 23–34. Mischel, W. (1968). Personality and assessment.
Kimura, D. (1999). Sex and cognition. Cambridge: New York: Wiley.
MIT Press. Mischel, W., & Shoda, Y. (1999). Integrating dispo-
Klahr, D. (2000). Exploring science: The cognition sitions and processing dynamics within a unified
and development of discovery processes. Cam- theory of personality: The Cognitive-Affective
bridge: MIT Press. Personality System. In L. A. Pervin and O. P. John
Kuhn, D. (1989). Children and adults as intuitive (Eds.). Handbook of personality theory and re-
scientists. Psychological Review, 964, 674 – 689. search (pp. 197–218). New York: Guilford Press.
Kuhn, D. (1993). Connecting scientific and informal National Science Foundation (1999). Women, minor-
reasoning. Merrill-Palmer Quarterly, 39, 74 –103. ities, and persons with disabilities in science and
Kuhn, D., E. Amsel, & M. O’Loughlin. (1988). The
engineering: 1998 (NSF 99 – 87.) Arlington, VA:
development of scientific thinking skills. Orlando
National Science Foundation.
FL: Academic.
Nisbett, R. E., & Ross, L. (1980). Human inference:
Kuhn, D., & Pearsall, S. (2000). Developmental or-
Strategies and shortcomings of social adjustment.
igins of scientific thinking. Journal of Cognition
New York: Prentice-Hall.
and Development, 1, 113–129.
Nosek, B. A., Banaji, M. R., & Greenwald, A. G.
Lehman, H. C. (1953). Age and achievement. Prince-
(2002). Math ⫽ male, me ⫽ female, therefore
ton, NJ: Princeton University Press.
Lehman, H. C. (1966). The psychologist’s most cre- math ⫽ me. Journal of Personality and Social
ative years. American Psychologist, 21, 363–369. Psychology, 83, 44 –59.
Lippa, R. (1998). Gender-related individual differ- O’Brien, V., Martinez-Pons, M., & Kopala, M.
ences and the structure of vocational interests: The (1999). Mathematics self-efficacy, ethnic identity,
importance of the people-things dimension. Jour- gender, and career interests related to mathematics
nal of Personality and Social Psychology, 74, and science. Journal of Educational Research, 92,
996 –1009. 231–235.
Long, J. S. (Ed.) (2001). From scarcity to visibility: Over, R. (1982). Is age a good predictor of research
Gender differences in the careers of doctoral sci- productivity? Australian Psychologist, 17, 129 –
entists and engineers. Washington, DC: National 139.
Academy Press. Over, R. (1989). Age and scholarly impact. Psychol-
Mallinckrodt, B., Gelso, C. J., & Royalty, G. M. ogy and Aging, 4, 222–225.
(1990). Impact of the research training environ- Parloff, M. B., & Datta, L. (1965). Personality char-
ment and counseling psychology students’ Holland acteristics of the potentially creative scientist. Sci-
personality type on interest in Research Profes- ence and Psychoanalysis, 8, 91–105.
sional Psychology: Research and Practice, 21, Piaget, J. (1952). The child’s concept of number.
26 –32. New York: Norton.
McCrae, R. R. (1991). The Five-Factor Model and its Piaget, J. (1972). Intellectual evolution from adoles-
assessment in clinical settings. Journal of Person- cence to adulthood. Human Development, 15,
ality Assessment, 57, 399 – 414. 1–12.
McCrae, R. R., & Costa, P. T. (1985). Comparison of Piedmont, R. L., McCrae, R. R., & Costa, P. T.
EPI and psychoticism scales with measures of the (1991). Adjective check list scales and the Five-
Five-Factor Model of personality. Personality and Factor Model. Journal of Personality and Social
Individual Differences, 6, 587–597. Psychology, 60, 630 – 637.
 SPECIAL ISSUE: DEVELOPMENTAL AND PERSONALITY INFLUENCES 181

Portes, A., & Rumbaut, R. G. (2001). Legacies: The Simonton, D. K. (1991). Career landmarks in sci-
story of the immigrant second generation. Berke- ence: Individual differences and interdisciplinary
ley, CA: University of California Press. contrasts. Developmental Psychology, 27, 119 –
Prediger, D. J. (1982). Dimensions underlying Hol- 130.
land’s hexagon: Missing link between interest and Simonton, D. K. (2000). Methodological and theo-
occupations? Journal of Vocational Behavior, 21, retical orientation and the long-term disciplinary
259 –287. impact of 54 eminent psychologists. Review of
Reis, S. M., & Park, S. (2001). Gender differences in General Psychology, 4, 13–21.
high-achieving students in math and science. Jour- Smithers, A. G., & Batcock, A. (1970). Success and
nal for the Education of the Gifted, 25, 52–73. failure among social scientists and health scientists
Reskin, B. F. (1977). Scientific productivity and the at a technological university. British Journal of
reward structure of science. American Sociological Educational Psychology, 40, 144 –153.
Review, 42, 491–504. Sperber, D. (1994). The modularity of thought and
Roe, A. (1952). The making of a scientist. New York: epidemiology of representations. In L. A. Hir-
Dodd, Mead. schfeld & S. A. Gelman (Eds.). Mapping the mind:
Roe, A. (1953). A psychological study of eminent Domain specificity in cognition and culture (pp.
psychologists and anthropologists, and a compar- 39 – 67). Cambridge, England: Cambridge Univer-
ison with biological and physical scientists. Psy- sity Press.
chological Monographs: General and Applied, 67, Sternberg, R. J. (1985). Beyond IQ: A triarchic the-
1–55. ory of human intelligence. Cambridge, England:
Roe, A. (1965). Changes in scientific activities with Cambridge University Press.
age. Science, 150, 313–318. Stumpf, H., & Stanley, J. C. (2002). Group data on
Rosenberg, E. L. (1998). Levels of analysis and the high school grade point averages and scores on
organization of affect. Review of General Psychol- academic aptitude tests as predictors of institu-
ogy, 2, 247–270. tional graduate rates. Educational and Psycholog-
Rosser, S. (Ed.) (1988). Feminism within the science ical Measurement, 62, 1042–1052.
and healthcare professions: Overcoming resis- Subotnik, R. F., Duschl, R. A., & Selmon, E. H.
tance. Exeter, England: A. Wheaton and Co. (1993). Retention and attrition of science talent: A
Royalty, G. M., & Magoon, T. M. (1985). Correlates longitudinal study of Westinghouse Science Talent
of scholarly productivity among counseling psy- Search winners. International Journal of Science
chologists. Journal of Counseling Psychology, 32, Education, 15, 61–72.
458 – 461. Subotnik, R. F., & Steiner, C. L. (1994). Adult man-
Rushton, J. P., Murray, H. G., & Paunonen, S. V. ifestations of adolescent talent in science: A lon-
(1983). Personality, research creativity, and teach- gitudinal study of 1983 Westinghouse Science Tal-
ing effectiveness in university professors. Sciento- ent Search winners. In R. Subotnik & K. D. Arnold
metrics, 5, 93–116. (Eds.). Beyond Terman: Contemporary longitudi-
Ryan, J. (2005, March 3). Brains of men and women nal studies of giftedness and talent. Creativity re-
only part of the story in science. San Francisco search. (pp. 52–76). Norwood, NJ: Ablex Publish-
Chronicle. (also found on the Internet at http:// ing Corp.
www.sfgate.com/cgi-bin/article.cgi?file⫽/chronicle/ Sulloway, F. (1996). Born to rebel: Birth order,
archive/2005/03/03/BAGSKBJI981.DTL&type⫽ family dynamics, and creative lives. New York:
printable). Pantheon.
Schwanenflugel, P. J., Stevens, T. P. M., & Carr, M. Summers, L. (2005, January 14). Remarks at NBER
(1997). Metacognitive knowledge of gifted chil- conference on diversifying the science and engi-
dren and nonidentified children in early elementary neering workforce. Retrieved on March 28, 2005,
school. Gifted Child Quarterly, 41, 25–35. from http://www.president.harvard.edu/speeches/
Shore, B. M., & Dover, A. C. (1987). Metacognition, 2005/nber.html.
intelligence, and giftedness. Gifted Child Quar- Van Zelst, R. H., & Kerr, W. A. (1954). Personality
terly, 31, 37–39. self-assessment of scientific and technical person-
Simonton, D. K. (1988a). Scientific genius: A psy- nel. Journal of Applied Psychology, 38, 145–147.
chology of science. Cambridge, England: Cam- Waller, N. L., Lykken, D. T., & Tellegen, A. (1995).
bridge University Press. Occupational interests, leisure time interests, and
Simonton, D. K. (1988b). Age and outstanding personality: Three domains or one? Findings from
achievement: What do we know after a century of the Minnesota Twin Registry. In D. J. Lubinski
research? Psychological Bulletin, 104, 251–267. and R. V. Dawis (Eds.). Assessing individual dif-
Simonton, D. K. (1990). Creativity in the later years: ferences in human behavior: New concepts, meth-
Optimistic prospects for achievement. Gerontolo- ods, and findings (pp. 233–259), Palo Alto, CA,
gist, 30, 626 – 631. Davies-Black Publishing, Inc.
 182 FEIST

Webb, R. M., Lubinski, D., & Benbow, C. P. (2002). Zachar, P., & Leong, F. T. L. (2000). A 10-year
Mathematically facile adolescents with math-sci- longitudinal study of scientists and practitioner
ence aspirations: New perspectives on their educa- interests in psychology: Assessing the Boulder
tional and vocational development. Journal of Ed- model. Professional Psychology: Research and
ucational Psychology, 94, 785–794. Practice, 31, 575–580.
White, B. Y., & Frederiksen, J. R. (1998). Inquiry, Zimmerman, C. (2000). The development of scien-
modeling, and metacognition: Making science ac- tific reasoning skills. Developmental Review, 20,
cessible to all students. Cognition & Instruc- 99 –140.
tion, 16, 3–118. Zuckerman, H. (1996). Scientific elite: Nobel laure-
Zachar, P., & Leong, F. T. L. (1992). A problem of ates in the United States. (2nd edition). New
personality: Scientist and practitioner differences in Brunswick, NJ: Transaction Publishers.
psychology. Journal of Personality, 60, 667– 677.
Zachar, P., & Leong, F. T. L. (1997). General versus
specific predictors of specialty choice in psychol-
ogy: Holland codes and theoretical orientations. Received January 30, 2006
Journal of Career Assessment, 5, 333–341. Accepted February 8, 2006 䡲

View publication stats