Multiple Intelligences in 2026: Q&A for a Chinese Journal

Recently, I participated in a Q&A-style interview with a visiting scholar to Harvard’s Graduate School of Education named Junjin Hu. In the thoughtful exchange below, we consider the origins of my theory of multiple intelligences, core issues, and educational applications. This interview was also published in a recent edition of Research MI.

PART I. The Theoretical Origins of the Theory of Multiple Intelligences

Question: Dear Professor Gardner, it is a great honor to have the opportunity to ask you several questions about the theory of multiple intelligences. I would like to begin with your early research. Could you discuss the connection between your early research on the psychology of art and the eventual development of the theory of multiple intelligences? Before proposing the MI theory, you had already explored the idea of art as a cognitive mode of the human mind in works such as The Arts and Human Development (1973), Artful Scribbles (1980), and Art, Mind, and Brain (1982). These studies can be seen as an attempt to construct a “cognitive psychology of art.” I am curious how this early line of inquiry into art and the mind gradually led you toward formulating the theory of multiple intelligences.

Answer: In the middle 1960s, I became interested in the study of cognitive development—as pioneered by Jean Piaget and pursued as well by my mentor Jerome Bruner. As a young person interested in the arts, I was surprised—and disappointed—that for these and other scholars, being “cognitively developed” meant “thinking like a scientist”—using what Piaget called “formal operations.” I believed—and still believe—that artists are as cognitively developed as scientists—but they develop different mental faculties and deploy them in different ways.

Jean Piaget

And so, with the enthusiasm and ambition of a young scholar, I decided to think about and to study—empirically—the development in children of skills, knowledge, technique in the arts. As we phrased it in those days, to deem “artistic competence” as a form of development as important as “scientific competence.” My first scholarly book—published in 1973—was called The Arts and Human Development. And by that time, I was a researcher in artistic development and also the co-director of Harvard Project Zero.

Question: Could you reflect on the influence of Nelson Goodman on your work? Goodman was not only a leading figure in analytic philosophy and symbolic logic, but also a scholar with profound insights into the artistic world. His Languages of Art (1968) is considered a classic in analytical aesthetics, and at the time of its publication he had founded Harvard’s Project Zero. As a researcher in the philosophy of education, I am very interested in how Goodman’s ideas on symbol systems and the arts shaped your thinking and research trajectory.

Nelson Goodman

Answer: As just mentioned, as a young doctoral student, I was searching for a way to understand the development of artistic skills, abilities, creativity. Almost by accident, I learned that a distinguished philosopher named Nelson Goodman was launching a research project on education in the arts and was looking for young research assistants. I went to meet Nelson Goodman, we “hit it off” (liked each other) and in the fall of 1967, I became the first research assistant at Project Zero. I was soon joined by MIT computer science doctoral student David Perkins. When Goodman announced his retirement from Project Zero in the early 1970s, David and I became the co-directors and served in that capacity until 2000.

I was not and am not a philosopher. That was an advantage in working with Goodman. He was a tough personality and held his doctoral students in philosophy to exceedingly high standards. Goodman treated me more as a son than as a doctoral student. He tried to teach me about how he conceived of the arts and encouraged me to study artistic development empirically—for example, figuring out how young persons could perceive style in the arts.

Goodman was a very careful writer. He once said, “When I am reading a paper and I don’t understand a sentence, I stop reading.” That was a very high standard indeed—even today, six decades later, I can still hear those words!

Also, in the late 1960s, Goodman and I learned about how different mental/cognitive faculties were represented in the human brain. We invited the brilliant neurologist Norman Geschwind to speak at Project Zero. Geschwind made a convincing argument that the left hemisphere of the brain handles certain kinds of symbols (roughly speaking, linguistic and numerical symbols), while the right hemisphere of the brain handles spatial and textural materials. This finding fit into our own emerging thinking about artistic cognition. I was fortunate to be able to carry out postdoctoral research in neurology and aphasia under the direction of Geschwind.

It's a good quip in English—my scholarly life was transformed by two men with the initials NG: philosopher Nelson Goodman (1906-1998) and neurologist Norman Geschwind (1926-1984). I miss them both very much.

Question: How do you view the relationship between multiple intelligences and Nelson Goodman’s notion of “multiple symbol systems”? In Ways of Worldmaking (1978), Goodman argues that humans do not understand the world through a single logical language, but through multiple symbol systems such as language, images, sound, movement, numbers, diagrams, and models, each of which helps construct different “worlds.” It seems to me that the theory of multiple intelligences extends this idea into the domain of the mind: the human mind is not a single logical instrument but a set of relatively independent symbol-using capacities. In this sense, each intelligence functions almost like a symbolic system within the mind. Would you consider this interpretation accurate?

Answer: You have understood well the confluence between Goodman’s philosophical distinctions and my psychological speculations. Goodman used to quip that “cognitive psychology is the most interesting branch of philosophy,” and “a psychologist is just a philosopher with a research grant.” Indeed, Goodman learned about cognitive psychology during a year spent at the Center for Cognitive Studies at Harvard. His host there was Jerome Bruner, who also was one of my principal mentors.

Jerome Bruner

Indeed, at the 25th anniversary celebration of Project Zero, both Bruner and Goodman made an appearance and that was very meaningful to me. Somewhere there is a photograph of the two of them together.

Question: How did Piaget, structuralism, and the broader tradition of symbolic philosophy influence your development of the theory of multiple intelligences? The symbol-systems approach and the twentieth-century tradition of symbolic philosophy, represented by figures such as Ernst Cassirer, Susanne Langer, and Alfred North Whitehead, emphasize that symbolic activity lies at the heart of human cognition and creative development. Piaget’s genetic epistemology, however, focuses primarily on logical and rational structures and portrays the child as a “little scientist,” which makes it difficult to account for symbolic creativity in areas such as the arts. In The Quest for Mind (1973), you examined both Piaget’s and Lévi-Strauss’s ideas. I am curious how these intellectual traditions, including Piaget’s genetic epistemology, Lévi-Strauss’s structuralism, and the broader currents of symbolic philosophy, collectively shaped your formulation of multiple intelligences. I also noticed that in Frames of Mind (1983) you place particular emphasis on biological and anthropological perspectives.

Answer: When I try to describe my own skills as a scholar, I consider myself a synthesizer. I called my scholarly memoir A Synthesizing Mind and have written dozens of blogs about synthesizing, which you can read here.

As a student, I read all of the scholars whom you mentioned—and in fact, Susanne Langer had been the teacher of my first wife Judith Krieger Gardner, and I have a letter from Langer hanging in my office. I was fortunate to have personal relationships with Jerome Bruner and Claude Lévi-Strauss and also to have met and interviewed Jean Piaget a few times—letters from these men also adorn my office.

I began college as a “history major,” but was attracted by a relatively new field of study called “Social Relations” (shortened to “Soc Rel”)—an effort to synthesize theories and findings in the relatively new fields of psychology, sociology, and anthropology. So, I was quite prepared to encounter and appreciate the works of Lévi-Strauss. At the time (the middle 1960s), the French school of structuralism/structural analysis was not well known in the United States. I saw intriguing parallels between the works of Piaget and Lévi-Strauss. I thought that it would be worth spelling them out in a book The Quest for Mind: Piaget, Claude Lévi-Strauss and the Structuralist Movement—my other major publication in 1973. Also adorning my office are letters from these two savants—both dated April 10, 1970! (Piaget’s in French, Lévi-Strauss in almost perfect English.)

Question: How did your training in neuroscience influence the theory of multiple intelligences? In The Shattered Mind (1975), you analyzed cases of brain damage to examine how different symbolic capacities can be impaired or preserved. These findings seem to reveal the natural classification and structural organization of symbol systems in the brain. I am curious how this line of neuropsychological research contributed to your eventual formulation of the theory of multiple intelligences.

Answer: My work in neuroscience—particularly on the effects of damage to the brain as studied by Norman Geschwind and many other researchers at his aphasia research center—was crucial to the development of MI theory. 

The single most important facet of damage to the human cortex is the location of the lesion—left or right hemisphere, anterior or posterior, shallow or deep. And as it happens, brain lesions can affect different facets of cognition—processing of particular symbol systems and how those symbols are used. 

To the extent that I have made any contribution to basic science, it is in delineating the effects of brain damage on various kinds of artistic and symbol-using capacities. And probably the most important work was the identification of the role of the right hemisphere in understanding metaphoric (as opposed to literal) aspects of language use.  I carried out this work with Ellen Winner, to whom I have been happily married since 1982.

Question: What do you consider to be the most fundamental biological basis of the theory of multiple intelligences? In China, a common critique is that MI lacks solid empirical support. Yet your research on brain damage, along with your discussion of the “biological foundations of intelligence” in Frames of Mind (1983), suggests that you place considerable emphasis on neuroscientific evidence. This gives me the impression that MI is not merely a developmental psychology theory but one with a strong psychobiological orientation.

Answer: Thank you for that question. It is often said that MI lacks empirical support. That is nonsense! Frames of Mind, published in 1983, cites and builds upon hundreds of empirical studies. And in the preface to the new edition—published in April 2026—I cite intriguing new lines of work.

Cover of the latest edition of Frames of Mind (2026)

MI is not an experimental theory. One cannot do an experiment—or even a sheaf of experiments—to prove MI theory right or wrong. Instead, the theory of multiple intelligences involves the synthesizing of vast amounts of data from many studies in many scholarly disciplines. I am sure that MI theory could be refined in light of the forty years of psychological, neurological, and anthropological work done since the original publication—but I have no reason to apologize for—let alone to withdraw—the general approach and the major conclusions.

Indeed, with my treasured colleagues, Shinri Furuzawa and Annie Stachura, I have now been investigating animal intelligences, plant intelligences, and artificial intelligences. (You can read the article we co-authored here.) While IQ tests and success in certain kinds of scholarly work still have their uses, they seem to be “period pieces” in the age of AI. They miss the vast expanse of intelligences in the world of today…and tomorrow.

Question: I am also curious about the influence of neurobiology on your development of the theory of multiple intelligences. In your writings, you frequently refer to Conrad Hal Waddington’s concepts of canalization and plasticity. Canalization highlights the biological stability of developmental trajectories, whereas plasticity emphasizes their sensitivity and openness to environmental shaping. These concepts seem highly relevant to the theory of multiple intelligences. How does MI address the tension between the biological foundations of each intelligence and their cultural malleability?

Answer: This is an important issue and one that I alluded to in the previous question. Work by Miriam Hauptman and her colleagues indicates that certain human faculties are established and essential quite early in life while others are more susceptible to change—due to neural plasticity. As a simple example, it is not difficult to learn new languages when you are an adult, but very difficult to master accents, or tonality in various languages. While I have not myself carried out work on canalization vs. plasticity, I suspect that we will find those tensions at work across the spectrum of intelligences. (I suspect that a sense of perfect pitch is much easier to acquire in childhood than in adolescence or later.) 

Miriam Hauptman

Of course, with the rise of artificial intelligence and many computational systems, one can compensate for the loss of some faculties—I may not be able to have a good French accent, but my avatar does just fine.

Question: I am also curious about the relationship between Jerry Fodor’s theory of modularity and the theory of multiple intelligences. Fodor presented his modular view of the mind in The Modularity of Mind (1983), and your Frames of Mind published in the same year. Your formulation of multiple intelligences seems to suggest a kind of functional modularity. To what extent do you think MI can be understood as compatible with Fodor’s modularity at a higher functional level?

Answer: Fodor and I are roughly contemporaries—and, as you point out, our books were published in the same year. We were both greatly influenced by the linguistic work of Noam Chomsky. Superficially, our works can be grouped together: my “intelligences” arguably each contain one or more Fodorian “modules.” But Fodor was not much interested in psychology and actually declared that study of the brain was worthless for cognitive scientists. Nor—as far as I know—was Fodor interested in educational or developmental issues. In that sense, he was very different from Nelson Goodman, or indeed, from Noam Chomsky himself.

In the future, an historian of science should take a look at the different streams of knowledge catalyzed by Chomsky’s scholarly work. And now, in the age of AI, we have learned that a lot of Chomsky’s speculations about processes of human cognition apply not only to human beings but also to Large Language Models. This does not, in my view, minimize the importance of Chomsky’s work. The questions that scholars raise are vital, even as the answers will change over time—as we have observed with figures as pivotal as Newton, Darwin, or Einstein.

II. Core Issues in the Theory of Multiple Intelligences

Question: Among the many related concepts, why did you choose “intelligence” as the central focus of your work? Terms such as faculty, ability, skill, talent, and creativity also describe differences in human cognition, yet the word “intelligence” seems to exert a particular appeal for the public. If Frames of Mind (1983) had been titled with one of these other terms instead of “intelligence,” its impact might have been very different. Could you discuss your reasons and considerations for adopting this specific concept?

Answer: Your assumption about the importance of selecting the word “intelligence” is absolutely correct. In the “West” (roughly the Europe and the Americas), that lexical decision made all the difference. If I had developed a “theory of multiple talents,” even otherwise critical people would have said, Sure, there are multiple talents! And then let the topic drop. 

But since I used the word “intelligence” and never disavowed it, I joined a battle, a contest with psychometricians who believe that they alone can define intelligence, and measure it, that it is singular, and that speaking of multiple intelligences, is not only invalid, it’s an abuse of language.

Of course, any informed psychometrician will concede that linguistic, logical, and spatial faculties are separate. Strength in one does not predict strength or weakness in the other two faculties. And with the advent of the work of Daniel Goleman on emotional intelligence, both interpersonal and intrapersonal intelligences now have a place in psychometry. And so, the argument—which I have no interest in pursuing at this time—now revolves around the status of musical, bodily-kinesthetic, and naturalist intelligences.

Question: In Chinese, the term “intelligence” has been translated in several different ways, including “zhili,” “zhineng,” and “zhihui,” each carrying distinct meanings. “Zhili” refers to a general cognitive capacity for understanding, reasoning, and problem solving. “Zhineng” emphasizes information processing and operational abilities and can apply to both humans and machines. “Zhihui” points to an integrated mental state that combines rationality, emotion, and ethical judgment. Since different translations imply different understandings of your theory, which of these do you feel best captures your original conception of “intelligence”? 

Answer: Since I don’t speak or understand Chinese, my answer is necessarily tenuous.  Clearly I am not referring to zhihui—I see emotions and ethics as separate domains. 

I feel somewhat closer to zhineng—because I am using MI theory to try to understand computational systems (see the aforementioned work with Furuzawa and Stachura). 

As for zhili, it all depends. If you believe that there is a general cognitive capacity, we are in disagreement. But if you believe that the question of one or more cognitive capacities is an empirical one, then we are “on the same page.”

Question: In classifying different intelligences, you adopt a cross-disciplinary synthesizing approach that draws on evidence from multiple fields to support the existence of each intelligence. However, some experimental psychologists argue that this method is subjective, lacking testable hypotheses and rigorous experimental design, and therefore belongs to a “soft science” that does not rely on strict quantitative data. Some even view the theory of multiple intelligences as a kind of personally constructed theoretical myth. How would you respond to these methodological critiques?

Answer: You raise two critiques here. The first one is that my method does not conform to the traditional view of science. Not all science is experimental in the common sense of that word—in general, claims in geology, astronomy, evolutionary biology are not testable in the way that claims in organic chemistry or psychophysiology are. 

I call the kind of science or scholarship that I practice synthesizing. It involves putting together vast amounts of empirical data in ways that make sense and open up new questions and new kinds of inquiry.

I’ve written a great deal about synthesizing—dozens of blogs and a memoir called A Synthesizing Mind. Interested readers can consult these sources.

The second criticism is foolish, if not disingenuous. My book Frames of Mind is based on hundreds of studies in the areas of psychology, neuroscience, anthropology and other areas of scholarship. I don’t believe that anyone who speaks about “theoretical myths” has ever read (or even opened up!) my 400-page book Frames of Mind (which is being published with a new preface in April 2026). I hope that the new edition will be translated into Chinese.

Question: In Frames of Mind, you proposed eight “signs” for identifying an intelligence, including potential isolation by brain damage, uneven developmental profiles across individuals, the presence of an identifiable core operation or set of operations, a distinctive developmental trajectory from novice to expert, an evolutionary history and evolutionary plausibility, and sensitivity to symbol systems, among others. These signs provided the foundational criteria for classifying intelligences. Several decades have now passed. Have you revised, refined, or reconsidered any of these criteria in light of subsequent research and developments in the field?

Answer: I have not revisited the eight signs of an intelligence. These were worked out over the course of five years of study that led to the writing and publication of Frames of Mind. I have considered the evidence for various candidate intelligences and have concluded that there is sufficient evidence to anoint an 8th intelligence called the naturalist intelligence. This is the capacity to make consequential distinctions between one plant and another, one animal and another, one cloud formation and another—distinctions crucial for survival on our planet. I claim that this capacity—seemingly less crucial in an urban environment—continues to be drawn upon in our time to distinguish one commercial product from another, be it automobiles or sweaters or perfumes.

I have been asked to consider other candidate intelligences—including the possible spiritual or existential intelligence. Such candidates are plausible, but I am no longer engaged in the exercise of evaluating other intelligences. (It took a year of study to decide on the validity of a “naturalist intelligence.”) My criteria exist and others are welcome to draw on them—so long as they don’t attribute their conclusions to me!

Question: The theory of multiple intelligences emphasizes the relative independence of each intelligence, a “modular” stance that brings conceptual and biological clarity to the framework. However, many assessments reveal positive correlations among different abilities, and some scholars, such as Piaget, argue for the existence of a general, overarching intelligence. This issue parallels a broader tension in neuroscience between localizationist and holistic perspectives: even if intelligences rely on partially distinct neural regions, might there still be a central integrative mechanism at work? How do you reconcile the observed correlations among intelligences with your claim about their relative independence?

Answer: You raise important issues here. Let me take them one by one:

Certainly, in particular individuals there will be correlations among particular intelligences. If I were to be convinced that, say, musical and spatial abilities were highly correlated, I might combine those two intelligences. On the other hand, if I became convinced that logical capacities were different from mathematical capacities, I would separate them into separate intelligences—and then we would have 8 or 9 intelligences.

I have no objection to the positing of a central integrative mechanism—many scholars would allocate this task to the frontal lobes or pre-frontal lobes, and some might also nominate areas in the parietal lobes for certain kinds of integration.

What I have done is to describe the building blocks—and then to point out that in certain populations, an intelligence may exist in isolation—or, on the contrary, that in certain populations or under certain circumstances, two intelligences might be combined.

If I became convinced that there were a “central intelligence entity” that routinely combines the several intelligences, I would be prepared to revise my theory. But today, over forty years after its positing, I do not find convincing evidence that intelligences are routinely combined in specific combinations. Indeed, the more we learn about cognition, the more we learn about abilities that are impaired, or spared, in isolation. The whole field of neurodiversity—hardly visible a half-century ago—is now a vital part of medicine, education, and rehabilitation.

Question: In China, alongside the theory of multiple intelligences, Robert Sternberg’s triarchic theory of intelligence and his later theory of successful intelligence are also highly influential. The triarchic theory seeks to move beyond the traditional IQ emphasis on analytical abilities by distinguishing analytical, creative, and practical intelligence. However, Sternberg’s definition of intelligence still centers on problem solving. In contrast, your definition of intelligence encompasses not only solving problems but also creating culturally valued products within specific cultural contexts. Because of this cultural dimension, the MI framework remains open-ended, allowing for the potential inclusion of additional intelligences when justified by evidence. How do you view the fundamental differences between the theory of multiple intelligences and the triarchic theory?

Robert Sternberg

Answer: You have given a very good and convincing answer to the question that you raised. My goal is to describe 8 or so separate computational mechanisms—which can operate in isolation or be damaged in isolation. They are building blocks, so to speak.

As I understand it, Robert Sternberg is interested in the ways in which our computational mechanisms are deployed. For instance, linguistic intelligence can be used for analysis (in analyzing a poem), creativity (in writing a poem) and for practical purposes (selecting a greeting card).

Accordingly, I see the two theories as complementary. In fact, thirty years ago, Sternberg and colleagues worked with my research team at Harvard Project Zero, to develop a school program called “Practical Intelligence.” You can read about our efforts in my book Multiple Intelligences: New Horizons. In the course of that collaboration, I did not sense a major difference in how we think about human intellect—the differences were more in our research methods and pedagogical goals and how we pursued them. 

As an example, Sternberg was far more interested in developing curricula and tests—while I was more interested in creating experiences that would engage students in the activities of school.

Question: I would also like to raise a question about the cultural and ethical dimensions of intelligence. Within the framework of multiple intelligences, could cultural preferences for certain intelligences lead to ethical imbalances? In other words, does a society have a moral obligation to ensure that every form of intelligence has the opportunity to be developed? If cultural power structures consistently reward some intelligences, such as linguistic or logical-mathematical intelligence, while devaluing others, such as spatial, musical, bodily, or naturalistic intelligence, would this amount to a form of epistemic injustice?

Answer: I agree that cultures send strong signals about which intelligences they prioritize, and which accordingly get short shrift. As just one example, learning to sing and to play instruments has been much more important in Finland and Hungary than in other countries. Obviously, naturalist intelligence is more of a premium in a community far away from cities and set in a rural agricultural area. To a Westerner, Japan seems to be especially focused on interpersonal intelligence, and not as much on intrapersonal intelligence.

I would not go so far as to say that every society is obliged to nurture every intelligence. But now that these intelligences have been described, I certainly think that no society should block their development. Preferably, societies should be encouraged to nurture and develop the range of intelligences in all of their young people. I have long admired children’s museums in the United States and cultural palaces in China for providing such opportunities to young persons.

Also, now that we have overwhelming evidence for neurodiversity, it’s especially important to make sure that every child has exposure to a range of subjects and pursuits, and that educational entities should offer different approaches to subjects and topics deemed important.

Question: If you were to rewrite Frames of Mind today, which chapters or core claims would you revise? With the rapid development of artificial intelligence, does the concept of “intelligence” itself require updating? Do you foresee the possibility of adding new types of intelligence in the future?

Answer: I have two answers to this question:

  1. I would not try to rewrite the book though I would certainly update with new findings from neuroscience, psychology, and other disciplines.

    Instead, in the preface to the new edition (to be published in April 2026), I have indicated the ways in which my thinking has changed.

  2. On the other hand, I think that the exclusive focus on human intelligences—while important and fascinating—is much too narrow, too egocentric.  Accordingly, as mentioned, with my colleagues Shinri Furuzawa and Annie Stachura, I have written a lengthy essay on “Who Owns Intelligence?” In that essay, we review claims about animal intelligences, plant intelligences, and artificial intelligences, including Artificial General Intelligence (AGI).

That’s where I would focus my energies—especially if I were granted a decade or two of life, and of a reasonably functioning mind and body. I am aware that I am well into my ninth decade on the planet.

III. Educational Applications of the Theory of Multiple Intelligences

Question: In educational research, “intelligence” often serves as an analytical handle that brings diverse issues such as learning differences, curriculum design, equity and selection, assessment, and even educational aims into a unified framework. When you first conceived the theory of multiple intelligences, had you already anticipated or considered its potential applications in education?

Answer: Good question. When I began to work on the research that led to Frames of Mind and the theory of multiple intelligences, I was in a very different place. I saw myself as a broadly-based psychologist—spanning cognitive, developmental, and neuroscience—with an interest in a range of social sciences. I did not see myself as an educator—and in fact the chapters on education at the conclusion of the book were the least developed—almost an afterthought.

I was surprised at the widespread interest in Frames of Mind and especially surprised that it evoked such interest among educators—far more than among psychologists or other scholars. I am not insensitive to how others react to something that I’ve done, and so I began to focus more on educational issues, both in the United States and abroad.  In fact, during the 1980s, when the work was first known, I made several trips to China, including a three-month stint in 1987, and much of the time was spent with educators and visiting schools. I wrote a book about my experiences called To Open Minds: Chinese Clues to the Dilemma of Contemporary Education.

While I became interested in educational issues I was hesitant to make specific educational recommendations, let alone start a school or devise a curriculum. Instead, when educators announced such ambitious interests and goals, I said I would be happy to learn from them and to help them, but that they were the educators. Indeed, I was the scholar, the social scientist who had developed ideas but that they were the ones—the informed practitioners—who could and should create and implement them. And that’s what I have tried to do for forty years.

Question: What do you consider to be the most successful applications of the theory of multiple intelligences in American classrooms, curricula, and assessment practices?

Answer: Whether in the United States or elsewhere, the best applications of MI theory draw on two concepts:

Individualization—Knowing as much as possible about each learner and presenting materials in ways that address the strengths and proclivities of each child. Of course, this is far easier when classes are small and when appropriate technological instruments are available.

Pluralization—Teachers should decide which ideas, concepts, practices are most important and then prepare to spend considerable time on them—and here is where “MI” comes in—to approach these important educational priorities in several ways. If one does this, if one approaches energy in science or revolution in history in numerous ways, one reaches more students and one gives every student more than one way to think about a topic.

Pluralization could always be done but it’s much easier to do at a time when we have powerful technology which can teach science, math, history, drawing, indeed the whole gamut of subjects and disciplines in many different ways.

In the latter years of the 20th century, I worked particularly with two schools—The Key School (later the Key Learning Community) in Indianapolis, Indiana and the New City School in St. Louis, Missouri. There is lots of written materials about both of these educational forays, see here and here, and I write about them in Multiple Intelligences: New Horizons, as well as the book I co-edited with Jie-Qi Chen and Seana Moran, Multiple Intelligences Around the World.

Question: How do you view the localized adoption of the theory of multiple intelligences in China and other East Asian regions? You visited China several times in the 1980s and discussed differences between Chinese and American approaches to arts education in To Open Minds (1989). MI theory has exerted considerable influence on China’s curriculum reforms, particularly in promoting arts education and aesthetic education, areas that have traditionally been relatively underdeveloped in the Chinese curriculum system. How do you assess these localized interpretations and applications?

Answer: I am very pleased to learn that my research ideas and my educational proposals have had influence in China and other parts of East Asia—including, in recent years, a great deal of interest in India. I make no pretense of evaluating how skillfully the work has been done. As a long-time art lover and a student of the arts, I am especially pleased to learn that aesthetic education has been bolstered—that was the principal reason for my several visits to China in the 1980s.

Around 2000, I was visiting some sites in China. I met with a journalist and asked her why there was so much interest in “MI” in China. She gave me a very memorable answer:

“In the United States, parents can look at each child and decide what to focus on in that child’s education—which intelligences to favor, to bolster. In contrast, in China, it’s just eight areas that we have to make sure that our child excels in.”

Even if this response was hyperbole, or perhaps a joke, it still indicates the greater emphasis in the West on individual differences and on individual flourishing. It may not be consistent with a Confucian or Buddhist or Communist approach to human development and flourishing.

Question: In the educational application of multiple intelligences, which aspects do you think are most prone to misunderstanding? For example, in China it is common to map school subjects directly onto specific intelligences, such as assigning mathematics and physics to logical-mathematical intelligence, language courses to linguistic intelligence, or music courses to musical intelligence. How do you view this type of simplified correspondence?

Answer: As you imply, this approach seems to be at odds with my whole educational philosophy. Of course, one can assign a particular intelligence to a particular subject matter—but if that’s all you do, nothing has been gained.

As I indicate in my answers to other questions, an “MI approach” to education entails mobilizing a student’s stronger intelligences so that they can master important concepts and processes. If one child learns history better via linguistic methods, a second child via art or music, a third child via logic, a fourth child via a focus on particular persons, heroes, villains, these should be the entry points to a better and more rounded education.

Otherwise, as you describe it, we are simply putting new labels on the traditional subjects rather than taking advantage of the fact that children have different strengths, different proclivities and tastes. Education (human and technological) should take advantage of these different profiles.

Question: The assessment of multiple intelligences has long been one of the most widely discussed issues in educational practice. Traditional examinations rely on objective and decontextualized measurements, whereas MI theory emphasizes that intelligences are potentials activated within specific cultural and situational contexts. Nevertheless, many educators still hope for a unified standard to avoid overreliance on subjective judgment, and some even attempt to use paper-and-pencil tests to assess multiple intelligences. How do you view the use of traditional assessment tools in evaluating MI? In your view, what pathways should be taken to assess multiple intelligences more appropriately?

Danfoss Universe theme park in Denmark

Answer: Ideally, the best way to assess intelligences is through a vehicle like a children’s museum—or in Denmark, a theme park called “The Explorama.” These are rich environments with lots of elements, aliments, and materials, which persons of any age can explore in their own way, as deeply as possible, pose questions, try out experiments, revisit, etc. Forty years ago, we undertook such an “MI education” with Project Spectrum—where colleagues and I created a rich preschool environment, encompassing the full range of intelligences, followed children over the course of a year, and even followed up with those children a year later. There are three books on this program and readers are well advised to consult them. (See more about these volumes here.)

As for paper and pencil tests, they are OK for the “standard intelligences”—language, logical-mathematical, and perhaps spatial—but are not appropriate for the other intelligences. I would not take seriously a test that purports to measure interpersonal or intrapersonal intelligence. Games and other rich environments provide some useful information, but evaluations are best done by individuals who know children well—parents, relatives, current and former teachers, coaches—and, if properly coached, children themselves—especially as they get older…and especially if they have good intrapersonal intelligence!

So long as you use multiple choice or other short answer instruments, you are really testing “test-taking intelligences” and not personal or bodily-kinesthetic or musical intelligences and you are assessing spatial intelligences in a suboptimal way.

Question: I have long been concerned with the question of whether the theory of multiple intelligences can be compatible with an exam-driven education system. In China, basic education is organized around competitive examinations, with a strong emphasis on logical-mathematical intelligence as the primary criterion for selecting the “best” students who will advance into higher levels of schooling. Yet according to MI theory, there is no single “most intelligent” person; individuals differ only in the areas in which their intelligences are strongest. MI aims to identify human potentials, to recognize children’s profiles of strengths at an early stage, and to reveal the diverse structures of intelligence that each child possesses. From this perspective, I sense a powerful value orientation toward educational equality embedded within the MI framework.

Answer: I appreciate this sympathetic question—which anticipates well how I think about these issues. So long as higher educational opportunities are restricted to those who do well on IQ-style measures—as has long been the case in China and other East Asian countries—you may as well just administer IQ tests and look at performances in a certain kind of traditional school.

I prefer to think about this question from a different perspective. What kind of a society do we want to have—aspire to—and how can we best achieve that kind of society, given our population and our educational options? We already have answers to this question based on the last IQ-centered century…and it’s not one that I’d like to repeat or extend!

I do think that an MI educational system and an MI-conscious society would likely be a healthier society—certainly more people would feel appreciated and would try to make contributions—including to the common good.  

To be sure, life is not fair. And some individuals have a multitude of talents and intelligences while others have far fewer. (A Leonardo da Vinci appears only rarely!) But that’s no reason not to have an equitable society—one that recognizes individual talents, aspirations, needs, deficits—and tries to accommodate them as much as possible. (This is a principal idea behind philosopher John Rawls’ important work on A Theory of Justice.) That’s the kind of society that I would like future generations—including my own family—to live in. And, just possibly, that’s the one that MI can help to bring about.

Question: While the theory of multiple intelligences emphasizes individual differences and diverse potentials, schools still need to teach certain forms of “common cultural literacy,” such as the core cultural knowledge proposed by Eric Donald Hirsch. However, common cultural knowledge often reflects the power structure of the dominant culture, whereas individualized education seeks to honor each child’s unique intelligence profile. How do you think education should balance the transmission of shared cultural reference points with the respect for intellectual diversity and individual potential? Does MI theory require a rethinking of Hirsch’s notion of cultural literacy in order to avoid cultural assimilation and the reinforcement of dominant cultural norms?

Answer: You raise two questions. It’s fine to have common cultural knowledge, but that should not be the focus on school. School should focus on developing the skills and approaches that you need to succeed in life and to be a positive part of your community—including the world community.

What Hirsch recommended is now available via a touch of any available keyboard—no need to memorize geographical locations or the names of presidents or the sites and outcomes of war. So no need to include common cultural knowledge in curriculum—just let students know it’s available and how to locate it—and importantly, what questions to ask of it…since Hirsch’s list did not pay attention to societal diversity and to changing political and cultural agendas—it was like a short version of an encyclopedia published half a century ago.

As I am reflecting on these issues, public television in the United States is broadcasting a six-part series on the War for the American Revolution (1775-1781) and the founding of the United States of America. It’s an excellent series—trying to account for the good, the bad, and the ambiguous aspects of that epoch-making event. I am reminded of the remark allegedly attributed to Premier Zhou Enlai. Asked whether the French Revolution (1789-1795) had been a success, he paused for a while and then apparently responded: “It’s too soon to tell.” 

Writing for a Chinese audience at the end of the first quarter of the 21st century, I would add: “It’s up to the human beings on the planet to determine whether we are to have good work and good citizenship.”

Question: In The Closing of the American Mind (1987), Allan Bloom argues for a return to the Western canon as a way to restore moral and spiritual order, whereas the theory of multiple intelligences emphasizes cultivating the diverse profiles of abilities that each student possesses. How do you view Bloom’s insistence on a single, “proper” path of spiritual and moral education? In your perspective, where does the deepest conflict lie between this canon-centered, elitist model of education and the MI framework that values individuality and diverse potentials? Is it possible for education to respect intellectual diversity while still maintaining a commitment to moral education and enduring spiritual traditions?

Allan Bloom

Answer: I do believe that if our planet is to survive, we need to have an education—formal or informal—that develops the moral and ethical potentials of human beings.  For thirty years, as part of the Good Work Project (now called The Good Project) my colleagues and I have been devising and administering models and curricula that help young persons to become good workers and good citizens. Those individuals need to be excellent in their work, engaged in their work, and willing and able to deal responsibly with ethical and moral dilemmas.

Those are “the Three Es of Good Work and Good Citizenship”: Excellence, Engagement, and Ethics. You can learn more about this endeavor on our website thegoodproject.org and in the ten books that we have written, including Good Work (2001) authored by Gardner, Csikszentmihalyi and Damon.

In all candor, I have to say that Bloom’s ideas—developed in the middle of the 20th century—are no longer part of discourse in any circle of which I am aware. There are perhaps strands of what he thought and wrote that might still be of interest—but, remaining in candor, I think that readers’ time would be much better spent learning about The Good Project and the curricula that we have developed for various contexts.

Question: In an era marked by rapid advances in artificial intelligence and a growing emphasis on personalized learning, do you think the theory of multiple intelligences can offer new frameworks or perspectives for the future of education?

Answer: I don’t think that MI theory or any theory can provide the answer to how best to educate in the future. But I believe that the pluralism that MI theory recognizes and promotes is precisely what we need if we are to have a peaceful and healthy planet, that makes a place for educating all human beings to the best of their potential. In that sense, it is entirely appropriate that the grant that supported the research for Frames of Mind was a strand of a larger project at Harvard, called “The Project on Human Potential.” If the ideas that we developed on that project contribute to better education for all persons, that would be my dream.

Interviewer: Thank you very much for your generous and illuminating reflections. Your answers make it clear that the value of the theory of multiple intelligences has never been about offering a fixed educational formula, but rather about reminding us to acknowledge the diverse potentials of human beings and to embrace a more open and inclusive way of understanding learning, ability, and development. In an era of rapidly evolving artificial intelligence, this exploration of “human possibilities” feels more urgent than ever. I hope to bring your insights back to educational research and practice in China, so that more students may be understood, supported, and empowered because of their unique potentials. Thank you again for your time, and for the intellectual light you continue to offer to all who care about human development.

AI and Kinesthetic Intelligence

© Mia Keinanen 2025

Martha Graham

“The body never lies,” said Martha Graham, one of modern dance’s most eminent pioneers. Her company recently launched its centennial celebrations. Graham’s claim suggests that a person's movements and expressions will reveal their true emotions and intentions—even when words fall short. Indeed, according to Howard Gardner’s theory, our intelligences are multiple, each representing different ways of processing information and solving problems. 

I want to tweak Howard’s formulation: What if the other intelligences (linguistic, logical-mathematical, musical, spatial, intrapersonal, interpersonal and naturalistic) cannot exist without the kinesthetic intelligence, without the human body? Indeed, in the race toward General Artificial Intelligence (AGI), this very quandary has been a major challenge. It’s been termed Moravec’s paradox: While AI can solve complex cognitive tasks, the sensorimotor capacity or the ability to generalize skills may be much lower than that of a child.

Case in point: Last year at the opening of MIT’s Quest for Intelligence (November 2024), an interdisciplinary center that probes how the brain produces intellect and how it could be replicated in artificial systems, a claim was made: “natural intelligence” still trumps artificial intelligence. By natural intelligence, they refer to kind of intelligence that biological organisms—especially humans—possess. Natural intelligence leverages sensory data to construct internal models of the world, enabling effective generalization from minimal information. Natural intelligence can make a lot from a little. In intriguing contrast, artificial intelligence, or Large Language Models (LLMs) make a little from a lot by synthesizing vast amounts of data succinctly.

The question arises: What if embodiment allows us to access deeper truths that are not expressed linguistically or mathematically? Indeed, as I construe it, cognition is contextual. It blossoms from our physical interactions with the world—put differently, the human body-in-context makes cognition possible. Yet, the human body is conspicuously absent from the LLMs that could make AGI possible. At the aforementioned conference, the mission statement proclaimed: “Imagine if we could build a machine that grows into intelligence the way a person does—starts like a baby and learns like a child…Imagine systems, like ChatGPT, which you could ask any question and expect to get reasonable answers—but inside they are not just black boxes trained on all of humanity’s data. Instead, they are our best current models of how human minds and brains think and communicate.” Assuming this stance: No human mind thinks or communicates without a body; the body is not there just to carry our heads around.

Boston Dynamics made a video of robots dancing to the song, “Do You Love Me?” (by The Contours)—and the presentation went viral. Their three kinds of robots (humanoid, robot dog, and warehouse robot) perform their dance moves in a synchronized manner. The feat required advanced robotics engineering, precise programming and sophisticated motion planning. 1X, a Norwegian humanoid robot start up, on the other hand, believes in building robots that have a “human form factor”. These entities adapt to the world humans have built and learn accordingly from humans by co-existing with us—just as a child does. We mere mortals have not seen these robots dance yet, but surely this’ll happen soon. (Children begin dancing as early as 9 months of age and many never stop!)

Neo 1X humanoid robot promotional photo

These contrasting examples foreground two schools of thought for the AGI effort. Boston Dynamics maintains that the internet already contains enough data—especially when paired with the vast amount of online video, massive computational power, and clever engineering—to eventually reach the holy grail of AGI. In contrast, 1X contends that true general intelligence requires direct interaction with the physical world. It requires learning through human form in the environment designed for humans.

Two weeks ago, 1X’s launch of the humanoid robot Neo sparked a global frenzy. Neo is designed to interact with humans in a highly lifelike and responsive manner; it is also soft and weighs only 114 pounds. Bernt Kristensen, the founder of 1X, also underscores that the robots must reflect human ethical values. Humanoid robots need to be accountable in how they engage with people and that human-robot interaction (HRI) should prioritize empathy and mutual respect.

Memes spread instantly, The Wall Street Journal put it on the front page, and The Daily Show with Jon Stewart roasted it—the ultimate badge of cultural relevance. Why the hype? For the first time, a humanoid robot opened for public preorders. With humanoids projected to become more common in households than cars, this marks a major milestone: the beginning of true human–humanoid coexistence, where we can finally learn from one another. Moreover, as humanoids learn by mimicking human actions and “learning by doing,” it could offer fresh insights into AI development—and possibly bring us closer to Artificial General Intelligence (AGI).

As a former professional dancer, (and also, one of Howard’s doctoral students), I believe that body and movement are not mere afterthoughts—they are fundamental to the nature of intelligence itself.  What’s most fascinating about AI vis-à-vis kinesthetic intelligence is that AI has no body, and the body doesn’t compute—yet we humans do both. One mode is lost in ecstatic dance, the other in deep thought. Yet it’s the body’s movement that gave rise to cognition, which, no matter how abstract, remains grounded in the physical self. And the body does not know how to lie.

As Howard noted in his recent talk with Anthea Roberts, AI may replicate the disciplined, synthesizing, and even aspects of the creative mind—but the respectful and ethical minds remain firmly human responsibilities. I agree, AI requires the human element (feet on the ground) to anchor its power and potential in terms of real-world human values, needs, and concerns. The true value of AI is as a mirror—AI will not be about our escape, but rather our striving to grasp, know, and appreciate the human condition more deeply.

As Anthea Roberts pointed out, because AI is being trained on the vast pool of written and visual content available online, creative professionals need new strategies to protect their work. Indeed, this is a chilling point which urgently needs attention and resolution. I note that, for once, dance is not the lowest on the totem pole. It cannot be copied and reproduced as literature, visual art, and design can be—at least not in terms of live performance. As such, it’s not equally threatened by AI. I predict a boom in live performance and sports, as people seek out the thrill of watching trained human bodies do the extraordinary—raw, unpredictable, and beyond what machines or AI-generated TV series or movies can offer. We feel it in our bodies.

Humanoid robots may one day outshine us—not because they are us, but because they aren’t. Unbound by the limits of flesh, their movements may be flawless, their emotions simulated with precision. Yet for all that perfection, will we truly feel them? Perhaps, paradoxically, we’ll connect even more—drawn to an "upgraded" version of our own emotional blueprint. Either way, for now, I’ll take the fragile, fleeting beauty of a live dance—which, by the way, I’m off to see tonight.

 

ACKNOWLEDGMENTS

For comments on earlier drafts, I thank Howard Gardner, Torkel Engeness, and Noah Riskin.

FOOTNOTE

“Natural intelligence” bears some similarity to what Howard Gardner has termed the “naturalist intelligence.” See Intelligence Reframed (1999) and other writings.

For those interested in further reading: “AI Business Review.” Aibusinessreview.org, 20 Feb. 2026, www.aibusinessreview.org/2026/02/20/fei-fei-li-spatial-intelligence/. Accessed 10 Mar. 2026.

Multiple Intelligences: New Strands of Evidence from Neuroscience

Howard Gardner © 2025

Some necessary background

Harvard Magazine Cover, September - October, 1990.

Caption: “Brown with re research associates Benjamin Gardner, Wolfgang, and Ernie.”

It’s well over half a century ago—in 1969—since I, a budding psychologist, first became interested in the question: Does the human BRAIN have anything to tell us about the human MIND? With the benefit of hindsight, you might think that was a stupid or ill-considered or unnecessary question. But at the time, the case was actually the opposite—from the behaviorist B. F. Skinner to my own beloved adviser Roger Brown, I received a clear message: We psychologists should leave the study of the brain to those who were interested in rats or, perhaps, in the sense of smell.

Norman Geschwind

At any rate, in that fateful year, I became acquainted with the writings of Norman Geschwind, a behavioral neurologist. I was fascinated by what psychologists might learn

from the study of patients who had suffered a particular form of brain injury—an accident, a stroke, tumor—and in turn had significant impairment or even the loss of a particular cognitive capacity—say, aphasia (language), alexia (reading), amusia (music) , prosopagnosia (facial recognition), and the like. I also learned about the amazing insights that were emerging from the study of individuals whose brains had been surgically split into two halves in an effort to control epilepsy!

Making a life-changing decision, I decided to learn about the human brain, and I asked Dr. Geschwind whether he would be my preceptor. Having received a positive response and a post-doctoral fellowship, I had the opportunity to work at a hospital alongside a ward of brain-damaged patients—the Boston Veterans Administration Hospital (BVAH).

By the middle 1970s, I thought that I had learned enough to share some insights and conclusions with readers. I wrote and published The Shattered Mind, with a tantalizing subtitle: The Person after Brain Damage.

I could have moved on, but I felt that there was more to say on the topic. (In fact, I worked at the BVAH for another 15 years). In particular, I was interested in exploring how knowledge of the brain—and of various forms and sequelae of brain damage—might illuminate the “kinds of minds” that members of our species possess and can elect to develop.

Thanks to the awarding of a major grant to the Harvard Graduate School of Education—one that provided five years of funding for several of us—I was able to undertake a major study. That concentrated research undergirded the book for which I remain best known: Frames of Mind:  The Theory of Multiple Intelligences. In that book, based in significant measure on evidence from studies of the effects of brain damage, I contend that human beings have evolved over the millennia to have at least seven different forms of cognition: linguistic, logical-mathematical, musical, spatial, bodily-kinesthetic, interpersonal, and intrapersonal—the so-called “multiple intelligences.” Even though I—along with many others—have pondered the possibility of adding intelligences, in the intervening years, I have confirmed only one additional intelligence:  a form that I’ve dubbed “naturalist intelligence.”

Multiple Intelligences Infographic

From time to time, I spoke of the intelligences as “computational capacities.” As I’ve conceived it, the mind/brain is best thought of as a set of semi-independent and semi-autonomous computational devices. A few individuals—call them omnibus geniuses or “da Vincis”—may be blessed with a set of seven or eight excellent computers. Some unfortunate individuals may have a set of computers that do not work well at all. But most of us constitute a mixed computational bag: some intelligences are strong, others are average, still others do not operate well at all. (To a person persuaded by the validity of “MI theory,” it should come as no surprise that many persons who are a whiz at logic and mathematics seem to be deficient in understanding of other persons… or of themselves.)

One assumption that I was well aware of half a century ago: The intelligences, the computational systems, are not necessarily yoked to a single sensory system. Most of us may rely on our visual system for spatial tasks, but one can be spatially skilled even when one is blind. Most of us use our auditory system for language, but one can be linguistically competent if one is deaf—by signing, learning to read braille, and so on. 

The one exception to this pattern seems to be musical intelligence. Certainly, the making and appreciation of music is enhanced—and perhaps enabled—by having adequate hearing. And yet, upon more careful examination, many aspects of music—rhythm, texture, accent—can be appreciated even by individuals who are deaf. By the same token, acute hearing does not in any way predict one’s musical capacity—we all know individuals whose hearing is otherwise fine but who are considered tone-deaf… or at least asked to listen or “mouth lyrics” while participating in a chorus. The 20th century Russian composer Alexander Scriabin was a firm believer in synesthesia—the integral co-operation of various sensory systems. Some of his most valued pieces were composed (and are performed in our time) along with complex visual forms that are carefully intertwined with streams of sound.  Even an individual who is totally deaf can appreciate most facets of Scriabin’s music; just as individuals who are blind are able to appreciate sculpture and, with appropriate tutelage, many aspects of drawings, paintings, murals, and the like.

So much for background!

Intelligences vs. sensory systems

At the time that the idea of multiple intelligences was first put forth, I could have made most of these points. And from the beginning, I objected strenuously when individuals—often with the best of motivation or intention—spoke about “visual” intelligence or “visual” learners. These spokespersons confused intelligences with sensory systems. I had always insisted that these hypothesized computational capacities were NOT indissolubly yoked to the eyes, the ears, a sense of touch, or a sense of smell.

But at the time that I was first learning about the cognitive effects of damage to the cortex (and other neural regions), our knowledge of these systems was modest. At most, when someone had a stroke, we could look at a CT scan or at patterns of EEGs, various forms of FMRI, or other more sophisticated brain-observing technologies were yet to be widely available. 

Now of course, we have many ways of studying the damaged brain. Indeed, even the brains of individuals who appear to be perfectly well can be studied while these individuals are engaged in a wide range of tasks. These technological advances make it possible both to understand the ways in which the human brain (and its associated sensory systems) ordinarily acquire skills, as well as the ways in which those whose brains (or sensory systems) were abnormal from birth are nonetheless able to navigate the world reasonably well.

Case in point: As it happens, I am visually very impaired: near-sighted, incapable of stereoscopic vision, color-blind, and prosopagnosic (unable to recognize faces)—and yet I have somehow compensated for these handicaps—even passing the color-recognition portion of the test for obtaining a license to drive. And much of my research has been about the arts—including the visual arts.

New ways of thinking about cognition

Miriam Hauptman

As one who has continued to ponder these issues, I was pleased to receive an insightful article by cognitive neuroscientist Miriam Hauptman and her colleagues Yun-Fei Liu, and Marina Bedny. These researchers describe the forms of cognition that are nonetheless enabled even when the normal or typical modes of learning and practice are not available to an individual (or, indeed, to groups of individuals). Human-typical functioning depends upon neural systems that have been prepared by many centuries of evolution. But those systems have the capacity termed plasticity—such flexibility enables members of our species to engage in cognitive activities that were invented in recent history and/or to engage in ways that had not been anticipated by evolutionary history.

Consider these examples: Most of us learn language relying on our auditory system. But individuals who are deaf for various reasons readily learn the signed system of their culture—and if such a system does not exist, they may actually develop a usable sign language. Most of us navigate using our visual systems, but individuals who are blind use tools like canes for navigation, braille for reading, and screen readers for searching the internet.

What we are now coming to understand—at the neural level as well as at the experiential level—is that important tasks are not necessarily associated with specific neural substrates. Rather, cortical systems are able to carry out abstract computations that can apply across disparate domains of knowledge and various types of information. As the authors put it: “Neural wetware can acquire different software as a function of experience, that is, that there are no immutable cognitive wirings. Cognitive flexibility and [my emphasis] specialization of cortical circuits co-exist,” (Hauptman, et al., 2025). Or, to choose another example: “social learning” (HG: in my terminology: interpersonal intelligence) depends on multiple neurocognitive systems, each with a different neurobiological substrate, including the so-called mentalizing system, which supports our understanding of the minds and actions of others.”

Were I to stop here, I could claim that each of the so-called intelligences can draw on circuits “prepared” by evolution, while having the option of mobilizing other ones. But the authors also posit the existence of domain-general reasoning abilities that are supported by frontoparietal circuits and prefrontal cortex. These neural structures enable the mastery and use of abstract rules and deductive reasoning—which, as the authors point out, are important for mathematical thinking and computer programming.

Scholars critical of “MI theory” could point to these structures as key to IQ—but in my terminology, these capacities simply constitute logical-mathematical (one form of) intelligence. They are not key to the several other forms of intelligences, ranging from musical to intrapersonal. Absent other developed neural geography, individuals with highly honed computational skills might well qualify as the classical savants but would be unable to perform proficiently in other domains. And of course, to the extent that Large Language Instruments can carry out such forms of computation more proficiently than all members of our species, the importance of such computational capacities for survival and thriving are undermined… while other less classically computational forms of intelligence may come to be more valorized.

Stepping back

As one who became fascinated over half a century ago by the study of the brain, I’m gratified when I have the opportunity to update my knowledge—and, I hope, my understanding—of human cognition and the human mind. I am grateful that the basic intuitions of “MI theory” seem to have withstood the passage of time reasonably well. I am equally gratified that the continued study of the brain—its evolution, its plasticity, its connections—has deepened our understanding of the human mind in its dazzling complexity.

It’s only fair to note that 2025 is not the same as 1969 or 1983. Nowadays, when my colleagues and I study intellect, we look not only at human intelligences, but also at animal intelligences, plant intelligences, and artificial intelligence. (See the blog my colleagues and I wrote on the topic here.)

It’s also important to point out that, in the second quarter of the 21st century, with artificial intelligence evolving so rapidly, the educational landscape will doubtless change in ways that are scarcely imaginable (Read my most recent blogs on this here, here, and here.) It’s up to those of us enmeshed in scholarship to nuance our views and conclusions in the light of new evidence, and it will be task of historians—should any remain!—to point out when we were confused, when we were wrong, when we were on to something, and when we saw further ahead.

Acknowledgements

For their useful comments on an earlier draft of this essay, I thank Miriam Hauptman, Annie Stachura, and Ellen Winner

References

Gardner, H. (1975). The shattered mind: the person after brain damage. Knopf.

Gardner H. (1983). Frames of mind: The theory of multiple intelligences. Basic Books.

Gardner, H. (2026). Introduction from Frames of mind: The theory of multiple intelligences. Basic Books.

Hauptman, M. et al. (2024). Built to adapt: Mechanisms of Cognitive Flexibility in the Human Brain. Annual Review of Developmental Psychology. 6, 133-162.

To Stream or Not to Stream: Academic Tracking and Benefits of an MI-based Approach

Contributed by Annie Stachura and Branton Shearer, in response to Ahmadie Thaha

Over the years, we’ve received many suggestions for how to use the concept of multiple intelligences in the classroom. These proposals arrive in our inbox from all over the world; they often present us with new contexts, and thus, new lenses for considering what interventions might support student learning and achievement in different arenas. We try our best to lend a helpful ear to these correspondents—to acknowledge their hard work and offer support and guidance where we can.

Recently, we read a thoughtful piece that appeared in KBA News, titled Reborned Academic Streams: Reviving the Stigma. Columnist Ahmadie Thaha takes issue with a new policy being contemplated by Indonesia’s Ministry of Education, Culture, Research, and Technology. As described, this policy reinstates academic tracking or “streaming”—requiring high school students to select and stick to one area of study, effectively narrowing their future career path.

The policy clashes with the still-nascent “Merdeka Curriculum” (meaning “the freedom curriculum”)—a recently developed framework that prioritizes flexibility, individuation, long-term growth, and comprehensive development in the classroom. Critics of the Merdeka Curriculum worry that this freedom can beget confusion for students, who may ultimately end up in the “wrong” field due to indecision, lack of knowledge of their areas of strength, and/or societal pressures. As described, The Ministry of Education’s new academic streaming policy seeks to resolve these concerns.

Logo for Indonesia’s Ministry of Education, Culture, Research, and Technology

Thaha criticizes the restrictive policy-in-progress and warns of its consequences. He quips: “It’s like saying, ‘Because so many people choose the wrong life partner, marriage will now be assigned by the neighborhood head.’”

Thaha puts forth an alternative way of determining a student’s strengths: an interest and aptitude assessment he’s designed based around multiple intelligences. 90 “Yes/No” questions aim at yielding a reasonably accurate map of a student’s various potentials. In addition, rather than yielding an inflexible stream, the assessment should be used in conjunction with structured observation to curate personalized learning strategies and help students gain insight into their own skills and possible vocations. Thaha asserts that this type of support system would be well-worth the time and effort it would take to implement.

The article concludes with a moving call-to-action from Thaha:

We’re not against structure—we oppose lazy structure. We’re not anti-tracking—we’re anti-boxing. If students need a framework, let’s provide one. But it should be flexible—less like a Victorian corset that restricts breathing, more like an adaptable map.

Tracking should be a guide, not a hammer. Let science students study sociology, let social studies kids peek into math. The world isn’t as black-and-white as our textbooks.

We found this essay thoughtful and well-considered, and believe it put forward concrete ideas about assessment and how it could help with educational and vocational planning.

To get his take, we consulted with our colleague Branton Shearer. Branton is a researcher who many years ago developed the MIDAS (Multiple Intelligences Developmental Assessment Scales). This instrument assesses the range of intelligences in students and, upon request, connects the findings to study and career options. We shared Ahmadie Thaha’s pieces for KBA News with Branton, and here’s what he had to say:

Branton Shearer

I am writing to lend my support and experience to Ahmadie Thaha’s suggestion that Indonesia adopt a multiple intelligences inspired approach to their Merdeka Curriculum. Guiding students through their education to a successful adult role and career path is a daunting task. There are no simple answers. Parents are the primary guides along with other family members, traditionally; but school personnel can be crucial. This is especially true in our high-tech digital information era when the jobs of tomorrow may look very different from what their parents do today.

[...]

I observed firsthand the benefits of using an MI assessment to guide middle school students (13+ years old) as they prepared to enter high school. Students completed an MI inventory (www.MIResearch.org) and were given short lists of jobs / careers matched to their top two MI strengths. They were encouraged to choose an 8 hour Career Shadowing Experience matched to a combination of their strengths. At the end of the program, students who did this were overwhelmingly more likely to have a satisfying positive experience and gain self-insight into potential career paths.

This finding was echoed during a 10-year project with undeclared university students who were confused about their major course of study. Every semester 75% or more of students reported that learning about their MI profiles enhanced both self-understanding and the fit with career requirements. Useful college courses to promote skill development were also highlighted.

The multiple intelligences have been called “the brain’s tool kit” and every brain is uniquely configured. We all have our particular strengths and limitations that can change given time and attention. Research has found that adults may have several distinct careers over the course of a lifetime so it makes sense to educate students about their own “MI Tool Kit” that can be employed to achieve success in an array of professions. The eight intelligences are much more than mere hobbies or enjoyable talents.

Another advantage of MI is that it provides a common-sense language to describe “thinking-in-everyday-life”, classrooms, and on-the-job. The MI terms can be scaled to fit across the lifespan and situation. As Thaha states, an MI profile used for career planning can also be employed by classroom teachers (and parents) to promote academic achievement. An MI profile is truly a multi-purpose power tool.

That’s the good news. One has to concede there are frequently barriers and difficulties integrating MI into a traditional school program. The retrofitting process can be perilous. It takes resilient and creative leadership to make it happen and then to sustain it.

The Indonesian government is to be commended for focusing the attention of their Department of Education on helping young people to gain career awareness and direction. [...] An MI assessment can be a powerful framework to inform a “dialogue of discovery” among the student, parents and classroom performance. The goals of this collaborative conversation are to deepen self-understanding (intrapersonal intelligence) and awareness of the skills that successful professionals use (interpersonal) and select the next best steps. From this dialogue a career development plan can emerge with a high probability of success.  

This is not always a quick or easy process. Only the lucky few know at a young age what their career path will entail and how to stick to it. For the majority of us, it is a constant adventure that zigs and zags through life. When we coach young people with a deep and realistic MI profile there will be fewer dead ends, higher intrinsic motivation and many benefits to society in the long run.

Concluding Thoughts

In agreement with Ahmadie Thaha and Branton Shearer, we hope that the Indonesian government will reconsider implementing its policy to reintroduce academic streams, which limit students’ ability to explore various subjects. As Thaha writes, this new system will not only cause stress to pick the “right” career far too early on for most, but also prevent students from gaining knowledge in diverse arenas and—in the process—becoming interdisciplinary individuals, capable of excelling in more than just one field.

MI Assessments of the sort endorsed by Thaha and Shearer would help students identify their interests and strengths and choose fitting subjects and careers, much better than standardized testing can. However, for a few reasons, it’s important to be cautious about self-assessment:

  1. There is no reason to believe that most people, especially young students still developing their senses of selves, have particular insights into their own strengths;

  2. Most persons do not understand the differences between what you like to do (preferences), what you are interested in, and how powerful your computational capacities are. Only the latter indicates the strength of an intelligence.

This being said, we recognize the desire for some kind of instrument—such as the one proposed by Thaha, or the MIDAS process. And we do see the benefits of using data, particularly when discussing with policy-makers the pros and cons of different curricular and assessment options.  

We recommend a system that relies heavily on triangulation—that is, using more than one source of data. For example, if individuals rate themselves on their intelligences, but one also obtains ratings from those who know them well (family, friends, present and former teachers), the profile of intelligences would be more reliable.

The gold standard consists of performance measurements: in this type of assessment, you have the opportunity to demonstrate your intelligence profile and not just testify to it. As examples: Assess interpersonal intelligences by observing how a person handles a conflict situation or motivates others to pursue a certain course of action. Assess spatial intelligence by seeing how quickly a person masters an unfamiliar geographical terrain and how accurately he or she remembers it. 

If Thaha’s suggestion of an MI-based assessment were implemented, flexibility is imperative so that students don’t become “trapped” in one path. In general, we believe that a broad liberal arts education (as long practiced in Western countries) focused on nurturing students’ interdisciplinary potentials strengths are more valuable than those that pipeline students—“from cradle to grave.” 

 

RESOURCES SHARED BY BRANTON SHEARER

Research and application reports on MI and The MIDAS assessment: https://miresearch.org/midas-archives/research/research-reports-papers/

Shearer, C. B. (2001). Enhancing a Career Exploration Program for 8th Grade Students with an Assessment for the Multiple Intelligences. Presented at AERA Annual Conference in Seattle, WA on April 12, 2001. Republished on ERIC. https://files.eric.ed.gov/fulltext/ED452212.pdf

Shearer, C. B., Luzzo, D. (2009). Exploring the application of multiple intelligences theory to career counseling.  Career Development Quarterly58, 1, 3 – 13. https://doi.org/10.1002/j.2161-0045.2009.tb00169.x     

Shearer, C. B. (2011). Exploring the relationship between intrapersonal intelligence and university students’ career confusion: Implications for counseling, academic success, and school-to-career transition. Journal of Employment Counseling46, 52-61. https://doi.org/10.1002/j.2161-1920.2009.tb00067.x

The Conversation that I Wish I’d Had with Leonard Bernstein

Howard Gardner © 2025

When I—as a promising young pianist—was growing up in Scranton Pennsylvania in the 1950s, there was no figure in contemporary classical music who was as salient, as charismatic, as Leonard Bernstein (almost always called his nickname “Lenny.”) Lenny was already legendary: he had become an instant celebrity in 1943 when on a few hours notice, he had skillfully conducted the New York Philharmonic Orchestra. Lenny was a composer of serious classical music; the brilliant creator of the scores for West Side Story and On the Waterfront; the much acclaimed conductor of the “New York Phil;” and an educator in the arts without equal. Millions of young people, including me, looked forward on Saturday morning television to his introductions to the world of classical music.

Also, as I grew older, I sensed personal ties. Like me, Lenny was Jewish and proudly so. He attended Harvard College (as did I) at a time when not many Jews had been admitted. He was also sympathetic to left-wing causes and critical of authoritarian regimes. In the mid-1970s, at the height of his unequaled career, he returned to Harvard to give the prestigious Charles Eliot Norton lectures—where he sought to analyze classical music using the linguistic theories of Noam Chomsky. I eagerly attended his lectures. And then, after the final lecture, I was privileged to be invited with a small group of friends and colleagues to a conversation with “Maestro.” Without question, he was the star—and he acted as one might have expected. Probably too intimidated to say anything substantive, let alone argumentative, I did engage in a brief conversation with him—though I would be amazed if it made any impression on him whatsoever.

Bernstein did not take good care of himself physically—to put it mildly—and he died too early, at the age of 72. A year or two before his death, he performed Brahms’ First Symphony at Tanglewood— the greatest live performance that I’d ever heard—perhaps of any major symphony. And it turned out that Joseph Horowitz, a classical music critic for The New York Times agreed with that evaluation.

Sadly, today most young people know of Bernstein only through the recent sensational biopic Maestro. (Alas, that’s the way that posterity will remember most personalities—JFK, Churchill—and most events—Pearl Harbor, The Titanic, 9/11.)

Little did I ever expect that I was linked to Leonard Bernstein in another way. In turns out that shortly before his death, Bernstein gave a rare lengthy interview to Jonathan Cott, a journalist quite knowledgeable about classical music. Among the many things Bernstein mentioned, was this:

“In an essay entitled ‘Children’s Conceptions (and Misconceptions) of the Arts,’ the psychologist Howard Gardner wrote: ‘We would not expect children to learn to understand computers by having them examine a terminal or a printout. Yet that is the way we expect the young to become sensitive to ballet, theater, and the visual arts. Schools bus them to plays and museums; Leonard Bernstein offers youth concerts on television; and somehow artistic understanding is supposed to result.’ And Gardner doesn’t seem to believe that it will. But, as I said in the six lectures I gave at Harvard University in 1973 [published in the book The Unanswered Question], all kids are born with a language and a musical competence. Otherwise you wouldn’t be able to account for a two-year-old child’s saying, ‘I like the green ice cream better’ in any language, whether it’s in Swahili or in Dutch. Every child can say it in the language of its parents—'I like the green ice cream better.’ That’s the Pentecostal alphabet I was speaking about before—the letters of fire that God gave us. The greatest gift he could give man was the ability to talk and communicate. And a big part of communication is music. 

Every kid is born with a sense of rhythm and has the ability to tune in on the overtone series. It’s part of the air we breathe, part of our bodies. The harmonic series is in everybody—the octave, the fifth, the fourth, the third, the major and minor seconds. This is provable through physical principles. An infant knows the interval of an octave because his or her mother sings a note or a melody one octave higher than the father does. And every kid knows the fifth, and every kid knows the first two different overtones of the harmonic series. In every country of the world, kids tease each other with the same tune: nya-nyanya-nya—the first two different overtones of the harmonic series. And every child is born with the knowledge of one-two, one-two—he has two hands, two feet, two eyes, he knows two nipples in his mother’s breasts, he breathes in and out, he knows up and down, left and right, and he can march: toddle-toddle, toddle-toddle!”

This blog introduces my first mea culpa. Foolishly and unnecessarily, I had “put down” the very young persons’ concerts from which I – and countless others—had gained so much. I had been searching for examples of “one night stands” in the arts—attending ONE art exhibit, participating in ONE dance, attending ONE concert. Unless the child is unusually gifted or has prior interests and inclinations that are consonant with that single exposure, it’s unlikely to have much effect…let alone any long-term impact. But of course, Lenny’s Saturday morning shows were designed to be seen as a whole series—and not as a one-shot exposure. A poorly selected and insufficiently articulated example on my part. No doubt there are scattered examples of young people who are inspired by a single powerful artistic experience—“chance favors the prepared mind.”

On the other hand: Bernstein was actually dwelling on another one of his favorite points—one brought up frequently over the course of the Cott Conversations. And that is his theory of cognitive development—a field in which I do have expertise. Bernstein wanted to expound—indeed to glorify—the mind and spirit of every child: openness to new experiences; willingness to play (physically or mentally) with whatever information is presented; often a keen memory for sensory and motor experiences; a desire to create, to build upon, whatever information proves enticing.

In the best of all worlds, Bernstein’s depiction has validity. As so many developmental psychologists have demonstrated, the minds—and the brains—of young children are amazing entities. But two qualifying points are necessary:

Bernstein with wife Felicia Montealegre and two of their children

  1. Most children throughout the world and throughout history were not born and raised in supportive conditions—and so the ideal just sketched is unlikely to be realized. Bernstein may well have had in mind his own three children—who had heredity (their mother was an internationally acclaimed singer), environment, historical era, and location (the upper West side of Manhattan) on their side.

  2. There are profound individual differences in cognitive profiles. While even children who are hearing-impaired can gain something from music or dance, not every child has—or can develop—a notable musical intelligence. Indeed, I’ve been married to two women—both of them far more artistic than I am or could ever be—but neither of them was able to carry or recognize a familiar tune. And while Leonard Bernstein’s concerts might well have appealed to them, they would likely not have benefited from them in the way that I (and others with ampler amounts of musical intelligence) could and did.

Being immigrants who had escaped from Nazi Germany in the nick of time, my parents had no money and no time for the arts. But when I was five years old, we visited a family in the neighborhood, and I simply began to pick out tunes on the piano. A parent in that family said “You must buy a piano for Howard!”—and so my parents scrounged up $30 (yes, thirty dollars!), bought an upright piano, and supported my piano lessons for the next six years—at which time, I stopped formal lessons but continued to play, and even to teach piano for decades. So, I was the kind of child that Lenny had in mind—but he universalized “musical intelligence” in a way that was aspirational rather than typical.

Lenny is long dead, and I am well into my ninth decade. We’ll not be able to continue this conversation, but thanks to the internet, I’ve at least had the chance to respond to one of my lifelong heroes.


ACKNOWLEDGMENTS

For comments on an earlier version of this blog, I thank Shinri Furuzawa, Annie Stachura, and Ellen Winner.

REFERENCES

Bernstein, L. (1976). The unanswered question: Six talks at Harvard. Harvard University Press.

Cott, J. (2013). Dinner with Lenny: The last long interview with Leonard Bernstein. Oxford University Press.

Horowitz, J. (1985, July 22). Music at Tanglewood: Bernstein and Brahms. The New York Times. https://www.nytimes.com/1985/07/22/arts/music-at-tanglewood-bernstein-and-brahms.html