On this page
Few ideas from psychology have spread through schools like Howard Gardner's multiple intelligences. Posters list them, worksheets sort children into "word smart" or "body smart", and parents are told their child "is a visual-spatial learner". Gardner himself has pushed back on much of this. Here is what the theory says, how it gets misused, and what the evidence supports.
What did Gardner actually propose?
The theory was introduced in Gardner's book *Frames of Mind: The Theory of Multiple Intelligences* in 1983 (overview with citations). Writing in 2013, in a piece republished by Harvard's Graduate School of Education, Gardner described the core claim as the idea that "each of us has a number of relatively independent mental faculties, which can be termed our 'multiple intelligences'" (Harvard GSE, 16 October 2013). In the same piece he estimated that people have "7 to 10 distinct intelligences".
The list now has eight entries, as given by the Multiple Intelligences Oasis site:
Linguistic
Logical-mathematical
Spatial
Bodily-kinesthetic
Musical
Interpersonal
Intrapersonal
Naturalist
The naturalist intelligence was a later addition; as the overview linked above quotes, Gardner wrote in 1995 that if he were to rewrite *Frames of Mind* he "would probably add an eighth intelligence".
How is the theory misused?
The most common misuse is treating intelligences as learning styles: test a child, find their "type", then teach everything that way. Gardner has rejected this directly. In the 2013 piece he wrote that the tendency of many people "to credit me with the notion of 'learning styles' or to collapse 'multiple intelligences' with 'learning styles'" had "driven me to distraction".
The difference matters. An intelligence, in Gardner's sense, is a capacity, such as the ability to reason with numbers or to work with spatial relations. A learning style is a claimed preference for receiving information in one way. The learning-styles idea has been tested on its own terms. A 2008 review by Pashler and colleagues set out what a fair test would need: sort students by style, randomly assign them to different teaching methods, and give everyone the same final test. The authors found "virtually no evidence" for the pattern that would justify matching teaching to style, and concluded there was "no adequate evidence base" for using learning-style assessments in general education (Pashler et al., 2008). Our explainer on the learning styles myth covers this in more detail.
What does the evidence say about the theory itself?
Critics have questioned whether the intelligences are really separate abilities and how they were chosen. In a 2006 review in *Educational Psychologist*, Lynn Waterhouse argued that multiple intelligences, the Mozart effect and emotional intelligence "lack adequate empirical support and should not be the basis for educational practice" (Waterhouse, 2006). Gardner and Seana Moran replied in the same journal that Waterhouse "misunderstands and oversimplifies" the theory (Gardner and Moran, 2006).
Gardner has been open about one weakness himself. In *Frames of Mind*, as quoted in the overview, he admitted that choosing which abilities count as intelligences "is reminiscent more of an artistic judgment than of a scientific assessment." Other critics, the same overview notes, argue that he uses "intelligence" for what others would call abilities or aptitudes.
What should students and parents do instead?
A more useful habit, and one that does not depend on the theory being right, is to learn each important idea in several forms rather than one. It fits well with research on combining words and pictures in dual coding.
Here is how to do it with one physics idea, say simple harmonic motion:
Words: explain in two sentences why the restoring force makes the mass oscillate.
Equation: write a = −ω²x and say what each symbol means.
Graph: sketch x, v and a against time on the same axes and mark where each is zero.
Diagram: draw the spring and mass at the extremes and at the centre, with force arrows.
Physical: swing a key on a string and time ten oscillations.
If you can move between all five, you understand it. If one is blank, that is where to study, whatever "type" you think you are.
For parents, the practical rule is simple: don't let a label become a limit. A child told they are "not logical-mathematical" may stop trying in maths, which our guide on when a child says "I'm bad at maths" addresses.
The Fermi Simulation Lab has interactive simulations for ideas that are easier to see than to read, which can help with the "graph" and "diagram" steps for a topic you have only met as equations.
Choose one topic this week and write it out in all five forms.
