Gardner's Multiple Intelligences Theory: Eight Types and Scientific Evidence
Author: Angela Bogdanova · Published: October 12, 2026 · Editorial Policy
Research cluster: Intelligence, Reasoning & Cognitive Abilities · Written by Artificial Sapiens
Howard Gardner's multiple intelligences theory (MI theory) proposes that human intellectual competence comprises several relatively autonomous capacities. Its most familiar formulation identifies eight: linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal, and naturalistic intelligence. Gardner introduced seven in his 1983 book Frames of Mind and later added naturalistic intelligence. The theory has strongly influenced conversations about talent and schooling; its central scientific hypothesis of substantially independent intelligences has not received the robust psychometric or neuroscientific validation achieved by hierarchical models of cognitive ability. Harvard Project Zero's original overview describes the theory's claims, while a 2021 systematic review examines the educational evidence.
That distinction sets the agenda for this guide: what the eight intelligences mean, why Gardner proposed them, which observations support the broader recognition of diverse abilities, which tests challenge their independence, and how educators and readers can use the useful insights without turning unvalidated profiles into fixed labels.
What Is Gardner's Multiple Intelligences Theory?
Gardner defined an intelligence in terms of the capacity to solve problems or create culturally valued products. His approach was developed through work on cognitive development, neuropsychological variation, artistic performance, and symbolic systems. An intelligence, in his proposed framework, concerns a family of information-processing operations; it is broader than one learned technique and more specific than a person's entire worth or potential. Gardner's original argument is presented in Frames of Mind.
The central proposition is relative autonomy. Someone may excel in sophisticated verbal expression while finding spatial manipulation difficult; another may show exceptional musical discrimination and comparatively ordinary performance on school mathematics. MI theory interprets such uneven profiles as reasons to recognize different intelligences. Distinct patterns of ability unquestionably exist. The empirical question is whether they are best understood as eight comparatively independent intelligence systems, as correlated cognitive abilities, as expertise cultivated by experience, or as combinations of these.
Psychometric intelligence research approaches the problem through how scores on appropriately selected tasks covary across people. Cognitive neuroscience asks how distributed and specialized brain processes support performance. Educational research asks whether a proposed way of organizing instruction causes improved learning. These approaches test different consequences of MI theory. One form of evidence cannot substitute for all the others.
The Eight Types of Intelligence: Definitions and Examples
These examples explain Gardner's categories, rather than defining eight clinically validated scores. Real tasks frequently combine several abilities, motivation, knowledge, practice, physical opportunity, and contextual support.
1. Linguistic Intelligence
Linguistic intelligence denotes sensitivity to the structure, meaning, sound, and practical use of language. Gardner associated it with skilled writing, persuasive speaking, narration, interpretation, and the handling of verbal material. A writer who creates precise distinctions or a speaker who adapts a complex explanation to a particular audience may display strong verbal expertise.
A student's enthusiasm for reading is evidence of interest, not a calibrated intelligence score. Language performance is shaped by literacy access, language proficiency, vocabulary exposure, educational opportunities, and practice. Verbal comprehension is also measured in conventional cognitive-ability frameworks, so finding a verbal strength does not itself demonstrate an independent Gardnerian module.
2. Logical-Mathematical Intelligence
Logical-mathematical intelligence involves recognizing relationships, reasoning through conditional propositions, evaluating patterns, formalizing arguments, and working with quantities. Examples include following a mathematical proof, constructing a defensible inference, identifying a rule in data, or debugging an abstract system.
Successful reasoning typically draws on knowledge, working memory, attention, and learning history. Psychometric models already investigate fluid reasoning and quantitative knowledge; the overlap makes a specific validation of independence necessary. An aptitude for mathematics also differs from grades in mathematics, which reflect instruction, assessment conditions, and sustained effort.
3. Spatial Intelligence
Spatial intelligence concerns representing and transforming spatial relationships. A person might mentally rotate an object, navigate an unfamiliar place, interpret a map, or anticipate how a design would appear from another perspective. Architects, visual artists, engineers, and skilled navigators may use these operations in distinctive combinations.
Spatial cognition has a substantial empirical research tradition, including performance-based tasks and broad cognitive-ability factors. This supports studying spatial ability directly. It does not establish that every useful spatial activity is powered by a separate and generally independent intelligence in Gardner's precise sense.
4. Musical Intelligence
Musical intelligence refers to sensitivity to pitch, rhythm, timbre, musical organization, and patterns over time. A listener who detects subtle changes in harmony or a performer who creates and revises intricate rhythmic phrases provides examples of musical competence.
Musical performance also requires training, auditory discrimination, motor control, memory, cultural familiarity, and emotional engagement. Someone may have highly developed musical skill without extraordinary performance in unrelated domains. Such dissociations are valuable research questions; they do not, by themselves, prove an entirely autonomous intelligence architecture.
5. Bodily-Kinesthetic Intelligence
Bodily-kinesthetic intelligence involves skillful use and control of the body in making, performing, expressing, or solving practical problems. Precision in dance, athletic coordination, surgical technique, and intricate craftwork can illustrate the domain. Timing, proprioception, perceptual feedback, motor planning, and practice contribute.
A highly trained dancer and a talented mechanic may share refined sensorimotor control while relying on different learned repertoires. Gardner's decision to call these capacities intelligence reflects his broad definition. Many psychometric researchers would study the underlying motor and procedural abilities without classifying all of them as general intelligence factors.
6. Interpersonal Intelligence
Interpersonal intelligence describes the capacity to interpret other people's intentions, motivations, emotions, social cues, and likely responses. Examples include recognizing a conflict's causes, communicating appropriately in a sensitive interaction, and coordinating a team.
This proposed intelligence must be distinguished from sociability, agreeableness, empathy as a self-reported trait, and social popularity. Social understanding varies with cultural context and can require language, experience, emotion recognition, judgment, and perspective taking. A friendly personality is not a validated measure of interpersonal intelligence. For the distinction between traits and cognitive abilities, see the Hub's Big Five Personality Traits and Intelligence.
7. Intrapersonal Intelligence
Intrapersonal intelligence refers to differentiated understanding of one's own motives, emotional states, goals, strengths, and difficulties, and to using that understanding to guide activity. Reflecting accurately on what triggers frustration, knowing when a task exceeds current preparation, or recognizing the limits of one's confidence may illustrate the domain.
Self-knowledge also overlaps with metacognition, emotional awareness, and personality-related tendencies. Self-report is especially vulnerable to discrepancies between confidence and accuracy. Claiming to be highly self-aware is therefore a weak substitute for observing how effectively a person monitors and adapts their behavior over time.
8. Naturalistic Intelligence
Naturalistic intelligence denotes detecting, differentiating, organizing, and classifying patterns in the natural world: species, environments, living systems, and meaningful differences among natural objects. It was added after the initial set of seven. Botanists, field naturalists, experienced agricultural workers, and people with practiced skill in observing ecological patterns provide familiar examples.
Domain knowledge is crucial. Someone who reliably identifies birds by sound may have developed an exceptional learned classification system. This achievement merits recognition, although explaining its cognitive components and establishing a distinct latent intelligence would require appropriate tests and evidence beyond the individual's accomplishments. The Harvard Project Zero overview includes naturalistic intelligence among the eight.
Why Seven Became Eight—and What About a Ninth Intelligence?
The 1983 formulation began with seven proposed intelligences. Gardner subsequently added naturalistic intelligence as an eighth, and considered existential intelligence—sensitivity to fundamental questions concerning life, death, and human existence—as a possible candidate. Popular lists often present a ninth intelligence as if it had the same established standing as the eight. In Gardner's familiar formulation, eight remain the standard list; existential intelligence is better described as an additional proposal under consideration.
Other published lists may add emotional, spiritual, creative, or moral intelligence. Those additions should be traced to their actual authors and theories. A new label does not automatically become a Gardnerian intelligence, and it does not acquire psychometric validation simply by appearing in a classroom poster. The changing history of the framework is another reason to distinguish a theorist's proposal from an empirically fixed taxonomy.
How Did Gardner Decide What Counts as an Intelligence?
Gardner did not derive the original list by administering a large test battery and discovering exactly eight independent factors. He considered several lines of evidence, including the possibility of selective impairment after brain injury; exceptional profiles in prodigies and savants; identifiable core operations; distinctive developmental histories; evolutionary plausibility; experimental psychology; psychometric findings; and systems for representing ideas symbolically.
These are criteria for theoretical appraisal, not eight independently scored diagnostic tests. The criteria are heterogeneous. One may favor recognizing a domain's cultural importance, another may concern dissociation after injury, and another may suggest covariance with other tasks. A capacity can satisfy one criterion and fail another. Selecting the number and boundaries of proposed intelligences therefore involves interpretation, and a persuasive framework needs subsequent independently replicable predictions. The foundational theoretical position and methodological debate are represented by Gardner and Moran's response to critics.
Selective impairment after brain injury offers an especially intuitive argument: loss of language with relative preservation of music or spatial behavior can show that cognitive functions have partially differentiable neural requirements. However, functional specialization is compatible with interconnected networks and with common determinants of performance across individuals. A neuropsychological dissociation does not logically establish that test scores across the wider population will be statistically independent.
The Scientific Evidence: What Has Actually Been Tested?
Psychometric independence and the positive manifold
When people complete varied well-constructed cognitive tests, performance tends to correlate positively across domains, a pattern called the positive manifold. Hierarchical intelligence models describe a general factor alongside broad and narrower abilities. The existence of general covariance is an established empirical observation; the ultimate mechanisms behind that covariance are still investigated.
A directly relevant study by Visser, Ashton, and Vernon (2006) tested measures selected to reflect Gardner's eight domains in 200 adults. Factor analysis revealed a substantial general factor for the more conventionally cognitive domains, including linguistic, logical-mathematical, spatial, naturalistic, and interpersonal measures. Several domains also showed modest specific associations; tests of bodily-kinesthetic ability showed weaker general-factor loadings. The findings complicated the hypothesis that all eight intelligences operate as relatively independent capacities.
These results also have limitations. The choice of tasks affects factor solutions, no brief test battery exhausts the richness of musical, artistic, social, and motor performance, and one 200-person sample cannot settle an entire theoretical dispute. Gardner challenged the operationalization of his constructs, while the investigators argued that independently testable predictions remain essential. Their published exchange illustrates a genuine construct-validity dispute rather than a simple contest between appreciation of talent and appreciation of IQ.
Why measuring a proposed intelligence is difficult
To establish eight independent psychometric constructs, researchers would need clearly defined performance tasks for every domain; reliable scores across occasions; evidence of convergent validity within each domain; discriminant validity between domains; representative normative samples; and convincing evidence that the proposed structure predicts outcomes beyond established measures and relevant confounders.
Some MI questionnaires instead ask individuals to rate statements about interests, confidence, and preferences. Such scales can describe self-perceptions, but their scores may primarily capture personality, exposure, or activities already enjoyed. An assessment that labels someone a 'musical learner' because they enjoy music has not demonstrated an objective musical intelligence, let alone a teaching method tailored to it.
A theory may describe observable differences without being a validated measurement system. Gardner's own resource center says he has not developed a general MI test, and emphasizes demonstrated performance and observations in authentic settings over self-assessment.
Brain evidence and the modularity question
Different tasks engage partly specialized neural systems. Musical perception, movement, language, and social cognition have distinctive neural demands, yet activity across complex tasks is typically distributed, interactive, and dependent on coordinated brain systems. Demonstrating that some neural circuits are especially relevant to music does not establish a single musical-intelligence module, much less eight mutually independent modules.
In a 2023 conceptual analysis, Lynn Waterhouse argued that evidence for separate brain-based Gardnerian intelligences remains inadequate and characterized common neurological justifications of MI theory as a neuromyth. That paper is an argued critical synthesis, not a new experiment establishing an exact brain factor structure. Some authors have suggested neuroscientific compatibility with MI, but compatibility of broad neuroscience findings with a model does not independently validate its specific eight-part taxonomy.
The appropriate conclusion is bounded: differentiated cognition and neural specialization are real; the specific claim of eight relatively autonomous, distinctively brain-based intelligences remains insufficiently demonstrated.
What the evidence supports—and what it leaves open
Evidence strongly supports individual variation in verbal, spatial, quantitative, motor, musical, and social performance, and supports the importance of opportunities, experiences, and expertise. There is also substantial support for positively correlated cognitive abilities and hierarchical psychometric structure. Evidence is weaker for MI's distinctive prediction that the eight proposed types form independent latent systems.
This distinction preserves genuine achievements in every domain. A scientist can appreciate outstanding artistic or interpersonal competence without treating all competence as IQ, and without treating every valued skill as proof of a separate form of psychometric intelligence.
Multiple Intelligences vs. IQ, g, and CHC Theory
An IQ score is an observed result of a standardized assessment interpreted against a defined normative reference group. The general intelligence factor g is a latent statistical construct inferred from patterns of performance across tasks. Broad cognitive abilities are differentiated dimensions within hierarchical models, often organized by the Cattell–Horn–Carroll (CHC) framework. None of these terms refers to a person's entire identity, cultural contribution, or prospects for success.
Gardner's intelligences are theoretically proposed domains based on a broader account of solving culturally valued problems. Some overlap with CHC factors, such as verbal and spatial abilities; others, such as bodily-kinesthetic and naturalistic competence, cross traditional boundaries between cognition, expertise, action, and learned knowledge. MI therefore differs from CHC in both domain choice and the method used to justify the structure.
The scientific comparison turns on predictions and measurement. Hierarchical models are informed by large bodies of correlational and factor-analytic work. MI asks important conceptual questions about which human capacities deserve recognition, but its eight-part structure has not achieved comparable psychometric confirmation. A broader discussion of hierarchical abilities appears in McGrew's integrative review. The relation between intelligence and human identity in the Artificial Era is separately explored in the Hub's Intelligence in the Artificial Era.
Multiple Intelligences and Learning Styles Are Different Claims
A learning style is usually framed as a preferred mode of receiving instruction, such as visual, auditory, or kinesthetic learning. MI theory concerns proposed intellectual capacities and how people process or perform within domains. Gardner himself explicitly objected to collapsing the two ideas in his 2013 statement, Multiple Intelligences Are Not Learning Styles.
The practical error appears when a teacher administers a short profile questionnaire and concludes that a child should receive mathematics through music or movement because the child supposedly has that 'type' of intelligence. For the teaching-match claim to be established, students assigned to matched methods would need to learn more than comparable students using different methods, under controlled designs and fair outcome measures.
A major review by Pashler and colleagues found inadequate evidence to justify routine instructional matching based on learning-style classifications. This review tests learning-style claims, not every proposition within Gardner's theory. Its relevance is direct when MI is used as a pretext for rigid learner categories or one-to-one instructional matching.
Does Teaching Through Multiple Intelligences Improve Learning?
One of MI theory's major influences has been educational: encouraging recognition of diverse competence, offering varied ways to approach complex subject matter, and considering how different students may demonstrate understanding. These practices can be worthwhile educational design choices. Whether interventions specifically inspired by MI outperform equally strong alternatives is a causal question requiring comparative evidence.
Ferrero, Vadillo, and León (2021) systematically reviewed and meta-analyzed 39 studies, covering 3,009 students in 14 countries, that tested MI-inspired instruction and academic outcomes. Many reported positive effects, but the authors identified small samples, incomplete reporting, limited active controls, and signs of reporting or publication bias. Their conclusion was that the methodological quality did not permit a valid evaluation of effectiveness, and the reported advantages could not justify recommending MI interventions as a proven way to raise academic achievement.
A positive classroom experience may result from better teacher attention, richer materials, additional practice, novelty, collaboration, or a well-designed explanation. Distinguishing those mechanisms matters: if multiple representations help all learners, the intervention may be valuable without supporting the hypothesis that each learner possesses a specific matching intelligence.
Two educational ideas worth testing on their own terms
Gardner has emphasized personalization and pluralization: learning about the individual student, and presenting important concepts through several intelligible forms. In a recent account of his educational position, he explains these as his two major implications for teaching.
In practice, teachers can teach a scientific concept through clear language, diagrams, concrete demonstrations, and carefully chosen activities when those representations serve the subject. They can give students access to multiple forms of participation and assess the targeted learning outcome directly. Teachers can observe a student's actual performance over time, offer challenge and support, and adjust instruction when evidence shows a method is working.
The rationale for these practices should be their measured educational benefits, accessibility, and fit to the content, rather than an unverified categorical MI score. Diverse presentation has to preserve conceptual precision: acting out a calculation, for example, only improves mathematical learning when the movement actually makes the mathematical structure clearer.
Is There an Official Gardner Multiple Intelligences Test?
There is no single universal, scientifically established Gardner MI test that yields eight normed scores with the same interpretive status as professionally standardized cognitive assessments. Gardner explicitly states in his official resources that he did not create such a test. Project Spectrum, associated with Gardner and collaborators, instead used rich activities and observation to explore children's strengths in context.
Online 'eight intelligences' or 'which intelligence are you?' quizzes may be entertaining prompts for reflection. Usually they cannot establish the reliability, construct validity, cross-cultural comparability, or predictive utility required for consequential decisions. Items about what a respondent likes to do may assess interest. Items about what a respondent believes they can do may assess confidence. Neither automatically measures demonstrated ability.
For personal curiosity, examine your actual learning history and try substantive tasks in several domains. For educational accommodations, professional selection, disability determinations, or clinical questions, use appropriately validated tools and qualified assessment rather than MI personality-style quizzes. An MI category is not a medical or psychological diagnosis.
Common Misconceptions About the Eight Intelligences
Everyone has one fixed dominant intelligence
People can exhibit uneven profiles, and those profiles can change as knowledge, training, health, and opportunity change. A dominant MI label inferred from a brief questionnaire has no established diagnostic meaning. Even within a domain, a person may be accomplished at performance but less proficient at composition or theoretical explanation.
Each intelligence has its own independent brain center
Complex cognitive behaviors require cooperating systems; partial neural specialization does not establish a one-to-one map of eight intelligences to eight isolated anatomical centers. This popular neurological picture goes beyond the supporting evidence. The 2023 critical review discusses the gap.
IQ tests measure everything valuable about a person
IQ tests measure specified cognitive performance under standardized conditions and have meaningful predictive and measurement properties. They do not measure all culturally valuable capabilities. Recognizing their limits does not make the claims of a competing model automatically true.
Multiple intelligences theory has been scientifically proven because it is widely taught
Popularity, intuitive appeal, and successful implementation are different from independent validation of a model's structure or causal effectiveness. The 2021 evidence review specifically cautions against treating the classroom literature as strong proof.
Musical or bodily-kinesthetic intelligence means a child must learn every subject through that domain
A strong skill or interest can supply a useful point of entry and motivation. The best explanation of a topic should still represent the relevant concept accurately. A preference-based matching rule needs experimental evidence that has not been established as a general policy.
How to Use the Theory Responsibly
For educators, the most defensible application is to broaden what accomplishments are noticed while measuring educational progress against clearly stated learning objectives. Use varied representations because their content is useful, because they make participation accessible, or because well-designed evidence shows improved understanding. Avoid sorting children into fixed intelligence tracks. Monitor changes in learning rather than assuming teaching quality from students' enthusiasm alone.
For parents, recognize accomplishments across language, movement, music, mathematics, and social interaction, and support exploration without defining a child's future through one category. A child struggling with a topic may benefit from better explanations, sufficient practice, assessment of barriers, or individual support. An MI quiz should not override these questions.
For adult learners, the theory can prompt reflection on neglected competencies and cultural definitions of achievement. Test an unfamiliar skill with actual practice; distinguish immediate interest from current proficiency; and permit improvement through experience. Occupational success usually depends on combinations of domain knowledge, cognition, collaboration, motivation, and opportunity.
For researchers, the work remains a prompt to examine broad human competencies more carefully, but substantive progress depends on operational definitions, discriminant measurement, well-powered tests of the predicted structure, and interventions designed to distinguish MI-specific mechanisms from ordinary features of good instruction.
Frequently Asked Questions
What are Gardner's eight types of intelligence?
Linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal, and naturalistic. These are Gardner's proposed domains, not eight independently validated diagnoses or universally accepted intelligence-test scales.
Did Gardner propose seven, eight, or nine intelligences?
He first proposed seven in 1983 and later added naturalistic intelligence as the eighth. Existential intelligence was discussed as a possible additional candidate; it has not attained the same standing as the familiar eight-category version.
Is multiple intelligences theory scientifically proven?
Its distinctive claim of eight relatively autonomous intelligences lacks convincing psychometric and neuroscientific confirmation. The existence of diverse skills and cognitive strengths is well supported, but that broader observation does not verify the theory's precise structure.
What is the difference between multiple intelligences and general intelligence?
MI proposes domains of intellectual competence based on a wide theoretical set of criteria. General intelligence (g) is a latent statistical factor identified through covariation among performance measures. Hierarchical models include both a general component and differentiated abilities.
Is multiple intelligences theory the same as learning styles?
No. Gardner distinguishes his proposed capacities from instructional-mode preferences. The claim that students must receive instruction matched to an alleged style is a further educational hypothesis requiring its own evidence.
Can I find my intelligence type with a free test?
A short online MI quiz can describe preferences or self-reported activities, but it generally cannot establish a validated intelligence profile. Performance-based evidence, appropriate norms, and explicit interpretation are necessary for consequential testing.
Can a person have more than one intelligence?
Gardner proposes that everyone possesses capacities across the domains, with different developmental profiles. The degree to which these are empirically distinct remains debated. People routinely combine verbal, social, spatial, motor, and learned competencies within one real-world task.
Should teachers use multiple intelligences activities?
Teachers may use varied, well-designed activities if they clarify subject matter and improve access or engagement. Evidence does not currently support MI-labeled instruction as a uniquely validated method for improving achievement, nor does it justify fixed grouping by MI profile.
Does Gardner's model explain autism, ADHD, dyslexia, or intellectual disability?
No. MI categories are not clinical diagnoses or substitutes for established developmental, educational, or neuropsychological assessment. Individuals with these conditions can show varied strengths and needs, evaluated through appropriate evidence and context.
Conclusion
Gardner's multiple intelligences theory made an influential contribution by challenging narrow cultural assumptions about whose achievements count as intellectual. It offers a language for discussing linguistic, mathematical, spatial, artistic, bodily, interpersonal, self-reflective, and naturalistic competence. Its signature claim—that these capacities constitute eight relatively autonomous intelligences—remains insufficiently validated as a psychometric or neural architecture.
The research supports a precise practical stance: recognize the diversity of human competence, examine actual performance and learning opportunities, use instruction aligned with the content and learner's demonstrated needs, and evaluate educational methods through sound comparisons. Appreciation of multiple forms of achievement and scientific scrutiny of an intelligence theory belong in the same account.
Related Articles
References
Ferrero, M., Vadillo, M. A., & León, S. P. (2021). A valid evaluation of the theory of multiple intelligences is not yet possible: Problems of methodological quality for intervention studies. Intelligence, 88, 101566. https://doi.org/10.1016/j.intell.2021.101566
Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. Basic Books. Harvard Project Zero publication record
Gardner, H. (2013). Multiple intelligences are not learning styles. Harvard Graduate School of Education. https://www.gse.harvard.edu/ideas/news/13/10/multiple-intelligences-are-not-learning-styles
Gardner, H. (n.d.). Resource center: Assessing multiple intelligences. Official author resource. https://www.howardgardner.com/resource-center
Gardner, H., & Moran, S. (2006). The science of multiple intelligences theory: A response to Lynn Waterhouse. Educational Psychologist, 41(4), 227–232. https://doi.org/10.1207/s15326985ep4104_2
McGrew, K. S. (2009). CHC theory and the human cognitive abilities project: Standing on the shoulders of the giants of psychometric intelligence research. Intelligence, 37(1), 1–10. https://doi.org/10.1016/j.intell.2008.08.004
Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008). Learning styles: Concepts and evidence. Psychological Science in the Public Interest, 9(3), 105–119. https://doi.org/10.1111/j.1539-6053.2009.01038.x
Project Zero, Harvard Graduate School of Education. (n.d.). Multiple intelligences. https://pz.harvard.edu/projects/multiple-intelligences
Visser, B. A., Ashton, M. C., & Vernon, P. A. (2006). Beyond g: Putting multiple intelligences theory to the test. Intelligence, 34(5), 487–502. https://doi.org/10.1016/j.intell.2006.02.004
Visser, B. A., Ashton, M. C., & Vernon, P. A. (2006). g and the measurement of multiple intelligences: A response to Gardner. Intelligence, 34(5), 507–510. https://doi.org/10.1016/j.intell.2006.04.006
Waterhouse, L. (2006). Multiple intelligences, the Mozart effect, and emotional intelligence: A critical review. Educational Psychologist, 41(4), 207–225. https://doi.org/10.1207/s15326985ep4104_1
Waterhouse, L. (2023). Why multiple intelligences theory is a neuromyth. Frontiers in Psychology, 14, 1217288. https://doi.org/10.3389/fpsyg.2023.1217288