Wednesday 17 July 2013

Creativity and technical innovation


I have kept my creativity to myself, but others are more generous. They publish papers, register patents, and generally exert themselves in public. What skills are required in order to be really creative?

“One reason creativity and innovation are difficult to
study longitudinally is that few people in the general
population create products deemed creative and innovative
by experts. Because of low base rates, large samples
are needed to generate findings with statistical stability
and real-world generalizability, especially given how
many different ways there are to develop products that
experts evaluate as creative.”

Readers of this blog will have already come across the Kell, Benbow and Lubinski gang, whose Study of Mathematically Precocious Youth longitudinal project is currently tracking 5 cohorts amounting to over 5,000 intellectually talented individuals throughout their lifespan.

Lubinski, D., & Benbow, C. P. (2006). Study of Mathematically Precocious Youth after 35 years: Uncovering antecedents for the development of math-science expertise. Perspectives on Psychological Science, 1, 316-345. View in PDF

Table 1 on page 319 of that paper gives the overall structure of their study, which shows a dedication to longitudinal studies normally restricted to Norwegians living in sunless fiords.

Now Kell, Lubinski, Benbow and Steiger are at it again. In this particular paper they are working on the 563 intellectually talented 13-year-olds (identified by the SAT as in the top 0.5% of ability) tested in the 1970s.

Creativity and Technical Innovation : Spatial Ability's Unique Role” Psychological Science published online 11 July 2013

DOI: 10.1177/0956797613478615

In this new paper they make a simple but powerful argument. If you do not  measure spatial ability you are missing out on a powerful predictor of later achievement, particularly as regards creativity and technical innovation.

This is no news to those of us who have been following their work for years. Of course you need to cover the major features of intelligence, in the way that the Wechsler face to face tests have always done.

Lubinki et al. have been publishing about all this for years. This is because, in addition to the US Scholastic Assessment Test (which does not test spatial ability) a spatial-ability composite score was calculated by equally weighting and summing scores on two Differential Aptitude Test subtests: Mechanical Reasoning and Space Relations. Composites such as these “tap a basic ability in spatial visualization”
(Carroll, 1993, p. 324).

The sample of 393 males and 170 females was 69% Caucasian, 6% Asian or Pacific Islander, 1% African American, and 1% other (23% of participants did not report their race-ethnicity).

So, what the authors are showing the general public, once again, is that if you were to add a spatial component to the U.S. SAT tests, then you would be able to find more bright young students with creative and patentable ideas. 

A two-step discriminant-function analysis revealed that the SAT subtests jointly accounted for 10.8% of the variance among these outcomes (p < .01); when
spatial ability was added, an additional 7.6% was accounted for—a statistically significant increase (p < .01). The findings indicate that spatial ability has a unique role in the development of creativity, beyond the roles played by the abilities traditionally measured in educational selection, counselling, and industrial-organizational psychology. Spatial ability plays a key and unique role in structuring many important psychological phenomena and should be examined more broadly across the applied and basic psychological sciences.

Here is a snapshot of their 2009 findings, showing the rate at which these young people went on to get higher degrees, and in which subjects. V S and M stand for Verbal, Spatial and Mathematical. (Maths still looks a pretty good predictor). Draw your own conclusions about the levels of intellect required in each discipline.


Wai, J., Lubinski, D., & Benbow, C. P. (2009). Spatial ability for STEM domains: Aligning over fifty years of cumulative psychological knowledge solidifies its importance. Journal of Educational Psychology, 101, 817-835. View in PDF

Few things illustrate the vast chasm between the popular (mis)understanding about IQ and the work being done by psychometric researchers than does this paper: to researchers this is a finding which has been well-known since 2006, if not far earlier.  It is obvious that one should test a broad range of intellectual abilities in order to identify and nurture the brightest minds. Bright minds improve economies and societies.

To the general public (or their appointed representatives) the additional test represents a dangerous idea. Spatial abilities are higher in males. There is also a significant difference between genetic groups. The results of the verbal and mathematical tests are already “wrong” (too many males, too few racial minorities). Why add a further test which gives the “wrong” answer, even if it might help us find the brightest citizens?


  1. In Roe's 1951 study of 64 eminent scientists, the average 'spatial IQ' was only 135. I wonder if the requirements of science are different now.

  2. If you've spent much time in good universities it comes as no surprise that the maths, engineering and physical science students are, by miles, the brightest - save that in a British university you'd have to add medics and vets, those being undergraduate degrees here.

    That diagram also reminds me that I've never had the first idea of what the distinction, if any, is between Humanities and Arts. Maybe the diagram alludes to an American distinction?