Is creativity an inborn personality trait or a set of learned skills? The answer determines whether we test for intrinsic talent or teach creative skills. Current research on screening for scientific creativity suggests why getting the answer right is critical to the future of women in science.
Many psychologists treat creativity as a fixed personality trait. Some people have the "genes" or the "brain wiring" or whatever to be creative, and some don't. Assuming that creativity is hard-wired, numerous researchers have attempted to develop tests that screen for and identify that trait among younger and younger children. The problem with most of these tests is that they are never validated in terms of whether children or youth with ‘high potential' for creativity actually exhibit more creativity in their adult lives or work than do children or youth with ‘low potential'. In fact, many tests for creativity are based on so-called ‘divergent' or ‘lateral' thinking exercises that are known to correlate very poorly or not at all with real-life creativity. In general, we think such tests are junk.
Some tests for creativity, however, do correlate well with real-life performance. So what's the problem? Whether predictive or not, the tests themselves tell us nothing about the origins or foundations of the creative potential they spot. We believe there is a danger that these predictive tests could be misused to strengthen the case for treating creativity as a personality trait, rather than a set of learnable skills. Why does this matter? Consider the implications of a forthcoming paper by David Lubinski and Camilla Benbow (see Holden, 2009). In our opinion it provides an interesting test case of direct relevance to our social understanding of whether or not women have what it takes to be creative scientists and engineers.
Let us begin by making it very clear that it is not the purpose of Lubinski or Benbow to argue that women can't be excellent scientists or engineers. They certainly do not draw this conclusion in their research. But their forthcoming paper on testing for scientific creativity is going to feed this ugly issue whether they want it to or not.
So let's look at what Lubinski and Benbow have found. For several decades, they have been studying students who achieve extremely high scores (e.g., greater than 700 on the verbal or math section) on the SAT by the time they are 13 years old (Lubinski, 2009). Their previous studies have shown weak correlations between such precocious SAT scores and later success in a science or technology field. They also found, however, that although they had a significant group of women who achieved such amazing scores, almost none of them went into science, mathematics, or technology, period. The vast majority of these women chose law and medicine instead (Lubinski & Benbow, 2006). Lubinski and Benbow subsequently found one reason (Holden, 2009): High scores on visual thinking tests are far more predictive of successful science careers, important publications, and obtaining patents than SAT scores, but women, as a group, tend to score significantly more poorly on visual thinking tests than men. Perhaps this lack of a necessary skill causes them to avoid science and technology careers.
(As an example of a visual imaging test take a look at the figures here. Can each of these objects be rotated to fit perfectly into one another?)
You can see where this might lead (and again, we stress that this is NOT where Lubinski and Benbow go, nor is it where we believe we, as a society, should be going!). If one interprets visual thinking ability as being an intrinsic mental trait, and if one acknowledges that visual thinking ability is highly correlated with scientific and technological creativity and success, then it follows logically that women are not, as a group, mentally equipped to succeed in science and technology. This is, in fact, the proposition that got Lawrence Summers fired as President of Harvard University not too long ago. It's a hot-button issue.
The fallacy of such propositions is the assumption that tests measure intrinsic ability rather than trainable skills. We, too, have documented the fact that visual thinking ability correlates with scientific success (Root-Bernstein, et al., 1995). We have also discovered that scientific creativity is predicted by avocations that build visual thinking skills, such as visual arts, sculpting, modeling, photography, painting, wood- and metalwork, and other forms of tool use (Root-Bernstein, et al., 2008). This additional set of correlations strongly suggests that visual thinking ability can be trained. In fact, a number of science and technology educators have demonstrated in controlled studies that science and engineering students who initially test poorly on visualization tests (many of them women), and who are subsequently given mechanical or artistic drawing training, improve dramatically on retest and perform better overall in their science and engineering coursework. Women benefit particularly from such training (Lord, 1985; Deno, 1995; Sorby SA, Baartmans BG. 1996; Alias, et al., 2002; Sorby, 2009).
So what are visual thinking tests actually good for? We believe that they are good for differentiating between those who have already achieved a certain skill level and those who need further training. Testing, in our view, shouldn't be used to cull the ‘scientifically talented' from the ‘scientifically untalented', but rather to insure that our schools and colleges provide an adequate and appropriate education that will bring every student up to their full potential. If we assume that visual thinking ability is intrinsic rather than learned, we will fail to follow through for those students who can benefit from proven interventions such as crafts, drawing and computer modeling classes.
The distinction that we have drawn here between intrinsic ability and acquired skill is not, of course, as black and white as we have made it out, but we have drawn it this way in order to make an important point: It is all too easy to fall into dichotomous thinking. Lubinski and Benbow, for example, recently gave an interview about their research in which they proposed a much broader program of testing for visual thinking in order to "identify Edisons and Fords" overlooked by other forms of testing. Their language unfortunately sounds as if they are thinking of visual thinking as an inborn trait rather than a learnable skill. The point of such testing would be to find what's already out there, not to remedy educational deficits. Yet the notion that a single score on a single test taken at single point in our students' lives could identify all the scientific and engineering talent in the country seems absurd to us. We hope that Lubinski and Benbow also find it so.
Rather than waste money using visual tests to search for Edisons and Fords (both male!), we suggest modifying science and technology curricula to teach the ‘thinking tools' that underpin all creative thinking. Visual thinking (indeed, imaging and manipulative skills of many kinds that are valuable to scientists and engineers) can and ought to be taught to improve the skills of all students (especially women), thereby equitably enlarging the pool of potential innovators far beyond a few Edisons and Fords.
If we're right, more students will do better in their science and technology courses; women will come into their own in these professions; and the pool of innovators will expand. It all comes down to whether we assume the pool of talent is predetermined and the problem is to find it, or whether we believe that the pool of talent is determined by how well we teach. We're teachers. We believe in the latter (Root-Bernstein, 2009).
© Robert and Michele Root-Bernstein
Alias M, Black TR, Grey DE. (2002). Effect of instructions on spatial visualization ability in civil engineering students. Inter Ed J 3(1): 1-12.
Deno JA. (1995). The relationship of previous experiences to spatial visualisation ability. Engineering Design Graphics J, Autumn: 5-17.
Holden, Constance. (4 Sep 09). Science needs kids with vision. Science, v. 325, p. 1190.
Lord TR. (1985). Enhancing the visuo-spatial aptitude of students. J Res Sci Teaching, 22 (5): 395-405.
Lubinski D. (2009). Exceptional cognitive ability: the phenotype. Behavioural Genetics, 39: 350-358.
Lubinski D, Benbow C. (2006). Study of mathematically precocious youth after 35 years. Perspectives on Psychological Science, pp. 316-345.
Root-Bernstein RS, Bernstein M, Garnier H. (1995). Correlations between avocations, scientific style, work habits, and professional impact of scientists. Creativity Res J 8(2): 115-137.
Root-Bernstein RS, Allen L, Beach L, Bhadula R, Fast J, Hosey C, Kremkow B, Lapp J, Lonc K, Pawelec K, Podufaly A, Russ C, Tennant L, Vrtis E,Weinlander S. (2008). Arts foster success: Comparison of Nobel prizewinners, Royal Society, National Academy, and Sigma Xi members. J Psychol Sci Tech 1(2):51-63.
Root-Bernstein R. (16 Oct 2009). Teaching, not testing, for scientific vision. Science, v. 326, pp. 365-6.
Sorby SA, Baartmans BG. (1996) .A course for the development of 3D spatial visualization skills. Engin Design Graphics J 60(1): 13-20.
Sorby S. (2009). Developing spatial cognitive skills among middle school students. Cogn Process 10 (Suppl 2):S312-S315. DOI 10.1007/s10339-009-0310-y