The Origins and Ends of Giftedness

Ellen Winner
Boston College and Project Zero, Harvard Graduate School of Education

Five issues about giftedness are discussed. First, the ori- current state of knowledge about giftedness, focusing on
gins of giftedness are explored. The view that gifiedness is five issues. The first and major focus of this article con-
entirely a product of training is critiqued. There is indirect cerns what is known about the origins and causes of gift-
evidence for atypical brain organization and innate talent edness. I discuss and critique the view that giftedness is
in gifted children: Many gifted children and savants have entirely a product of training and practice and argue that
enhanced right-hemisphere development, language-related there is indirect evidence for atypical brain organization
difficulties, and autoimmune disorders. Second, the intense and innate talent in gifted children. The same claim is made
motivation of gifted children is discussed. Third, it is ar- as well for savants, autistic and retarded individuals with
gued that gifted children have social and emotional diffi- extremely high levels of ability, usually in only one area. I
culties that set them apart. Fourth, evidence for the often argue that the indirect evidence suggests that many gifted
uneven cognitive profiles of such children is presented. children and savants have enhanced right-hemisphere de-
Finally, the relationship between childhood g(ftedness and velopment, concomitant language-related difficulties, and
"domain" creativity in adulthood is discussed. Few gifted autoimmune disorders.
children go on to become adult creators because the skills In the second section, I briefly discuss the motiva-
and personality factors required to be a creator are very tional characteristics of gifted children, showing that these
different from those typical of even the most highly gifted children are far more intrinsically driven than are average
children. children. Third, I discuss the particular social and emo-
tional difficulties of gifted children that set them apart from
others. Fourth, I describe what is known about the often
uneven cognitive profiles of gifted children.

G ifted children and prodigies display near-adult

Plevel skills and interests. They may begin to read
fluently at the age of three or four, without any
extended instruction; they may play a musical instrument
as skillfully as a highly trained adult; they may turn every-
I conclude by proposing three ways to think about the
"ends" of giftedness. The end of giftedness has a deliber-
ately triple meaning. I use this term in a positive sense to
refer to the adult endpoint of the development of a gifted
child. I use it in a negative sense to refer to the potential
day experiences into mathematical problems to play with,
disappearance of giftedness after childhood. Finally, I use
moving from arithmetic to algebra before their peers have
it to refer to the goals I believe we should expect gifted
learned to carry numbers in addition (cf. Feldman, 1991;
children in particular to fulfill if we are to give them extra
Radford, 1990; Winner, 1996a).
societal resources.
Psychologists have always been interested in the de-
viant. As a result we know much more about deviance at Origins and Causes of Giftedness
the negative than at the positive end of the spectrum. Just
as we know more about depression and fear than we do The Disputed Role of Training
about happiness and courage, we also know far more about
The topic of giftedness inevitably awakens the nature-
retardation and learning problems than we do about gift-
nurture debate. Most people in our culture subscribe to the
edness. Research on retardation is more advanced and more
nativist view of giftedness, in which giftedness is believed
integrated into the field of psychology than is research on
to be a product of inborn high ability. However, this folk
giftedness. Research on retardation is more likely to find its
psychology view of the origins of giftedness has recently
way into mainstream developmental journals than is re-
come under sharp attack by psychologists who argue that
search on giftedness, which is often to be found in special-
giftedness (in any domain) is entirely a product of what is
ized and hence less widely read journals. This phenomenon
referred to as goal-directed hard work, or deliberate prac-
is part of the wider phenomenon of psychology's focus on
the pathological rather than the healthy. It also surely
reflects the view that retardation is a problem researchers Ellen Winner, Department of Psychology, Boston College, and Project
may eventually learn to alleviate, whereas gifts are privi- Zero, Harvard Graduate School of Education.
leges to be admired or envied rather than problems in need Some of the research reported here was supportedby a grant from the
of solutions. International Dyslexia Association. I thank Julian Stanley for his insight-
ful comments on an earlier version of this article.
In previous work I have examined the myths and Correspondence concerningthis article should be addressed to Ellen
misconceptions held by laypeople and psychologists alike Winner, Department of Psychology, Boston College, McGuinn Hall, 140
about the gifted (Winner, 1996a). Here I consider the Commonwealth Avenue, Chestnut Hill, MA 02467-3807.

January 2000 • American Psychologist 159

Copyright 2000 by the American Psychological Association. Inc. 0003-066X/00/$5.00
Vol. 55, No. I. 159-169 DOE IO. I0371IOlN)3-O66X.55,1.159
 ever, does this mean that hard work is all that is needed or
that anyone can engage in the kind of hard work that will
lead to Newton's creative breakthroughs?
Second, Roe (1951, 1953a, 1953b) found that out-
standing achievement in science was predicted by the par-
ticipants' capacity for endurance, concentration, and com-
mitment rather than their level of intellectual ability. How-
ever, Roe's scientists were all high in intellectual ability to
begin with. Her studies thus show that high ability is not
sufficient for exceptional achievement; rather, one needs
both high ability and perseverance.
Third, Bloom (1985) showed that eminent adults in a
variety of domains did not achieve high levels of perfor-
mance without a long and intensive period of training.
Their training began in early childhood with warm and
loving teachers, who were then supplanted by more de-
manding and rigorous master teachers. Bloom's study
might be taken as evidence that the high levels of achieve-
ment attained were entirely the result of the rigorous train-
ing. However, a careful look at the descriptions of these
Ellen Winner eminent individuals as children shows that at a very young
Photo by Jerry Bauer age, prior to any regimen of training or deliberate practice,
signs of unusual ability were present. The musicians were
described as quick to learn the piano, and both their parents
and their teachers recognized they were special. The sculp-
tice (Ericsson, Krampe, & Tesch-Romer, 1993; Ericsson & tors said that they drew constantly as children, usually
Lehman, 1996; Howe, Davidson, & Sloboda, 1998; Howe, realistically. The mathematicians recalled being obsessed
Davidson, Moore, & Sloboda, 1995; Sloboda, Davidson, & with gears, valves, gauges, and dials and were considered
Howe, 1994). Ericsson et al. (1993) showed that levels of "brilliant" as children. Thus, Bloom's work, like that of
expertise in piano, violin, chess, bridge, and athletics cor- Roe (1951, 1953a, 1953b), allows us to conclude only that
relate directly with the amount of deliberate practice. They intensive training is necessary for the acquisition of exper-
also argued that there is no systematic and verifiable evi- tise; it does not sufficiently explain children's high level of
dence for high abilities emerging prior to extensive periods achievement.
of deliberate practice. They discounted as unreliable anec- The same criticism can be leveled at the work of
dotal reports about the childhood feats of prodigies such as Ericsson and his colleagues (Ericsson et al., 1993). Hard
Mozart, Gauss, and Menuhin. work and innate ability have not been unconfounded.
Ericsson and Faivre (1988) have also sought to ac- Those children who have the most ability are also likely to
count for savant gifts in terms of deliberate practice. Sa- be those who are most interested in a particular activity,
vants are individuals who are retarded (with IQs between who begin to work at that activity at an early age, and who
40 and 70) and also either are autistic or show autistic work the hardest at it. Ericsson's research demonstrated the
symptoms. Savants are typically found in the domains of importance of hard work but did not rule out the role of
arithmetic calculation, music, and realistic drawing, and innate ability.
they often surpass child prodigies in their level of achieve- Although Ericsson and his colleagues (Ericsson et al.,
ment. For instance, at ages three and four, the drawing 1993) consider the stories of early (pretraining) achieve-
prodigy Nadia drew more realistically than any known ments of child prodigies to be unreliable, there are simply
"normal" child prodigy at the same age (Selfe, 1977). too many such reports that are too consistent with one
Savants work obsessively in their area of ability, and it is another for them to be easily discounted. In addition, these
the countless hours they spend drawing, doing mental reports come not only from potentially biased parents but
calculation, or playing an instrument that have led to the also from careful case studies of young prodigies (cf.
suggestion that the savant's skills are the product of delib- Feldman, 1991; Milbrath, 1998; Winner, 1996a). If excep-
erate practice. tional abilities emerge prior to intensive instruction and
Consistent with this contemporary nurture view of training, then these abilities are likely to reflect atypical,
giftedness are several other earlier findings. First, case innate potential.
studies of creative people such as those by Csikszentmi- The claim that savants achieve their astonishing level
halyi (1996), Gardner (1993a), and Gruber (1981) show of performance because they have practiced their skill for
that all great achievement is associated with years of deep countless hours leaves unexplained the fact that, like gifted
and prolonged work. For example, Gruber (1986) pointed children, savants show extremely high abilities right from
out that it took Newton 20 years to go from his preliminary the start, before they have spent much time working at their
ideas to his magnum opus, Principia Mathematica. How- gift. In addition, this claim cannot explain why savants are

160 January 2000 • American Psychologist

found only in domains that are highly rule governed and the model is inconsistent. However, whether or not the
structured rather than in looser domains such as higher Geschwind and Galaburda model best accounts for the
mathematics, abstract painting, philosophy, or creative associations just described, we cannot discount the exis-
writing. Thus, it seems more likely that savants and gifted tence of these associations, which suggest gifted children,
children owe their gifts at least in part to innate abilities child prodigies, and savants are not made from scratch but
that in turn reflect atypical brain organization. Recently, are born with unusual brains that enable rapid learning in a
Miller (1999) has made the same point: The motivation of particular domain.
savants may be the result rather than the cause of high
ability. The Role of Families
Indirect evidence indicates that gifted children and
savants have atypical brain organization (whether as a The notion that giftedness is a product of intensive training
result of genetics, the in utero environment, or after-birth reflects an overly optimistic view of the power of nurture.
trauma). First, giftedness in mathematics, visual arts, and A more negative view of the power of nurture is reflected
music is associated with superior visual-spatial abilities, in another common claim: that gifted children are created
and children with mathematical gifts show enhanced brain by driving, overambitious parents, There is concern that the
activity in their fight hemisphere when asked to recognize end result of such extreme pushing will be disengagement,
faces, a task known to involve the fight hemisphere bitterness, and depression. Parents of gifted children are
(O'Boyle, Alexander, & Benbow, 1991; O'Boyle & Ben- advised to let their children have a normal childhood.
bow, 1990). Thus, giftedness in these domains may involve However, most gifted children do not become bitter
enhanced right-hemisphere development. Second, individ- and disaffected. Moreover, it is impossible to drive an
uals with gifts in mathematics, visual arts, and music are ordinary child to the kinds of high achievements seen in
disproportionately nonright-handed. Again, this finding gifted children. In addition, gifted children typically report
suggests atypical brain organization, because nonright- that their family played a positive, not a negative, role in
handedness is a rough index of anomalous brain dominance their development (Van Tassel-Baska, 1989). Today we
(Annett, 1985; Geschwind & Galaburda, 1987).Third, stud- know quite a bit about the family characteristics of gifted
ies have shown that mathematically and musically gifted children, at least of those in our society. These character-
individuals have a more bilateral, symmetrical brain orga- istics are positive ones, as described below, but the research
nization than is usual, with the right hemisphere participat- does not allow us to conclude that particular family char-
ing in tasks ordinarily reserved for the left hemisphere acteristics play a causal role in the development of gifted-
(Gordon, 1970, 1978, 1980; Hassler & Birbaumer, 1988; ness. There are two reasons why no causal conclusions can
O'Boyle, Gill, Benbow, & Alexander, 1994). Fourth, gift- be drawn from the existing data. First, there is the lack of
edness in spatial areas is accompanied by a disproportion- relevant control groups. Second, if causality exists, its
ate incidence of language-related learning disorders such as direction could be either from parent to child or from child
dyslexia, a finding reported for artists (Winner & Casey, to parent.
1993; Winner, Casey, DaSilva, & Hayes, 1991), inventors The families of gifted children are child centered,
(Colangelo, Assouline, Kerr, Huesman, & Johnson, 1993), meaning that family life is often totally focused on the
and musicians (Hassler, 1990), Finally, youths with very child's needs (e.g., Freeman, 1979; Winner, 1996a). How-
high IQs have an increased incidence of autoimmune prob- ever, the fact that parents spend a great deal of time with
lems (Benbow, 1986; Hildreth, 1966; Kolata, 1983; Tem- their gifted children does not mean that they create the gift.
ple, 1990); a link between immune disorders and giftedness It is likely that parents first notice signs of exceptionality
in music has been suggested but not firmly established and then respond by devoting themselves to the develop-
(Hassler & Birbaumer, 1988; McNamara, Flannery, Obler, ment of their child's extraordinary ability. Of course, we
& Schachter, 1994; but see Hassler & Gupta, 1993). In have no information on the number of child-centered fam-
summary, giftedness in abilities that involve the right hemi- ilies that do not produce gifted children (i.e., the control-
sphere may be associated with enhanced right-hemisphere group problem).
development and hence with anomalous brain dominance. Gifted children typically grow up in enriched family
Individuals with such gifts are more likely to be nonright- environments with a high level of intellectual or artistic
handed, to have language represented bilaterally, and to stimulation (e.g., Csikszentmihalyi, Rathunde, & Whalen,
have language-related and immune system disorders. 1993; Freeman, 1979; Gottfried, Gottfried, Bathurst, &
These disparate signs and symptoms are accounted for Guerin, 1994; Moss, 1990). Of course, these findings are
by Geschwind and Galaburda's (1987) theory of the pa- correlational. We cannot conclude that stimulation and
thology of superiority, in which an association between enrichment lead to the development of giftedness. First,
spatial (right-hemisphere) gifts, linguistic (left-hemisphere) gifted children may need an unusual amount of stimulation
deficits, nonright-handedness, and immune disorders is ar- and may demand enriched environments, a demand to
gued to be a consequence of testosterone-induced alter- which their parents respond. Thus, the child's inborn ability
ations of the fetal brain. Testosterone is argued to inhibit could be the driving force, leading the child to select
some areas of the brain while stimulating other areas. This enriched environments (cf. Scarr & McCartney, 1983).
theory has come under sharp attack (cf. Bryden, McManus, Again, how many children of enriched environments dis-
& Bulman-Fleming, 1994), and the evidence in support of play no signs of giftedness?

January 2000 • American Psychologist 161

 Parents of gifted children typically have high expec- most European and East Asian nations (Mullis et al., 1998).
tations and also model hard work and high achievement The gap between American students and others is greatest
themselves (Bloom, 1985; Csikszentmihalyi et al., 1993; for those at the highest levels of ability. According to a
Gardner, 1993a). It is logically possible that gifted children widely cited government report, about half of the top one
have simply inherited their gift from their parents, who also percent of our students are underachieving (Ross, 1993).
happen to be hardworking achievers. Parents of children in Any educational solution for the gifted should be
performance domains like music and athletics are the most made in the context of educational reform for all students
directive; parents of children in the visual arts are the least (Winner, 1996a, 1996b, 1997a). Standards and expecta-
directive; parents of children gifted in an academic domain tions are not only too low for the gifted, they are also far
fall somewhere in between (Bloom, 1985). To achieve in a too low for the rest of our students. If our schools were as
performance domain, one must submit to rigorous and rigorous as those in Western Europe and East Asia, then
early training; even the most gifted children might not stick many of our moderately gifted students, who are currently
to such a rigorous schedule without a directive parent who bored, tuning out, and underachieving, would be appropri-
insists that time be spent on practice. Families of children ately challenged. Those still not challenged enough ought
gifted in the visual arts may be the least directive because to be able to take advanced classes in their domain of gift.
of the low value our culture places on being an artist. Such advanced classes exist in high schools (e.g., advanced
Parents of gifted children grant their children more placement courses), but they should be available at all
than the usual amount of independence (Colangelo & Dett- levels.
man, 1983; Karnes, Schwedel, & Steinberg, 1984; Terman Schools are not the only agents that should hold gifted
& Oden, 1947). However, we do not know whether grant- children to high standards. Parents also play a critical role.
ing independence leads to high achievement, or whether it Too often parents fear pushing their children too hard.
is the recognition of the child's gift that leads to the They fear they may rob their children of a normal child-
granting of independence. It is also possible that gifted hood if they make them work too much and instead allow
children are particularly strong willed and single-minded their children unlimited access to television, video games,
and thus demand independence. malls, and such (Damon, 1995). To be sure, many gifted
Gifted children who grow up in "complex" families-- children are so driven than they focus on work in the area
those that combine both stimulation and nurturance--are of their ability or talent whether or not their parents push
happier, more alert, more engaged, and more goal directed them to do so. However, we do not know how many more
than are gifted children who grow up in families with only high-potential children never develop their ability because
one or neither of these traits (Csikszentmihalyi et al., they are not challenged but are instead captured by the
1993). Gifted children from complex families report more potent messages from their peer culture to avoid work and
states of flow and high energy and were rated by teachers be like everyone else.
as original, independent, and working up to their potential. Similarly, although we cannot conclude that any par-
Youths who dropped out of their domains of talent reported ticular kind of family can create a gifted child, the corre-
having parents who were either too directive or too unin- lational findings reported by Csikszentmihalyi et al. (1993)
volved. However, we do not know whether a combination certainly suggest that given a high-potential child, certain
of stimulation and nurturance causes gifted children to kinds of family constellations are most likely to succeed in
remain engaged, because we do not know whether the maintaining and nurturing the gift. Although it is not
parents are reacting to the child, or the child to the parents. proven that complex families, which combine nurturance
and stimulation, are causally implicated in maintaining and
Implications for Education and Child Rearing developing a gift, it is also not proven that they are n o t
Research on the nurture hypothesis has failed to demon- causally implicated. Unless we have evidence to show that
strate that giftedness is a product of hard work and inten- family environment plays no causal role, it seems prudent
sive training or that any particular kind of family environ- to urge families to strive to combine the two qualities most
ment causes giftedness. Nonetheless, the research just re- typically associated with gifted youths who remain en-
viewed has implications for the nurturance of giftedness. gaged in their area of ability.
To be sure, no research has demonstrated that hard work,
perseverance, and practice is sufficient to explain the exis- Motivational Aspects of Giftedness
tence of giftedness. Yet, these qualities have been shown to Gifted children have a deep intrinsic motivation to master
be necessary for high achievement, because we have no the domain in which they have high ability and are almost
documented cases of high achievement reached in the manic in their energy level (Winner, 1996a, 1997b). Often
absence of long training and many cases showing the one cannot tear these children away from activities in their
association of high achievement with training. area of giftedness, whether they involve an instrument, a
Thus, parents and schools ought to hold and model computer, a sketch pad, or a math book. These children
high expectations if gifted children are to reach their po- have a powerful interest in the domain in which they have
tential. All too often, American schools do not sufficiently high ability, and they can focus so intently on work in this
challenge their students. International comparisons show domain that they lose sense of the outside world. They
that American children, no matter what their ability level, combine an obsessive interest with an ability to learn easily
perform below the levels of comparably aged students in in a given domain. Unless social and emotional factors

162 January 2000 • American Psychologist

interfere, this combination leads to high achievement. This students on a variety of scales. Teachers may welt have
intrinsic drive is part and parcel of an exceptional, inborn been subject to a halo effect, perceiving the students they
giftedness. had nominated as gifted as being generally better on all
This "rage to master" characterizes children we have dimensions. In addition, because almost a third of the
traditionally labeled gifted: children with high IQs who Terman children were drawn from professional, middle-
excel in school. It also characterizes children we have class families, giftedness was confounded with social class.
traditionally classified as talented, children who excel in Another early researcher of gifted children, Leta
art, music, or athletics. Hollingworth, argued that children with profoundly high
The distinction in terminology between gifted and IQs (over 180) had special social and emotional problems
talented suggests two different subtypes of children, but (Hollingworth, 1942). In a more recent report it was esti-
this is a distinction with no basis. No matter what the mated that the rate of social and emotional difficulties
domain, gifted or talented children show a rage to master: experienced by profoundly academically gifted children is
Musically gifted children spend hours voluntarily working about twice the rate found among the nongifted, with
at their instrument, artistically gifted children draw when- almost a quarter of such children having such difficulties
ever they are allowed, just as mathematically gifted chil- (Janos & Robinson, 1985). Extreme levels of giftedness
dren willingly spend their time solving existing math prob- lead to isolation. Hence, in middle childhood profoundly
lems and discovering new ones (Winner, 1996a, 1997b). gifted children may try to hide their abilities in the hopes of
The intense drive characterizing gifted children should becoming more popular. Academically gifted girls are
be recognized, celebrated, and cultivated, not destroyed. more apt to do this than boys, and such girls report more
When children are not sufficiently challenged in school, as depression, lower self-esteem, and more psychosomatic
so often happens to gifted children, they sometimes lose symptoms than do academically gifted boys (Gross, 1993).
their motivation and become underachievers. When parents Teenagers with gifts in the visual arts, music, and
and schools try to force single-minded, driven children to athletics have as many difficulties with their peers as do
be well-rounded by curtailing activity in the children's those gifted in academic areas (Csikszentmihalyi et al.,
domain of giftedness and having the children spend time on 1993). These teenagers have been shown to be atypical
more "normal" activities, they may end up stifling the socially and emotionally in a number of respects: They are
children's drive. All children, not only the gifted, would be highly driven, nonconforming, and independent thinkers.
better educated if teachers sought to find out what moti- Gifted children in all domains also tend to be intro-
vates and excites individual students and then harnessed
verted. They spend more time alone than do ordinary
this drive toward learning.
adolescents. They gain stimulation from themselves more
The Social and Emotional Lives of than from others and report liking solitude far more than do
Gifted Children most other people (Csikszentmihalyi et al., 1993). Gifted
children are not only solitary because of their rich inner
The study of giftedness began in earnest in the early part of
lives, but also because solitude is requisite for the devel-
this century, when Lewis Terman initiated a large-scale
opment of their talent. Whereas ordinary children come
longitudinal study of over 1,500 high-IQ children. The first
home after school to play, gifted children come home after
volume about this group appeared in 1925 (Terman, 1925),
school eager to paint, play music, work on math problems,
a 40-year follow-up appeared in 1968 (Oden, 1968), and a
volume describing the survivors in their 80s appeared in read, or write.
1995 (Holahan & Sears, 1995). Terman's goal was to Despite liking solitude more than do ordinary chil-
dispel the myth that gifted children are maladjusted and dren, gifted adolescents also report a preference to be with
emotionally troubled. Terman tried to use his evidence to others rather than alone (Csikszentmihalyi et al., 1993).
show that the participants in his study were, in his words, Thus, although they gain more from solitude than do oth-
"superior to unselected children in physique, health and ers, they still yearn for peer contact. It is difficult for these
social adjustment; [and] marked by superior moral attitudes atypical children to find like-minded peers.
as measured by character tests of trait ratings" (Subotnik & The desire for like-minded peers is one of the stron-
Arnold, 1994, pp. 17-18). gest arguments for placing gifted children in advanced
To understand how Terman came to this conclusion, it classes. Advanced classes for gifted students are almost
is necessary to understand how the children were selected nonexistent at the elementary level, infrequent at the mid-
for the study. The first cut came from teacher nominations dle school level, and common at the secondary level. Such
of the brightest children and also the youngest children in opportunities come in the form of honors classes, advanced
their classes. Nominated students who scored in the top one placement classes, and college-level courses. Because
percent of the school population on an intelligence test meta-analyses of research show that ability grouping helps
were then admitted to the study. For students under high students academically and does not harm them socially
school age, this meant a score of at least 140 on the (Kulik & Kulik, 1997), schools should be increasing their
Stanford-Binet IQ test; for high school students, this meant offerings of advanced coursework and allowing such
a score within the top one percent on the Terman Group courses even at the elementary school level. Yet, all too
Test of Mental Ability. Personality and social and emo- often today schools are disbanding such offerings in the
tional adjustment were assessed by asking teachers to rate name of egalitarianism.

January 2000 • American Psychologist 163

 Advanced courses also exist in summer or weekend Lubinski, and Benbow (1996) found that 42% of students
programs at many universities around the country (Stanley, scoring in the top 0.5% on the SATs had math and verbal
1988). Since 1979, over 100,000 students have participated SAT scores over one standard deviation apart, whereas
in programs across the country now run by the Institute for 72% of students scoring in the top .01% had such a differ-
the Academic Advancement of Youth at the Johns Hopkins entiated profile. When the personal interests of the individ-
University (Johns Hopkins University, 1999). Students are ual children were added into the same equation, 82% and
selected on the basis of a high Scholastic Aptitude Test 94%, respectively, had differentiated profiles. Some math-
(SAT) or American College Testing (ACT) score earned as ematically gifted children identified by the Study of Math-
early as late elementary school. (Students also participate ematically Precocious Youth (SMPY; Stanley, 1988) are
in various annual regional talent searches based on the more gifted in math than verbal ability, although extreme
same model. In some of these talent searches students all cases of such discrepancies are not typical (J. C. Stanley,
the way down to the second grade are tested using down- personal communication, January 7, 1999). For further
ward extensions of the SAT and the ACT.) Students take evidence of unevenness of abilities associated with gifted
courses in their area of high ability, and they find the IQs, see Benbow and Minor (1990); Lewis (1985); Muel-
experience to be very positive, particularly because meet- ler, Dash, Matheson, and Short (1984); and Silver and
ing like-minded peers means they feel less isolated (Ben- Clampit (1990).
bow & Lubinski, 1997; Enersen, 1993). There are now It is not surprising that unevenness exists, because the
about a dozen residential state-supported high schools for abilities that underlie mathematical giftedness differ
the gifted, as well as an equal number of residential early- sharply from those that underlie verbal giftedness. Under-
entrance-to-college programs; these make it possible for lying mathematical but not verbal giftedness are spatial
highly gifted children to mix with equally gifted peers abilities (Benbow & Minor, 1990; Benbow, Stanley, Kirk,
(Boothe, Sethna, & Stanley, in press). & Zonderman, 1983; Casey & Brabeck, 1989; Gardner,
1983; Hermelin & O'Connor, 1986; Krutetskii, 1976).
Cognitive Profiles of Gifted and Mathematically gifted children show stronger recall for
Prodigious Children numerical and spatial information than for linguistic infor-
Psychologists typically assess academic giftedness with an mation, whereas verbally gifted children show the reverse
IQ test that yields subtest scores as well as an overall, pattern (Dark & Benbow, 1991).
global number. Children are usually defined as gifted if Jagged profiles also characterize children gifted in
their global IQ score rises above some arbitrary cutoff point music and art. A gift in music or art can exist alongside an
(often 130). The assumption underlying the use of a global average or even a subnormal IQ. Correlations between
score is that academically gifted children are generally musical ability and IQ are positive but low: Above an
gifted in all academic subjects. Some children justify this average IQ, intelligence is not particularly predictive of
assumption perfectly by demonstrating giftedness in read- musical ability. In the same vein, high musical ability is not
ing, math, and logical analytic thinking. These kind of predictive of a high IQ (Shuter-Dyson, 1982). Further, the
children are notationally gifted, able to master rapidly the existence of musical savants, individuals who are retarded
two kinds of notational symbol systems valued in school: and autistic but who have exceptional musical ability,
language and numbers. shows decisively that a high IQ is not a necessary compo-
Although globally gifted children certainly exist, nent of giftedness in music (Miller, 1999; Treffert, 1989).
many other academically gifted children present a much Yet, musically gifted children typically do very well
less balanced picture; unevenness between verbal and academically (Csikszentmihalyi et al., 1993). One possible
mathematical abilities may be the rule, not the exception. explanation for this conflicting set of findings is that all that
Many of Terman's participants had greater strengths and our knowledge of the relation between music, IQ, and
interests in either reading or math. Terman, however, ar- academic skills comes from studies of children taking
gued that the unevenness in ability among the gifted was no classical music lessons. These children are likely to come
more marked than the unevenness found in the general from educated parents who provide enriched family envi-
population: "The 'one-sidedness' of precocious children is ronments. In addition, such children learn to read music
mythical" (Terman, 1925, p. 339). and practice regularly, two activities that might transfer to
More recent research suggests Terman was wrong. school performance. Whether children who perform rebel-
When assessed with difficult tests without low ceilings, lious antiauthority music (rock, rap, etc.) and who do not
academically gifted children often reveal jagged profiles, read music also do well academically has not been inves-
and a gift in one scholastic area does not imply a gift in tigated, but I speculate that such children would not excel
another area. For example, Detterman and Daniel (1989) in school-related activities.
have found that the higher the IQ, the lower the correlation Children gifted in the visual arts and in athletics
among subtests of the IQ test. Thus, it is more common to typically show a lack of interest in academic achievement,
find mathematical ability far higher than verbal ability in a with those gifted in the visual arts even less committed
high-IQ individual than in a low-IQ individual. Wilkinson academically than those in athletics (Csikszentmihalyi et
(1993) reported sharp discrepancies between verbal and al., 1993). Savants who excel in drawing provide clear
performance IQ scores in children with IQs of 120 or evidence 1or the possible dissociation between giftedness in
higher. In a large-scale study of gifted adolescents, Achter, the visual arts and IQ (Miller, 1999; Treffert, 1989).

164 January 2000 • American Psychologist

 A gift in the visual-spatial area may bring with it a receiving formal training outside of school until or unless
language-based learning disability. Gifted children may they elect to attend an art school. Schools ought to offer
perform at an average level in some academic domain. rigorous and advanced training in the arts as well as in
Others are gifted in one domain and learning disabled in academics so that gifted children can advance in an art
another. Davis and Rimm (1985) estimated that there are form and have their gift taken as seriously as is academic
between 120,000 and 180,000 American schoolchildren ability.
who are both gifted and learning disabled. A recent study
by Reis, Neu, and McGuire (1995) found that all of the The Ends of Giftedness
academically gifted students in a University of Connecticut There are at least three senses in which to consider the ends
program for learning disabled students had a language- of giftedness: (a) the most positive endpoint of childhood
based learning disability. It is certainly not uncommon to giftedness, (b) the end or loss giftedness in adulthood, and
encounter high-IQ children who are also dyslexic. (c) the end or goals that we should expect gifted children to
It has also been argued that dyslexia is often accom- fullfil. In the following paragraphs I consider each of these
panied by gifts in the visual-spatial arena, a view consis- in turn.
tent with Geschwind and Galaburda's (1987) pathology of
superiority theory, and anecdotal reports on this association When Giftedness Ends in Big-C Creativity
abound (Galaburda & Kemper, 1979; Rimland & Fein, The highest possible endpoint of childhood giftedness is
1988; Sano, 1918; West, 1991). Consistent with this claim certainly creativity in the sense of domain-altering innova-
of association is the finding that there are disproportion- tion (which I refer to here as big-C creativity). Terman's
ately more dyslexics in populations of artists than in the children typically became experts in a well-established
population at large (Winner & Casey, 1993; Winner et al., domain (e.g., medicine, law, business, the academy). Al-
1991). Also consistent is Bloom's (1985) report that none though they may have been creative in the little-c sense
of the 20 world-class mathematicians he studied had (e.g., coming up with innovative approaches to problems),
learned to read before attending school (even though most they did not become major creators. That is, they neither
academically gifted children do read by that time) and that created a new domain nor revolutionized an old domain.
6 had had trouble learning to read. A retrospective study of Yet, expertise as an endpoint should not be lightly dis-
inventors (who are presumably individuals with high me- missed. Society needs experts, and we can neither expect
chanical and spatial aptitude) showed that as children these nor hope that all prodigies will become creators. Many
individuals struggled with reading and writing (Colangelo gifted children grow up to become happy and well-adjusted
et al., 1993). Also, late-talking children have been found to experts in their fields.
have high spatial abilities and to have relatives in spatial However, only a fraction of gifted children eventually
professions such as engineering (Sowell, 1998). become revolutionary adult creators. Those who do so must
Despite these intriguing findings, however, systematic make a painful transition from a child prodigy (a child who
studies of the spatial abilities in dyslexic populations have learns rapidly and effortlessly in an established domain) to
revealed mixed and inconsistent findings (for a review, see an adult creator (a person who disrupts and ultimately
von Karolyi, 1998a). Individuals with dyslexia show no remakes a domain; Gardner, 1993a, 1993b; Simonton,
spatial advantage on a wide variety of spatial tasks, includ- 1977).
ing spatial visualization, mental rotation, spatial memory, It is not surprising that most gifted children, even most
visual scanning, and spatial orientation (Malinsky & Win- child prodigies, do not go on to become adult creators. All
ner, 1999; Winner et al., 1999; for an exception, see yon young children, whether typical or gifted, think divergently
Karolyi, 1998b, 1999). and engage in fantasy play (cf. Richards, 1996, for a
Educational programs for the academically gifted that discussion of divergent thinking in ordinary children).
rely on global IQ scores as an entrance criterion are likely However, this kind of universal creative thinking is quite
to miss children who are unevenly gifted. Of course, ad- different from the kind of big-C creativity that is involved
mission by overall IQ means that children with mathemat- in reshaping a domain. Individuals who are creative in this
ical gifts are treated identically to those with verbal gifts. It big-C sense have a personality structure different from that
would make far more sense to admit children to special of the typical gifted (and nongifted) child: They are rebel-
programs that are tailored to the domain in which they are lious, they have a desire to alter the status quo, and they
gifted (Stanley & Benbow, 1986). have often suffered childhoods of stress and trauma (Gard-
Although our schools do little for the academically ner, 1993a; M. G. Goertzel, Goertzel, & Goertzel, 1978; V.
gifted, with those choosing the curriculum often insisting Goertzel & Goertzel, 1962; Sulloway, 1996). Their fami-
that all children are gifted and hence none need special lies are often a far cry from the complex families of
classes, our schools do even less for the musically or engaged gifted adolescents (Csikszentmihalyi et al., 1993).
artistically gifted child. It is assumed that schools nurture The disproportionate incidence of manic depression in cre-
academic but not artistic or musical abilities. Children with ative individuals also suggests a link (although not a nec-
gifts in an art form are expected to get extracurricular essary one) between creativity and psychopathology (Jami-
training. This is particularly true in the case of music, son, 1993: Ludwig, 1995).
whereas children gifted in the visual arts are likely to The biggest issue for profoundly gifted children is
experiment on their own time in their field of talent, not making the transition from precocity and technical exper-

January 2000 • American Psychologist 165

tise to the innovation of the big-C creator. Consider the When Giftedness Ends
very different situation of a prodigy in an academic or One noninevitable reason that prodigies may fail to make
artistic domain as compared with an athletic prodigy. An the transition is that they have become frozen into exper-
athlete's career is over relatively early in life because tise. This is particularly a problem for those whose work
physical strength and agility are so important. Also, in has become public and has won them acclaim, such as
sports, creativity plays far less of a role than it does in an musical performers, painters, or children who have been
art form or in a scholastic area such as mathematics. In publicized as "whiz kids." Expertise won them fame and
sports there is no transition to be made from technical adoration as child prodigies. It is then difficult to break
perfection to creative interpretation. For the athlete, tech- away from expertise and take the kinds of risks required to
nical perfection is most, if not all, of the story. In contrast, be creative.
in music, mathematics, writing, or the visual arts, the A second noninevitable reason is that some with the
situation is much more difficult for the prodigy. For exam- potential to make the transition do not do so because they
ple, a high-IQ six-year-old who can multiply three-digit have been pushed so hard by their parents, teachers, and
numbers in her head or solve algebraic equations wins managers that they lose their intrinsic motivation (Elkind,
acclaim. However, as a young adult she must come up with 1981; Winner, 1996a). At adolescence they begin to ask,
a new way to solve some unsolved mathematical problem "Who am I doing this for?" If the answer is that they are
or discover some new problems or areas to investigate to pursuing their gift for a parent or a teacher but not for
themselves, they may decide that they do not want to do it
make her mark in the domain of mathematics. Although
anymore and drop out (cf. Bamberger, 1986). The case of
she may remain in the domain of mathematics for her
William James Sidis, a math prodigy pushed relentlessly by
whole life as an excellent teacher, an accountant, an engi-
his father, is one such case among many (Montour, 1977).
neer, or a math professor, she will not have become a These last two reasons show us what can happen when
creator in the domain. Although she may not drop out of culture and greed overtake nature and stamp it out. Parents,
the field entirely, she will not fulfill the highest level of teachers, and psychologists all have an obligation to nur-
potential that a gifted child may reach, big-C creativity. ture prodigies through the potential transition from exper-
The situation is the same in art or music. Technical per- tise to creative innovation and to help them avoid four
fection wins the prodigy adoration, but if the prodigy does dangers:
not eventually go beyond technical perfection into origi- 1. The danger of pushing so hard that the intrinsic
nality, he or she sinks into oblivion. motivation and rage to master these children start out with
There are a number of reasons for prodigies' failure to become a craving for the extrinsic rewards of fame.
remake themselves into big-C creators. Two are inevitable, 2. The danger of pushing so hard that these children
but two are within our control and hence challenge us to later feel they missed out on having a normal childhood.
change how prodigies are nurtured so we may help them 3. The danger of freezing a prodigy into a safe, tech-
make this transition. nically perfect but noninnovative way of performing be-
One inevitable reason is that the funnel is small. There cause this is what he or she has been rewarded for doing so
is simply not enough room at the top for all prodigies to well.
become creators. Therefore, there is an inevitable weeding 4. The danger of the psychological wound caused by
out of those who do not make the cut, so to speak. Any the fall from being a famous prodigy who can perform
domain would be in chaos if there were as many creative perfectly to a forgotten adult who can do no more than
perform perfectly.
adult innovators as there are child prodigies.
A second inevitable reason is that the skill of being a The Ends, or Goals, That Gifted Children
prodigy is not the same as the skill of being a big-C creator. Should Be Held To
A prodigy is someone who can easily and rapidly master an I have argued here and elsewhere (Winner, 1996a, 1996b,
already-established domain with expertise. A creator is 1997a) that we should provide extra resources for the
someone who changes a domain. Personality and will are education of our most able students. The traditional argu-
crucial factors in becoming an innovator or revolutionizer ment for this has been a utilitarian one. These children are
of a domain. Creators have a desire to shake things up. our national resources, and we should cultivate them so
They are restless, rebellious, and dissatisfied with the status they can become our future leaders and innovators. How-
quo (Csikszentmihalyi, 1996; Gardner, 1993a; Simonton, ever, there is also a nonutilitarian reason for intervention:
1994; Sulloway, 1996). They are courageous (cf. Gruber's, We need to intervene for the happiness and mental health
1981, discussion of Darwin's courage) and independent of gifted students. For their emotional well-being, students
(Albert & Runco, 1986). They are able to manage multiple need an appropriate level of challenge. Otherwise, they are
related projects at the same time, engaging in what Gruber not only bored (which can lead to underachievement) but
also socially isolated, and they feel different from everyone
(1981) calls a "network of enterprise" (p. 105).
else. Schools can meet the needs of gifted students without
For these two reasons, we should never expect a
violating egalitarianism. Schools cannot be truly egalitarian
prodigy to go on to become a creator. The ones who do
unless they acknowledge learning differences, including
make this transition are the exception, not the rule.
those differences possessed by students of high ability.

166 January 2000 • American Psychologist

 However, if our schools are to provide specialized Csikszentmihalyi, M. (1996). Creativity: Flow and the p.L,,,chology of
education for the most able, then the most able must also discoveo: and invention. New York: HarperCollins.
Csikszentmihalyi, M., Rathunde, K., & Whalen, S. (1993). Talented
learn to give back to the society that grants them extra teenagers: The roots of success and failure. New York: Cambridge
resources. Thus, one of the ends of giftedness might be University Press.
argued to be service. Today there is a one-sided emphasis Damon, W. (1995). Greater expectations: OvercomhTg the culture of
on the self-actualization of the gifted child. All of the indulgence in America's homes and schools'. New York: Free Press.
Dark, V. J., & Benbow, C. P. (1991). Differential enhancement of work-
educational research on the outcomes of ability grouping,
ing memory with mathematical versus verbal precocity. Journal of
acceleration, pull-out programs, and so on focuses on one Educational Psychology, 83, 48-60.
primary issue: whether the gifted do better on some cog- Davis, G. A., & Rimm, S. B. (1985). Education of the gifted and talented.
nitive or emotional measure when given such education. Englewood Cliffs, N J: Prentice Hall.
An altogether different emphasis can be found at the Israel Detterman, D., & Daniel, M. (1989). Correlations of mental tests with
each other and with cognitive variables are highest for low IQ groups.
Academy of Art and Science, a school for gifted adoles- Intelligence, 15, 349-359.
cents in which students not only develop their abilities but Elkind, D. (1981). The hurried child. Reading, MA: Addison-Wesley.
also participate in community service, making use of the Enersen, D. (1993). Summer residential programs: Academics and be-
kinds of abilities in which they are gifted (Gardner, 1998). yond. Gifted Child Quarterly, 37, 169-176.
Ericsson, K., & Faivre, R. I. (1988). What's exceptional about exceptional
The moral value of service, of giving back to a society that
abilities? in L. Obler & D. Fein (Eds.), The exceptional brain: Neuro-
has devoted extra resources to the gifted, ought to be psychology of talent and special abilities (pp. 436-473). New York:
considered as important as the value of self-actualization of Guilford Press.
the gifted. All children should be taught the value of Ericsson, K. A., Krampe, R., & Tesch-Romer, C. (1993). The role of
service, and gifted children are no exception. deliberate practice in the acquisition of expert performance. Psycho-
logical Review, 100, 363-406.
Ericsson, K. A., & Lehman, A. C. (1996). Expert and exceptional perfor-
REFERENCES mance: Evidence of maximal adaptation to task constraints. Annual
Review of Psychology, 47, 273-305.
Achter, J., Lubinski, D., & Benbow, C. (1996). Multipotentiality among Feldman, D. H. (with Goldsmith, L. T.). (1991). Nature's gambit: Child
the intellectually gifted: "It was never there in the first place, and prodigies and the development of human potential. New York: Teach-
already it's vanishing." Journal of Counseling Psychology, 43, 65-76. ers College Press.
Albert, R. S., & Runco, M. A. (1986). The achievement of eminence: A Freeman, J. (t979). Gifted children: Their identification and development
model based on a longitudinal study of exceptionally gifted boys and in a social context. Baltimore: University Park Press.
their families. In R. J. Steinberg & J. E. Davidson (Eds.), Conceptions Galaburda, A. M., & Kemper, T. L. (1979). Cytoarchitectonic abnormal-
of giftedness (pp. 332-357). Cambridge, UK: Cambridge University ities in developmental dyslexia: A case study. Annals of Neurology, 6,
Press. 94.
Annett, M. (1985). Left, right, hand and brain: The right-shift theory. Gardner, H. (1983). Frames of mind: The theory, of multiple intelligences.
Hillsdale, NJ: Erlbaum. New York: Basic Books.
Bamberger, J. (1986). Growing up prodigies: The midlife crisis. In D. H. Gardner, H. (1993a). Creating minds: An anatomy of creativity seen
Feldman (Ed.), Developmental approaches to giftedness (pp. 265-279). through the lives q[Freud, Einstein, Picasso, Stravinsky, Eliot, Graham
San Francisco, CA: Jossey-Bass. and Gandhi. New York: Basic Books.
Benbow, C. P. (1986). Physiological correlates of extreme intellectual Gardner, H. (1993b). The relationship between early giftedness and later
precocity. Neuropsychologia, 24, 719-725. achievement. In G. R. Bock & K. Ackrill (Eds.), The origins and
Benbow, C. P., & Lubinski, D. (1997). Intellectually talented children: development of high ability (pp. 175-182). New York: Wiley.
How can we best meet their needs? In N. Colangelo & G. Davis (Eds.), Gardner, H. (1998, May). Gifts and responsibilities: A tribute to Raphi
Handbook of gifted education (2nd ed., pp. 155-169). Boston: Allyn & Amram. Paper presented at the Israel Arts and Science Academy,
Bacon. Jerusalem.
Geschwin& N., & Galaburda, A. (1987). Cerebral lateralization. Cam-
Benbow, C. P., & Minor, L. L. (1990). Cognitive profiles of verbally and
bridge, MA: MIT Press.
mathemarcally precocious students: Implications for identification of
Goertzel, M. G., Goertzel, V., & Goertzel, T. G. (1978). Three hundred
the gifted. Gifted Child Quarterly, 34, 21-26.
eminent personalities. San Francisco: Jossey-Bass.
Benbow, C. P., Stanley, J. C., Kirk, M. K., & Zonderman. A. B. (1983). Goertzel, V.. & Goertzel, M. G. (1962). Cradles of eminence. Boston:
Structure of intelligence in intellectually precocious children and in
Little, Brown.
their parents. Intelligence, 7, 153-161. Gordon, H. W. (1970). Hemisphere asymmetry in the perception of
Bloom, B. (1985). Developing talent in young people. New York: Bat- musical chords. Cortex, 6, 387-398.
lantine. Gordon, H W. (1978). Left-hemisphere dominance of rhythmic elements
Boothe, D., Sethna, B. W., & Stanley, J. C. (in press). Special educational in dichotically presented melodies. Cortex, 14, 58-70.
opportunities for able high school students: A description of residential Gordon, H. W. (1980). Degree of ear asymmetry for perception of dichotic
early-college-entrance programs. Journal of Secondary Gifted Educa- chords and for illusory chord localization in musicians of different
tion. Brain and Cognition. (1994). 26(2). levels of competence. Journal of Experimental Psychology: Perception
Bryden, M. P., McManus, I. C., & Bulman-Fleming, M. B. (1994). and Pe~ormance, 6, 516-527.
Evaluating the empirical support for the Geschwind-Behan-Galaburda Gotffried, A. W., Gottfried, A. E., Bathurst, K., & Guerin, D. W. (1994).
model of cerebral lateralization. Brain and Cognition, 26, 103-167. Gifted IQ: LYtrly developmental aspects: The Fullerton longitudinal
Casey, M. B., & Brabeck, M. (1989). Exceptions to the male advantage in study. New York: Plenum.
a spatial task: Family handedness and college major as factors identi- Gross, M. (1993), Exceptionally gifted children. London: Routledge.
fying women who excel. Neuropsychologia, 27, 689-696. Gruber, H. (1981). Darwin on man: A psychological study of scientific
Colangelo, N., Assouline, S., Kerr, B., Huesman, R., & Johnson, D. creativity (2nd ed.). Chicago: University of Chicago Press.
(1993). Mechanical inventiveness: A three-phase study. In G. R. Bock Gruber, H. (1986). The self-construction of the extraordinary. In R. J.
& K. Ackrill (Eds.), The origins and development of high ability' (pp. Sternberg & J. E. Davidson (Eds.), Conceptions of giftedness (pp.
160-174). New York: Wiley. 247-263). Cambridge, UK: Cambridge University Press.
Colangelo, N., & Dettman, D. G. (1983). A review of research on parents Hassler, M. (1990). Functional cerebral asymmetric and cognitive abilities
and families of gifted children. Exceptional Children, 50, 20-27. in musicians, painters, and controls. Brain and Cognition, 13, 1-17.

January 2000 * American Psychologist 167

Hassler, M., & Birbaumer, N. (1988). Handedness, musical attributes, and involvement during cognitive processing may relate to intellectual
dichaptic and dichotic performance in adolescents: A longitudinal precocity. Neuropsychologia, 28, 211-216.
study. Developmental Neuropsychology, 4, 129-145. O'Boyle, M. W., Gill, H. S., Benbow, C. P., & Alexander, J. E. (1994).
Hassler, M., & Gupta, D. (1993). Functional brain organization, handed- Concurrent finger-tapping in mathematically gifted males: Evidence for
ness, and immune vulnerability in musicians and nonmusicians. Neu- enhanced right hemisphere involvement during linguistic processing.
ropsychologia, 31, 655-660. Corwx, 30, 519-526.
Hermelin, B., & O'Connor, N. (1986). Spatial representations in mathe- Oden, M. H. (1968). The fulfillment of promise: 40-year follow-up of the
matically and in artistically gifted children. British Journal of Educa- Terman gifted group. Genetic Psychology Monographs, 77(1), 3-93.
tional Psychology, 56, 150-157. Radford, J. (1990). Child prodigies and exceptional early experience.
Hildreth, G. (1966). Introduction to the gifted. New York: McGraw-Hill. London: Harvester.
Holahan, C., & Sears~ R. (1995). The gifted group in later maturity. Reis, S. M, Neu, T., & McGuire, J. (1995). Talents in two places: Case
Stanford, CA: Stanford University Press. studies (~f high ability students with learning disabilities who have
Hollingworth, L. (1942). Children above 180 IQ, Stanford-Binet origin achieved (The National Research Center on the Gifted and Talented
and development. Yonkers, NY: World Book. Research Monograph No. 951 I30.) Storrs, CT: University of Connect-
Howe, M. J. A., Davidson, J. W., Moore, D. G., & Sloboda, J. A. (1995). icut.
Are there early childhood signs of musical ability? Psychology of Richards, R. (1996). Beyond Piaget: Accepting divergent, chaotic, and
Music, 23, 162-176. creative thought. New Directions for Child Development, 72, 67-86.
Howe, M. J. A., Davidson, J. W., & Sloboda, J. A. (1998). Innate talents: Rimland, B., & Fein, D. (1988). Special talents of autistic savants. In L.
Reality or myth? Behavioral and Brain Sciences, 21,399-406. Obler & D. Fein (Eds.), The exceptional brain: Neuropsychology of
Jamison, K. (1993). Touched with fire: Manic-depressive illness and the talent and superior abilities (pp. 341-363). New York: Guilford Press.
artistic temperament. New York: Free Press. Roe, A. (1951). A psychological study of physical scientists. Genetic
Janos, P., & Robinson, N. (1985). Psychosocial development in intellec- Psychology Monograph. 43(2), 121-235.
tually gifted children. In F. Horowitz & M. O'Brien (Eds.), The gifted Roe, A. (1953a). The making of a scientist. New York: Dodd Mead.
and talented: Developmental per~pectives (pp. 149-195). Washington, Roe, A. (1953b). A psychological study of eminent psychologists and
DC: American Psychological Association. anthropologists and a comparison with biological and physical scien-
Johns Hopkins University. (1999). Institute for the Academic Advance- tists. Psychological Monographs: General and Applied, 67(2, Whole
ment of Youth: History, mission, and goals. Baltimore: Author. Re- No. 352).
trieved December 3, 1999, from the World Wide Web: http://www. Ross, P. O. (1993). National excellence: A case for developing America's
jhu.edu/~gifted/news/mission.html. talent. Washington, DC: U.S. Department of Education, Office of
Karnes, M. B., Schwedel, A. M., & Steinberg, D. (1984). Styles of Educational Research and Improvement.
parenting among parents of young gifted children. Roeper Review, 6, Sano, F ( 1918). James Henry Pullen, the genius of Earlswood. Journal oJ
232-235. Mental Science, 64, 251-267.
Kolata, G. (1983, December 23). Math genius may have hormonal basis. Scarr, S., & McCartney, K. (1983). How people make their own environ-
Science, 222, 1312. ment,;: A theory of genotype-environment effects. Child Development,
Krutetskii, V. (1976). The psychology of mathematical abilities in school 54, 424-435.
children. Chicago: University of Chicago Press. Selfe, J. (1977). Nadia: A case of extraordinao' drawing ability in an
Kulik, J. A., & Kulik, C. -L. C. (1997). Ability grouping. In N. Colangelo autistic child. London: Academic Press.
& G. Davis (Eds.), Handbook of gifted education (2nd ed., pp. 2 3 0 - Shuter-Dyson, R. (1982). Musical ability. In D. Deutsch (Ed.), The psy-
242). Boston: Altyn & Bacon. chology of music (pp. 391-412). San Diego, CA: Academic Press.
Lewis, M. (I985). Gifted or dysfunctional: The child savant. Pediatric Silver, S., & Clampit, M. (1990). WISC-R profiles of high ability chil-
Annals, 14, 733-742. dren: Interpretation of verbal-performance discrepancies. Gifted Child
Ludwig, A. M. (1995). The price of greatness: Resolving the creativity Quarterly, 34, 76-79.
and madness controversy. New York: Guilford Press. Simonton, D. K. (1977). Creative productivity: A predictive and explan-
Malinsky, D., & Winner, E. (1999, April). Dyslexia and visual-spatial atory model of career trajectories and landmarks. Psychological Re-
ability: Is there a connection? Poster session presented at the biennial view, 104, 66-89.
meeting of the Society for Research in Child Development, Albuquer- Simonton, D. K. (1994). Greatness: Who makes histo~' and why. New
que, New Mexico. York: Guilford Press.
McNamara, P., Flannery, K. A., Obler, L. K., & Schachter, S. (1994). Sloboda, J. A., Davidson, J. W., & Howe, J. J. A. (1994). Is everyone
Special talents in Geschwind's and Galaburda's theory of cerebral musical? The Psychologist, 7, 349-354.
lateralization: An examination in a female population. International Sowell, T. (1998). Late-talking children. New York: Basic Books.
Journal of Neuroscience, 78, 167-176. Stanley, J. C. (1988). Some characteristics of SMPY's "700-800 on
Milbrath, C. (1998). Patterns of artistic development in children: Com- SAT-M before age 13 group": Youths who reason extremely well
parative studies ~ftalent. Cambridge, UK: Cambridge University Press. mathematically. Gifted Child Quarterly, 32, 205-209.
Miller, L. (1999). The savant syndrome: Intellectual impairment and Stanley, J. C., & Benbow, C. P. (1986). Youths who reason exceptionally
exceptional skill. Psychological Bulletin, 125, 31-46. well in mathematics. In R. J. Sternberg & J. E. Davidson (Eds.),
Montour, K. (1977). William James Sidis, the broken twig. American Conceptions of giftedness (pp. 361-387). New York: Cambridge Uni-
Psychologist, 32, 265-279. versity Press.
Moss, E. (1990). Social interaction and metacognitive development in Subomik, R., & Arnold, A. (Eds.). (1994). Beyond Terman: Contempo-
gifted preschoolers. Gifted Child Quarterly, 34, 16-20. ra~ longitudinal studies of giftedness and talent. Norwood, NJ: Abtex.
Mueller, H., Dash, U., Matheson, D., & Short, R. (1984). WISC-R subtest Sulloway, F. (1996). Born to rebel: Birth order, family dynamics, and
patterning of below average, average, and above average IQ children: A creative lives. New York: Pantheon.
meta-analysis. Alberta Journal ~?fEducational Research, 30, 68-85. Temple, C. M. (1990). Academic discipline, handedness and immune
Mullis, I. V. S., Martin, M. O., Beaton, A. E., Gonzales, E. J., Kelly, D. L., disorders. Neuropsychologia, 28, 303-308.
& Smith, T. A. (1998). Mathematics and science achievement in the Terman. L. M. (1925). Genetic studies of genius." Vol. 1. Mental and
final year of secondary school: lEA's Third International Mathematics physical traits of a thousand gifted children. Stanford, CA: Stanford
and Science Study. Boston: Center for the Study of Testing, Evaluation, University Press.
and Educational Policy, Boston College. Terman. L. M., & Oden, M. H. (1947). Genetic studies of genius: Vol. 4.
O'Boyle, M. W., Alexander, J. E., & Benbow, C. P. (1991). Enhanced The gifted child grows up. Stanford, CA: Stanford University Press.
right hemisphere activation in the mathematically precocious: A pre- Treffert. D. A. (1989). Extraordinary" people. New York: Bantam Press.
liminary EEG investigation. Brain and Cognition, 17, 138-153. VanTassel-Baska, J. L. (1989). Characteristics of the developmental path
O'Boyle, M. W., & Benbow, C. P. (1990). Enhanced right hemisphere of eminent and gifted adults. In J. L. VanTassel-Baska & P. Olszewski-

168 January 2000 • American Psychologist

 Kubilius (Eds.), Patterns of influence on gifted learners: The home, the Winner, E, (1996b, October 19). The miseducation of our gifted children
self, and the school (pp. 146-162). New York: Teachers College Press. [Commentary]. Education Week, 16, pp. 44, 35.
yon Karolyi, C. (1998a). Dyslexia and visual spatial talents: A review of Winner, E, (1997a). Exceptionally high intelligence and schooling. Amer-
the literature. Unpublished manuscript, Boston College. ican Psychologist, 52, 1070-1081.
von Karolyi, C. (1998b). Dyslexic visual-spatial strength: Rapid visual discrim- Winner, E. (lq97b). Giftedness vs. creativity in the visual arts. Poetics,
ination of impossible figures. Unpublished manuscript, Boston College. 24, 349-377.
yon Karolyi, C. (1999, April). Developmental dyslexia and visual-spatial
strength: Is there a connection? Poster session presented at the biennial Winner, E., & Casey, M. (1993). Cognitive profiles of artists. In G.
meeting of the Society for Research in Child Development, Albuquer- Cupchik & J. Laszlo (Eds.), Emerging visions: Contemporary, ap-
que, NM. proaches to the aesthetic process (pp. 154-170). New York: Cam-
West, T. (1991). In the mind's eye: Visual thinkers, gifted people with bridge University Press.
learning difficulties, computer images, and the ironies of creativity. Winner, W., Casey, M., DaSilva, D., & Hayes, R. (1991). Spatial abilities
Buffalo, NY: Prometheus Books. and reading deficits in visual art students. Empirical Studies of the Arts,
Wilkinson, S. C. (1993). WISC-R profiles of children with superior 9, 51-63.
intellectual ability. Gifted Child Quarterly, 37, 84-91. Winner, E., von Karolyi, C., Malinsky, D., French, L., Seliger, C., Ross,
Winner, E. (1996a). Gifted children: Myths and realities. New York: E., & Weber, C. (1999). Dyslexia and ~spatial ability: Is there a
Basic Books. connection Y Manuscript submitted for publication.

January 2000 ° American Psychologist 169