Pathways, not labels. Backed by research.
Velociti turns your material into multiple evidence-informed pathways — then adapts to what actually helps each learner make progress.
Beyond visual, auditory, and kinesthetic
Coffield and colleagues catalogued at least 71 learning-style models in 2004 — and most struggled when tested. The strongest reviews are skeptical of the “meshing” idea: that matching instruction to a learner's stated style reliably improves outcomes.
The stronger direction in learning science focuses on learner variability, task demands, cognitive load, self-regulation, prior knowledge, and observed performance — not fixed type labels.
What we actually believe.
- 01
Learner preferences are real — but preference alone is not proof that a format improves learning.
- 02
The meshing hypothesis is not well supported by the strongest review evidence.
- 03
Many learning-style instruments have reliability, validity, and pedagogical-impact problems.
- 04
Models can still help as design checklists — not as hard diagnostic categories.
- 05
Stronger directions: cognitive load theory, multimedia learning, learner variability, self-regulated learning, prior knowledge, performance-based adaptation.
Six dimensions, one lesson
Representational mode
Text-first, diagram-first, audio, animation, worked examples, etc.
Processing path
Global overview first, sequential steps, example-first, abstract-first.
Support level
Low load, standard, advanced, practice-heavy, scaffold-heavy.
Learning approach
Deep meaning, exam strategy, application, memorization, exploration, reflection.
Regulation support
Goals, checklists, progress, mistake review, confidence checks, spaced review.
Evidence loop
Adapt from observed performance, not labels alone.
The same material, multiple ways in
| Output style | What it generates |
|---|---|
| Visual map | Diagrams, concept maps, flowcharts, timelines |
| Audio script | Narrated explanation, podcast-style summary |
| Reading / writing | Notes, glossary, summaries, flashcards |
| Sequential guide | Numbered steps prerequisite → advanced |
| Global overview | Big-picture model before details |
| Example-first | Concrete example before formal rule |
| Quiz-first | Diagnostic questions before explanation |
| Retrieval practice | Flashcards, short-answer, spaced review |
| Low cognitive load | Chunked version with signaling |
| Application version | Use cases, projects, exercises |
Models we draw from
| Model | Evidence note | How Velociti.ed uses it |
|---|---|---|
| VAK / VARK | Weak evidence for fixed matching | Output formats, not fixed learner types |
| Felder-Silverman | Useful framework; matching evidence cautious | Active/reflective, sensing/intuitive, etc. |
| Vermunt ILS | Strong candidate for learner modeling | Meaning/reproduction/application-directed |
| Multimedia learning (Mayer) | Strong experimental tradition | Core design for dense source material |
| Universal Design for Learning | Inclusive framework | Multiple means of engagement/representation |
| Cognitive load theory | Strong instructional foundation | Chunking dense PDFs |
| Self-regulated learning | Strong learner-difference direction | Adaptive tutoring, dashboards |
| Retrieval practice | Very evidence-backed | Quizzes, spaced repetition |
Where to dig deeper
- Coffield et al. — Learning styles and pedagogy in post-16 learningRead source ↗
- Pashler et al. — Learning styles: concepts and evidenceRead source ↗
- Felder — Index of Learning StylesRead source ↗
- Mayer — Cognitive theory of multimedia learningRead source ↗
- CAST — Universal Design for Learning guidelinesRead source ↗
- Boyle et al. — Vermunt ILS researchRead source ↗
- Romanelli et al. — Learning styles: a review of theory and practiceRead source ↗
- Evans & Waring — Styles, approaches and learning environmentsRead source ↗
Discover pathways — not permanent styles
The most defensible promise: convert the same source into multiple evidence-informed pathways, then adapt using observed performance.
See it in your own material
Turn a PDF or notes into pathways built on learning science — free to start.