Studying Intellectual Outliers

Notes by Howard Gardner

As indicated by the succinct title, this article addresses two topics that have generated much discussion around the water cooler and in both professional and lay publications. The answers, expressed succinctly, is that sex differences in performance on tests of mathematical aptitude have decreased greatly in recent years; and that, overall, intellectual performance (as measured  by standardized tests) has gone up in recent years, even among those with high aptitude.

Many readers will know of the brouhaha which occurred in 2005 when then Harvard president Lawrence H. Summers claimed that gender differences in science and math were probably due to some extent to inherited (genetic/biological/brain) differences between males and females. Contrary to the widespread belief, this statement by Summers was not the primary reason that he was removed from the presidency. We do not know the causes of any such differences, but the dramatic change over the years dictates extreme caution before one evokes biological (and, hence, very difficult to affect) differences.

As for the overall rise in scores, it is probably due to several factors. Perhaps the least interesting, and yet possibly the most important, is that test takers (and teachers) have become familiar with certain kinds of tests, and hence, these tests become easier for test takers.

To read the original article in its entirety click here.

How is Number Represented in the Brain?

Notes by Howard Gardner

A principal component of the theory of multiple intelligences has been the belief that the human nervous system has evolved to facilitate a number of relatively independent computations. Rather than a single intelligence, on which one draws for the full panoply of skills, there are neural regions or networks that are more specifically dedicated to language, number, music, and the like.

A quite original aspect of the theory is that the intelligences are not yoked to a specific sensory capacity: for example, linguistic intelligence (known to be represented in the left hemisphere of right handed individuals) is mobilized whether linguistic information enters through the ear, the eye, or (in the case of the blind reading Braille) the fingertips.

When MI theory was developed over 30 years ago, most of our knowledge of the brain basis of cognition came from the study of individuals who had suffered brain damage. The findings were consistent, but at a very gross level; brain damage does not follow strict guidelines! In the intervening era, researchers have developed far more sophisticated means of studying the representation of capacities in the brain.

A study released in Science provides a number of fascinating findings. First of all, by using functional magnetic resonance imaging, it is possible to examine numerical capacities quite specifically in the human parietal cortex. Specific cortical areas, known as association cortex, are stimulated by numerical operations that occur across different sensory capacities; unlike, say, face recognition or tone discrimination, they are not restricted to a specific sensory cortex. Most remarkably, the size and manner of cortical representations actually reflects the size (technically, the numerosity) of the array; if you look at the pattern of neuronal arousal, you can tell whether the array contains few or many stimuli.

What I like best about the study is the clear implication that the human brain has evolved, not only to represent specific sensory input, but also to capture important distinctions that cut across the senses. I suspect that when scientists begin to study other intelligences, ranging from spatial to interpersonal, they will discover a clear rationale for the way these capacities are represented in the human brain.

To read the article in its entirety, click here.

---

Reference:
Harvey, B.M. et al. “Topographic Representation of Numerosity in the Human Parietal Cortex.” (September 2013). Science, 341 (6150), pp. 1123-1126.

The Teaching Intelligence

Notes by Howard Gardner

As some of you know, I have been speculating in recent years that there may be a ‘pedagogical’ or ‘teaching intelligence.’ I’ve been influenced in this direction by conversations with my friends and colleagues Antonio Battro and Sidney Strauss. This article, by Strauss and Ziv, lays out the basic argument for a separate ‘cognitive ability.’ The teaching faculty seems to be universal among human beings, while not detectable in non-human animals. Though, there are likely to be aspects of that faculty which can be observed in other primates and perhaps even in certain species of birds. What’s especially intriguing is that children as young as three already show some ability to adjust their ‘lessons’ in terms of the perceived knowledge, skills, and understanding of their ‘students.’

In our book The App Generation, Katie Davis and I argue that the nature of early teaching is very important. We cite the work of developmental psychologist Elizabeth Bonawitz, who has demonstrated an important phenomenon: children are likely to play with and explore a toy for a longer period of time if they have just had a short and obviously partial introduction to the toy, than if the ‘teacher’ purports to demonstrate the complete working of the toy. This line of research suggests that the model of teaching that we put forth in early life may have significant influence on how growing children conceive of the ‘teaching encounter.’

To read the article in its entirety click here.

Intelligence: New Findings and Theoretical Developments

This is an important review article by several of the leading researchers in the world: individuals with a long time interest in the study of intelligence from a psychological point of view. It undermines many widely held views about intellect, such as those made famous thirty years ago in The Bell Curve. For example, the authors report that intelligence is on the rise around the world, that adoptions can significantly increase intelligence when children of working class background are raised in middle-class homes, and that efforts to locate the gene(s) that regulate measured intelligence remain elusive. The review also outlines issues that remain to be worked out; for example, whether ‘a general intelligence factor could arise from initially largely independent cognitive skills.’

Quoted is the authority W.T. Dickens who argues “even if these skills are initially largely independent of one another, after people interact with their environments, these skills will no longer be independent. Someone who is good at any intellectual skills is more likely to end up in environments where all skills will be practiced, which will lead to the development of all skills.” An intriguing hypothesis. I would raise an opposite possibility: that perhaps a variegated intellectual profile may emerge from initially similar cognitive potentials. I regret that the article does not discuss theories of the multiple intelligence variety. The gulf between my work and that of mainstream psychometricians remains large.

To read the article in its entirety click here.

Fewer Tests, Not More

Notes by Howard Gardner

On the surface, it could appear as if I were in enthusiastic agreement with John Mayer. After all, he calls for more attention to creativity, and for tests of musical, spatial, and emotional (interpersonal/intrapersonal) intelligence.  Shouldn’t this hymn of praise to psychometrics be music to the ears of an advocate of Multiple Intelligences?

And yet, though I respect Mayer and Salovey’s work on emotional intelligence, I have no enthusiasm whatsoever for his recommendation of ‘more tests, not fewer.’  Already our kids are being over-tested in K-12, and when they don’t do well, rather than try to improve their performance, all too often we just test them again.

Next, Mayer assumes that we have adequate tests for these different strengths, dispositions, or intelligences. But most of the extant tests are simply multiple choice or short answer tests, and we know that such tests heavily tap linguistic and logical-mathematical intelligence. In my one effort to create measures of the various intelligences, my colleagues and I did not create short-answer instruments;  instead, in Project Spectrum, we created environments in which one could observe students as they work with different kinds of materials, what I call ‘intelligence-fair testing.’ So for bodily-kinesthetic intelligence, we look at how students master a new dance or for spatial intelligence, we see how they come to navigate an unfamiliar space.  One shudders to think of how such measures could be deployed on a massive scale.

Also, I identify conceptual problems with Mayer’s recommendations. He favors measuring creativity, but assumes that is the bailiwick of artists and musicians. As a matter of fact, individuals can be creative or non-creating in any domain; there are plenty of creative scientists, and many artists and musicians who are not creative at all.

Finally, let’s think about college admissions. At any selective school, admission officers are already deluged with information about the candidates when they can only admit one out of five, or perhaps even one out of 15! Imagine if they now received scores on several other tests of varying degrees of credibility. This might be a boondoggle for test makers and for those individuals who call themselves ‘college admission counselors’ or ‘coaches’ and charge outlandish fees for their services.  But would the students of colleges be enhanced?  I’d prefer to call for “Fewer Tests” and more trustworthy self-descriptions and letters of recommendations. But I also concede that such calls are unlikely to be heeded in the United States, to everyone’s loss.

To read John Mayer's original Op-Ed click here.