Personal Learning Preferences (Preferencias Personales de Aprendizaje)

What this is. What has traditionally been called “learning styles” is called Personal Preferences (Preferencias Personales) in Smoother. The renaming is deliberate and reflects our position on the evidence. They are the particular ways in which each person prefers to interact with the Learning Objects (Objetos de Aprendizaje — what is learned, in eight components), the Learning Tasks, the Activities and the Resources — and they are used as a menu of variety, never as a prescription.

Why we do not call them “learning styles”

The meshing hypothesis — that teaching each person according to their preferred style produces better learning — has no solid evidence to support it. The reference is Pashler, McDaniel, Rohrer and Bjork (2008), Learning styles: Concepts and evidence, in Psychological Science in the Public Interest.

Smoother fully adopts that critical position, and four consequences follow from it:

• Attending to preferences makes the experience more smoother — more attractive, comfortable, appealing and memorable — but does not necessarily produce more effective learning. They are different things and we do not confuse them.
• Even so, we firmly believe in attending to each person’s unique preferences. Engagement, comfort and appeal are valuable in themselves: they sustain motivation, reduce friction and make people want to keep learning.
• Most people do not have a single dominant preference, but affinities toward several, in varying degrees and contextually: they change with the topic, the moment, the mood and the situation.
• Preferences are a menu of variety, not a prescription. A good experience offers multiple representations, resources and tasks that serve everyone, not a subgroup according to their “type.”

What does determine effective learning are cross-cutting factors that apply to everybody: retrieval practice, spaced practice, interleaving, cognitive load management, dual coding, worked examples, productive failure, quality feedback and deliberate design for transfer.

This position is not a criticism of Fleming, Felder, Kolb or Gardner, whose contributions Smoother explicitly acknowledges. It is a calibration of what can be asserted when using their frameworks: drawing inspiration from them to build a catalog of coverage and variety, yes; promising that measuring a person’s “type” and matching the teaching to it produces better learning, no.

The catalog: 16 preferences in 6 categories, plus 4 Kolb profiles

Each preference is described with what appeals to the person who has it, the activities that take advantage of it and the resources that support it. Those lists are inventories of variety: a well-designed Learning Unit (Unidad de Aprendizaje) combines elements from several preferences; it does not pick one. The four Kolb profiles are not independent preferences but combinations of the perception and processing dimensions, and they are included because they help in understanding how several preferences cluster at once.

The role they play, unchanged

The change was of name and framing, not of function. The Personal Preferences still serve two things:

1. Coverage inventory. Are we offering visual elements? Auditory ones? Practical ones? Reflective ones? Social ones? Logical ones? That coverage guarantees richness, accessibility and variety.
2. Personalization as an invitation, not as a restriction. Recognizing a person’s preferences makes it possible to suggest first the formats that most appeal to them and to offer alternatives — never to limit what is available. All formats and tasks remain available to everyone.

The diagnostic, and why its instruments are our own

When the Personal Preferences diagnostic is offered to the person who is learning, it is composed exclusively of original Smoother instruments, inspired by public theoretical frameworks but without reproducing or adapting proprietary instruments: four administered instruments plus one derived profile, and around 58 items in total. The Kolb profile is not administered: it is computed as a derived view of the Felder dimensions already measured.

The reasons, stated plainly: no legal risk — VARK and the Kolb inventory are protected commercial instruments —; honesty about the evidence — we do not claim psychometric validation we do not have, and each instrument of our own is labeled Adapted or Emerging —; complete editorial control; and free iteration. The results page visually distinguishes the assessed, the derived and the self-assessed dimensions.

Three conventions govern the result: it never prescribes — it suggests entry points, and the full catalog remains available —; the catalog is separable from the diagnostic — a preference can be explored and declared without having taken the diagnostic —; and re-assessment is free, because preferences change with the topic, the mood and prior experience: nothing in the result is a permanent label.

How they are applied

Every Learning Unit offers a variety of resources, materials and tasks that cover multiple preferences. A person’s preferences are recognized in order to personalize — suggest first, offer alternatives — not to limit. And they are always complemented by the principles of Memory and Learning, which are the ones that do determine effectiveness.

Evidence

Level: Adapted, with reformulated and renamed use. The preferences come from several traditions: VARK, Gardner’s Multiple Intelligences and the Felder-Silverman dimensions; the four Kolb profiles combine preferences. The meshing hypothesis has no evidence (Pashler et al., 2008). That is why Smoother renamed them and uses them not as a prescription but as a menu of variety and a coverage inventory: attending to them makes the experience more smoother, not necessarily more effective.

That discipline — honoring without asserting effectiveness, with the label attached and visible at the point of decision — is the direct lineage of the rule of support without endorsement that Smoother applies to external frameworks.

Prohibited wording on every surface: “your learning style is X, therefore…” and any formulation that hints at meshing · “proven method” · “evidence-based” applied to personalization by preference.

Related

Memory and Learning · Learning Activities · Material Elements · Visualizations · Learning Units · Response Styles · Discovery

Source

protocol/smoother-preferencias-personales.md — version 0.39.0 (2026-08-14). Machine-readable data in data/preferencias-personales.json.