Beware the myth: learning styles affect parents', children's, and teachers' thinking about children's academic potential

The paper addresses the widespread but unsupported VAK learning styles myth—the belief that individuals have dominant modality-based learning styles (visual, auditory, kinesthetic/tactile) that predict learning outcomes. Despite its popularity across many countries and among educators and the public, empirical evidence does not support that VAK styles predict cognitive functioning or learning, nor do they align with validated individual differences in thinking. The authors propose an overlooked consequence of this myth: that describing children with learning style information may bias parents’, teachers’, and children’s judgments about students’ intelligence and prospects across school subjects. They situate this question within research on essentialist reasoning, whereby people treat categories (like “visual learner”) as biologically based and predictive of behaviors and outcomes, potentially fostering specific inferences (e.g., who is smart or good at math). Given that beliefs about intelligence and related stereotypes emerge early in childhood and have real-world consequences for achievement, the authors test whether labeling a child as a visual versus hands-on learner changes judgments of intelligence, sportiness/effort, and predicted grades across school subjects. They hypothesize that visual learners will be perceived as smarter and better at core subjects (math, language arts, social studies; with science as a possible exception), whereas hands-on learners will be viewed as sportier/harder-working and better at non-core subjects (art, gym, music).

The introduction reviews evidence that neuromyths, especially learning styles, are widely endorsed by educators and the public worldwide (e.g., United States, Turkey, Portugal, China, Switzerland, UK, Latin America). Foundational critiques show no credible empirical basis for VAK learning styles influencing learning outcomes, and they are unrelated to established constructs of individual differences. Prior domain-specific claims (e.g., that visual learners excel in applied science; that medical students have distinct learning styles) are criticized as unsupported yet influential in practice, including university teaching centers promoting associations between learning styles and specific subjects. The authors draw on psychological essentialism research indicating people often see learning styles as stable, biologically based categories that predict outcomes, suggesting these beliefs may enable specific inductive inferences about academic strengths. Existing work has not tested specific predictions about performance across subjects; this paper fills that gap by examining concrete inferences (intelligence, core vs non-core subject aptitude) made from learning style descriptions.

Design: Three experiments tested whether describing a student as a visual learner (learns best using eyes) or a hands-on learner (learns best using hands/touch) affects judgments of intelligence and academic aptitude. Descriptions used plain-language descriptors rather than labels to ensure familiarity across ages and ecological validity.

Experiment 1 (Children and Parents):

• Participants: Children: N=73, ages 6–12 years (M=9.06, SD=1.88), 48.1% female, predominantly monoracial white (67.5%), primarily mid-to-high SES (71.4% family income >$60k; 88.3% mothers college-educated); recruited via lab database and social media (mostly Greensboro, USA; 2 from Canada). Parents: N=94 US parents recruited via Prolific (51.1% female; age 20–79, M=42.25, SD=13.21). IRB approvals noted (UNCG IRB #20-0365 for children; UNCG IRB #20-0213 for adults). Exclusions: children (n=3) for incomplete/experimenter error; parents (n=4) for failed attention check.
• Materials/Procedure: Two vignettes describing Grade 5 students: one visual learner and one hands-on learner (silhouette images to avoid gender/identity cues). For each student, participants rated smartness and sportiness using a 4-point scale (1 not very smart/sporty; 4 really smart/sporty). Children saw narrated slides via Zoom/WebEx; parents completed a Qualtrics survey with randomized order of learner and question blocks. An attention check asked parents to identify which question had not been asked in the story.
• Analysis: Linear mixed-effects models (lme4 in R) with random intercepts for participant (1|id). For children: fixed effects included Age (continuous, centered, in months), Learning Style (visual vs hands-on), Question Type (smart vs sporty), and their interactions. For parents: model without Age. Cohen’s d effect sizes; Wald 95% CIs; Bonferroni-corrected post hoc simple effects (emmeans). Sensitivity analyses via simr for key interactions.

Experiment 2 (Parents and Teachers; Forced-choice):

• Participants: Parents: N=79; Teachers: N=94 (67% female; 75.5% white; age 20–78, M=34.94, SD=10.76), all US-based via Prolific and paid $1 for ~5 min. Screening: parents reported having children; teachers affirmed a teaching role. Exclusions: 6 teachers and 21 parents for failing attention checks or non-meaningful responses. A belief-in-learning-styles item was included; all parents and 85.1% of teachers endorsed the myth.
• Materials/Procedure: Same learning style introductions as Experiment 1. Two randomized forced-choice questions: (a) Which student is smarter? (b) Which student is sportier? Attention checks included brief descriptions of each learner after introductions. Open-ended exploratory questions asked participants to list three school subjects each learner would excel at.
• Analysis: Mixed-effects binary logistic regression with random intercepts for participant, predictors: Question (smarter vs sportier) and Group (parent vs teacher). Odds ratios reported. Word-frequency visualizations for open-ended responses.

Experiment 3 (Pre-registered; Parents and Teachers):

• Participants: Parents N=100 (73% female, 81% white; age 23–79, M=45.65, SD=12.23); Teachers N=100 (58% female, 66% white; age 19–75, M=38.76, SD=12.42). No exclusions; all passed attention checks. Teaching settings: 57% school (public/private), 30% university/college, 13% home/other. Main age groups taught: 35% Adults, 30% High school, 21% Elementary, 11% Middle school, 3% Preschool.
• Materials/Procedure: After introductions to Grade 5 learners, participants estimated each learner’s report card grades across seven subjects: math, science, language arts, social studies (core), and art, gym, music (non-core). Grades selected on a 10-point letter-grade scale (Below C- = 1, C- = 2, C = 3, C+ = 4, B- = 5, B = 6, B+ = 7, A- = 8, A = 9, A+ = 10). Forced-choice items followed: (a) Which student is smarter? (b) Which student works harder? Final open-ended item solicited reflections for attention screening. Pre-registered analyses.
• Analysis: Linear mixed-effects models as in Exp 1. Hypothesis 1: factors were Subject Type (core vs non-core), Learning Style, Sample (teacher vs parent), and interactions. Hypothesis 2A: Subject (math, social studies, language arts, science) × Learning Style. Hypothesis 2B: Subject (art, gym, music) × Learning Style. No Bonferroni correction per preregistration. Post hoc simple effects; sensitivity analyses confirmed power. Forced-choice analyzed with mixed-effects logistic regression (Question: smarter vs works harder; Group: teacher vs parent).

Experiment 1:

• Children: Significant main effect of Learning Style (b = 0.22, 95% CI [0.09, 0.34], p < 0.001, d = 0.47) and a significant Learning Style × Question Type interaction (b = −0.20, 95% CI [−0.32, −0.08], p = 0.001, d = −0.44). Simple effects: Children rated visual learners as smarter than hands-on learners (Mvisual = 2.88, SD = 0.99; Mhands-on = 2.26, SD = 1.09), t = 3.48, p_adjusted = 0.001, d = 0.58; no difference for sportiness (Mvisual = 2.04, SD = 1.17; Mhands-on = 2.23, SD = 1.21), t = 1.08, p = 0.540, d = 0.18. No age effects observed.
• Parents: Significant main effects of Question Type (b = 0.16, 95% CI [0.10, 0.23], p < 0.001, d = 0.56) and Learning Style (b = 0.10, 95% CI [0.03, 0.17], p = 0.007, d = 0.32), and a significant Learning Style × Question Type interaction (b = −0.33, 95% CI [−0.40, −0.26], p < 0.001, d = −1.11). Simple effects: Parents rated visual learners as smarter (Mvisual = 3.02, SD = 0.49; Mhands-on = 2.55, SD = 0.74), t = 4.66, p_adjusted < 0.001, d = 0.68; and hands-on learners as sportier (Mhands-on = 2.88, SD = 0.79; Mvisual = 2.03, SD = 0.99), t = 8.47, p_adjusted < 0.001, d = 1.24.

Experiment 2:

• Forced-choice mixed-effects logistic regression: Main effect of Question Type (smarter vs sportier) was significant (b = 1.61, 95% CI [1.29, 1.92], z = 9.95, p < 0.001; OR = 4.99). No main effect of Group (parent vs teacher; b = 0.25, 95% CI [−0.07, 0.56], p = 0.126; OR = 1.28). Significant Question × Group interaction (b = −0.43, 95% CI [−0.74, −0.11], p = 0.008; OR = 0.65), with teachers showing a larger effect. Percentages: For “Which student is smarter?”, parents chose visual 76.6% vs hands-on 23.4%; teachers chose visual 69.62% vs hands-on 30.38%. For “Which student is sportier?”, parents chose hands-on 94.68% vs visual 5.32%; teachers chose hands-on 82.28% vs visual 17.72%.
• Open-ended subjects: Visual learner strengths—Teachers: math 22.65%, history 14.89%, English 9.39%, art 8.74%, reading 5.83%; Parents: math 19.56%, English 9.96%, art 7.75%, history 7.75%, reading 5.54%. Hands-on learner strengths—Teachers: science 19.14%, art 16.83%, gym 11.55%, math 6.60%, music 5.28%; Parents: science 13.52%, art 11.03%, gym 6.76%, chemistry 5.34%, physics 4.63%.

Experiment 3 (pre-registered):

• Hypothesis 1 (Subject Type × Learning Style × Sample): Main effects of Learning Style (b = 0.07, 95% CI [0.004, 0.128], p = 0.036, d = 0.08) and Subject Type (b = −0.69, 95% CI [−0.75, −0.63], p < 0.001, d = −0.86); no main effect of Sample (b = −0.05, 95% CI [−0.17, 0.08], p = 0.456, d = −0.11). Significant Learning Style × Subject Type interaction (b = −0.60, 95% CI [−0.66, −0.54], p < 0.001, d = −0.75). Simple effects: Visual learners rated higher on core subjects, t = 12.89, p < 0.001, d = 0.65; hands-on learners rated higher on non-core subjects, t = 13.94, p < 0.001, d = 0.81.
• Hypothesis 2A (Core subjects): Significant main effects of Learning Style (b = −0.53, 95% CI [−0.60, −0.46], p < 0.001, d = −0.79) and Subject (b = 0.97, 95% CI [0.85, 1.09], p < 0.001, d = 0.83); significant Learning Style × Subject interaction (b = 0.62, 95% CI [0.50, 0.74], p < 0.001, d = 0.53). Simple effects (means on 1–10 scale): Language arts—visual 7.15 (SD 1.86) > hands-on 5.67 (SD 1.69), t = 10.23, p < 0.001, d = 1.02; Math—visual 7.36 (SD 1.79) > hands-on 5.82 (SD 1.91), t = 10.65, p < 0.001, d = 1.07; Social studies—visual 6.96 (SD 1.96) > hands-on 5.55 (SD 1.72), t = 9.72, p < 0.001, d = 0.97; Science—no difference (visual 7.62, SD 1.55; hands-on 7.80, SD 1.72), t = 1.21, p = 0.226, d = 0.12.
• Hypothesis 2B (Non-core): Main effects of Learning Style (b = 0.66, 95% CI [0.59, 0.74], p < 0.001, d = 1.06) and Subject (b = 0.73, 95% CI [0.62, 0.84], p < 0.001, d = 0.83); significant Learning Style × Subject interaction (b = −0.43, 95% CI [−0.54, −0.32], p < 0.001, d = −0.48). Simple effects: Hands-on > visual in Art (visual 8.61, SD 1.55; hands-on 9.09, SD 1.10), t = 3.47, p < 0.001, d = 0.35; Gym (visual 7.07, SD 2.04; hands-on 9.01, SD 1.23), t = 14.16, p < 0.001, d = 1.42; Music (visual 6.69, SD 1.87; hands-on 8.25, SD 1.56), t = 11.42, p < 0.001, d = 1.14.
• Hypothesis 3 (Forced-choice replication): Participants more likely to pick the visual learner as smarter and the hands-on learner as working harder. Logistic model: main effect of Question (b = −0.83, 95% CI [−1.04, −0.612], z = −7.59, p < 0.001; OR = 0.44); no main effect of Group (b = −0.071, 95% CI [−0.28, 0.14], p = 0.515; OR = 0.93) and no Question × Group interaction (b ≈ 0, p = 0.990). Percentages: Smarter—visual chosen 67–70% by teachers and 70–77% by parents; Works harder—hands-on chosen ~69–72% teachers, ~67–72% parents.

Across three studies, providing learning style descriptors systematically biased judgments: visual learners were perceived as more intelligent and better at core academic subjects, while hands-on learners were perceived as sportier/harder working and better at non-core subjects; science was an exception among core subjects where hands-on and visual learners were judged similarly. Children aged 6–12 also showed these inferences for intelligence versus sportiness, indicating early emergence of neuromyth-influenced reasoning. These findings suggest that learning style categories function like social categories that trigger essentialist inferences, potentially shaping educators’ and parents’ expectations about children’s abilities and academic prospects. Given the lack of scientific support for VAK learning styles, describing children in these terms may foster unwarranted stereotypes about intelligence and subject-specific aptitude. The results underscore the need for caution in using learning style language and for efforts to improve the scientific foundations of educational discourse and practice, as endorsements of the myth appear widespread and may influence both adult judgments and children’s self-concepts and choices.

The paper demonstrates that learning style descriptions (visual vs hands-on) lead parents, teachers, and children to draw specific, unwarranted inferences about a child’s intelligence and academic strengths: visual learners are seen as smarter and better at core subjects (math, language arts, social studies), while hands-on learners are seen as sportier/harder-working and better at non-core subjects (art, gym, music); science judgments were comparable across styles. These findings reveal a consequential pathway by which a pervasive neuromyth can shape educational expectations. Future research should examine real-world consequences for recommendations, admissions to specialized programs, and children’s academic identity and choices; explore intersections with other identities (gender, race); test how labels versus descriptors modulate effects; and assess interventions that reduce reliance on learning style categorizations in educational settings.

• Potential euphemism confound: “hands-on” might be used as a euphemism for lower academic competence; against this, hands-on learners received relatively high ratings in science and moderate ratings elsewhere.
• Alternative constructs: Participants might have been referencing real individual differences (e.g., visual/verbal cognition) rather than the VAK myth; however, the pattern (e.g., visual > hands-on for both math and language arts; hands-on > visual for art) does not map cleanly onto such constructs.
• Methodological inflation: Forced-choice and survey formats may amplify effects by requiring selections; nevertheless, learning styles are commonly linked to outcomes in real-world messaging, and participants spontaneously associated styles with subjects in open-ended responses.
• Sample overlap: Some teachers may also be parents, potentially attenuating group differences; notably, being a teacher did not protect against myth-based inferences.
• Labels vs descriptors: Only descriptors were used; labels might strengthen effects. The moderating role of labeling and description framing remains to be tested.
• Generalizability: US-based, predominantly white, middle-class samples may limit generalization across cultures and demographics.