A learning theory is an account of how people learn. You will never be asked to pick one and live in it, but you will be asked, in a QA review and in an interview, to say which one a piece of instruction leans on and whether that fits the people it is for. This lesson gives you the four theories that cover most of what you will see, and then the distinction that runs underneath this whole program: adults do not climb the ladder children climb.
By the end you should be able to summarize behaviorism, cognitivism, constructivism and connectivism, what each says learning is and what instruction looks like under it (3.1); distinguish andragogy from pedagogy and explain "structure to chaos, chaos to structure" (3.2); given a described lesson, name the theory it leans on and say whether that fits the audience (3.3); and place a described program with the frame 24/7 Teach uses for its audience, Learn-Do-Be for younger learners or the learning model for adults, and say why (3.4).
In this lesson 5 sections
Read alongside this lesson: Understanding Andragogy and Pedagogy: Structure to Chaos and Chaos to Structure(opens in a new tab) (10 minutes). It is the article the learning model itself is grounded in.
Hold one request in mind while you read. A director of branch operations at a regional bank asks you for a 20-minute eLearning module with a 10-question quiz to teach new branch managers how to coach. The video opens with her request, and the end of 3.3 comes back to it.
3.1 Four theories, and what each makes you build
Learning theories are principles and frameworks that explain how people learn and retain knowledge. In instructional design they are the foundation under every model, whether or not the model names them. Four cover most of the ground. For each, notice two things: what it says learning is, and what instruction built on it looks like. The second is how you will recognize it in someone else's work.
Behaviorism
Learning is a change in observable behavior, shaped by practice, reinforcement and feedback. The learner's inner state is not the object of study; what they do is. Instruction looks like clear performance objectives, practice with immediate feedback, drill on component skills, mastery before progression, and rewards or consequences tied to performance. Its fingerprints are on every model: Gagné's "elicit performance, provide feedback" and Mager's measurable objectives are behaviorist inheritances. Where it fits: procedures, safety and compliance, anything where the correct behavior is known and must become automatic. Where it strains: judgment, transfer to new situations, and anything where the right answer depends on context.
Cognitivism
Learning is the building and reorganizing of mental structures: schemas, memory, attention. The learner is an information processor with limits. Instruction looks like chunked content, worked examples, advance organizers that show the shape of what is coming, attention to how much is asked of working memory at once, and sequences from simple to complex. Lesson 4 is almost entirely cognitivist in its evidence base: cognitive load, spacing, retrieval and Mayer's multimedia principles all sit here. Where it fits: conceptual and technical content, anything complex enough that how it is presented changes whether it is learned. Where it strains: it can treat the learner as a vessel and forget that meaning is made, not received.
Constructivism
Learning is something the learner builds from experience, by connecting new material to what they already know and making meaning of it, often with others. Instruction looks like authentic problems and projects, cases and simulations, discussion, reflection, and a facilitator who guides rather than tells. Merrill's problem-centered principle, project-based learning and the whole "Do" beat of this program are constructivist in stance. Where it fits: judgment, complex skills, adult professional learning. Where it strains: pure novices, who can flounder without enough structure to make the struggle productive (Lesson 4 returns to this under the expertise reversal effect). The 24/7 Teach answer is not to abandon structure but to arrive at it last: build first, then name what you built.
Connectivism
Learning is, in George Siemens's account, the forming and navigating of networks: of people, of sources, of tools. Knowing where to find and how to evaluate is as important as what you hold in your head, because the field moves faster than any individual can. Instruction looks like communities of practice, curated networks of resources, social learning platforms, and explicit teaching of how to find, judge and connect. Where it fits: fast-moving domains and distributed workforces, and the way working instructional designers actually keep current. Where it strains: it describes the environment well and the individual's learning less precisely; it is younger and less tested than the other three.
| Theory | Learning is | You will see | Best fit |
|---|---|---|---|
| Behaviorism | Observable behavior change through practice and feedback | Objectives, drill, immediate feedback, mastery gates | Procedures, compliance, safety |
| Cognitivism | Building mental structures within the limits of memory and attention | Chunking, worked examples, organizers, load management | Concepts, technical content |
| Constructivism | Meaning built from experience, often socially | Problems, projects, cases, reflection, facilitation | Judgment, complex adult learning |
| Connectivism | Forming and navigating networks of people and sources | Communities, curation, social learning | Fast-moving fields, distributed teams |
No process model belongs to one theory. ADDIE can produce a constructivist course and SAM a behaviorist one. The theory shows up in what is built, not in the model that built it.
Check yourself: A module teaches forklift pre-start checks with a video demonstration, a ten-item practice checklist scored immediately, and a rule that you cannot proceed until you score 10 of 10. Which theory, and does it fit?
Behaviorism, and yes. The behavior is known, must become automatic, and the cost of getting it wrong is physical. Drill, immediate feedback and a mastery gate are exactly right here. The same design applied to "how to give difficult feedback to a direct report" would strain, because the right behavior there depends on context.
3.2 Andragogy and pedagogy: structure to chaos, chaos to structure
Pedagogy is the teaching of children; andragogy, a term Malcolm Knowles brought into wide use, is the facilitation of adult learning. Knowles rested his framework on several assumptions about adults: that their self-concept moves from dependent toward self-directed, that they bring a reservoir of experience, that their readiness to learn is tied to their social roles, that they are problem-centered rather than subject-centered, and that they are motivated more from inside than from outside. The framework is debated, including whether the line between adults and children is as clean as he drew it. The self-direction assumption is the best-supported part and the part that matters most here. Where andragogy assumes a learner who is already self-directed, heutagogy, a term Stewart Hase and Chris Kenyon introduced, goes one step further and makes the capacity to direct your own learning the goal of the design itself. Keep the name in mind for the AI lessons: a learner who can run the taxonomy without a designer is the outcome the chaos-to-structure move is aiming at.
The linked article puts the difference in a form this program is built on. Pedagogy starts at the bottom of Bloom's Taxonomy and climbs: Remember, Understand, Apply, and eventually Create. Children need that scaffold; they are building foundational knowledge with limited capacity to process it, so structure comes first and creativity later. Structure to chaos.
Andragogy runs the same taxonomy in the opposite direction. You start at Create, in a state of deliberate chaos, using the experience you already have to attempt something real. Then you work down: Evaluate, Analyze, Apply, Understand, and finally Remember. Chaos to structure. That is not skipping the fundamentals. It is arriving at them last, having built them yourself, which is why they hold. And "chaos" does not mean disorganization. It means an open environment with a real problem and a real deadline, because urgency is what turns exploration into invention rather than wandering.
Two consequences for you as a designer. First, when you design for adults at work, start from the problem they have, not from the content you have; the content arrives when the problem demands it. Second, when you design for children, or for adults who are complete novices in a domain (see the expertise reversal effect, Lesson 4), more structure up front is not condescension, it is what the evidence says they need. This course is itself an example of the balance: Units A and B are the structured floor that makes QA cycle one a productive struggle rather than a random one.
Check yourself: Why does this program put a two-week structured course before the first project, if adults learn chaos to structure?
Because chaos with no vocabulary is not productive chaos. Our learners are novices in instructional design even when they are experts in teaching; the evidence on novices (Lesson 4) says some guidance first makes the struggle productive. Units A and B are that guidance, kept to two weeks so the first real project comes fast. The theory then arrives when the project makes you need it, which is the andragogical move.
3.3 Reading a lesson for its theory
In QA you will read other people's designs and be asked why they do or do not work. The first move is to name the theory the design leans on and ask whether it fits the audience and the content. Three prompts do most of the work:
- What does the learner actually do? Recall and repeat (behaviorist). Process and organize (cognitivist). Build, solve, discuss (constructivist). Find, connect, curate (connectivist).
- Who is the audience and how expert are they? Novices in the domain need more structure than the design's theory may be giving them; experts are slowed by structure they no longer need.
- Does the theory match the outcome? A procedure taught through open-ended discovery wastes time; a judgment taught through drill produces people who know the rule and cannot apply it.
Practice on three short descriptions:
Try it: A. New managers read a case about a conflict on a team, propose what they would do, compare in small groups, and hear how an experienced manager handled the real event.
Constructivist: authentic case, meaning built through discussion, expert perspective offered after the learner's own attempt (this is also Merrill's activation and demonstration, which Lesson 6 covers). Fit is good for judgment content and an adult audience.
Try it: B. A pharmacology module presents drug classes in a chunked sequence with a worked example of a dose calculation, then three practice calculations of increasing difficulty with feedback.
Cognitivist, with behaviorist practice: chunking and worked examples manage load; practice with feedback builds the skill. Fit is good for technical content where the calculation must be right.
Try it: C. A leadership course consists of a curated set of articles, a discussion forum, and a directory of internal experts to contact, with no fixed sequence.
Connectivist. Fit depends on the audience: for experienced leaders who know what they need, strong; for new leaders who do not yet know what they do not know, it strains, because nothing tells them where to start. That is the diagnosis a QA reviewer would write.
Transcript
Welcome back. So, you've just tried diagnosing three example designs on your own. Now we're going to walk through a worked example together for Lesson 3, Section 3.3 of 24/7 Teach's Intro to Instructional Design course.
Today, I'm going to model exactly how a QA reviewer reads a real module. This particular one was built for school staff who give students their medication at school. Think of this explainer as me, an experienced reviewer, pulling back the curtain and showing you exactly how my brain works when I look at a course design.
Okay, prompt 1. What does the learner actually do? As reviewers, we always want to look strictly at the observable actions. If we look at this module's outline, we can inventory the exact steps the learner takes. Step one, they watch five videos. This sequence includes a two-part lecture on administration, a hand washing demonstration, a video on the rights of medication administration, and a final one on forms of medication. Notice how I'm not detailing the internal content of those videos. I'm simply observing the action itself. Step two, they complete a worksheet using a real pharmacy label. And finally, step three, they take a quiz. That's the entire sequence of actions from start to finish.
Pause here and answer the question below the video.
Alright, the answer is mostly behaviorism. We're seeing a very clear pattern here: a demonstration, a checklist of rights that absolutely has to become automatic, some practice, and then a quiz. Splitting the content into short, distinct videos does add some cognitivist chunking, which is great, but the foundation of this design leans heavily on behaviorism as a learning theory.
Moving right along to prompt two. Who is the audience? The audience consists of school staff. Most of them don't have clinical backgrounds, and they're completing required training. Because they are novices, having this kind of heavy directive structure right up front isn't condescending at all. It's exactly what they need to succeed.
And that brings us to prompt three: does the theory match the outcome? The rights of medication administration are a strict procedure. The correct sequence is completely known, it absolutely has to become automatic, and a mistake causes real harm. That makes it a perfect fit for behaviorism. And honestly, this brilliantly illustrates why that worksheet is the strongest activity in the module. It uses a real pharmacy label rather than a made-up one, to practice that exact high-stakes procedure.
Pause here and answer the question below the video.
All right. Welcome back. Here is our verdict. This design is a really good fit, but it has one gap. Most of the learner's time is spent just watching, and the procedure only gets practiced once, on one single label. Behaviorism as a learning theory tells us that a procedure becomes automatic through repeated practice with quick feedback.
Let's compare the current setup with our proposed change to see how this builds. In the current design, the procedure is practiced just once. Because of that, there's a crucial judgment moment the module never actually practices: what happens when one of the rights cannot be confirmed? That's exactly where drill alone strains.
So, here is the change we're making. We need to add several short label checks with instant feedback. This should include a few where something doesn't match, so the learner has to practice how to decide what to do next.
So, let's close by looking at the overall shape of our review. Three prompts, one verdict, and one change. That is exactly what a QA note looks like. Moving forward, your own QA notes will follow this exact same shape to clearly evaluate how any module uses learning theories.
As we wrap up this worked example, I want to leave you with this thought. You now have a practical, structured framework for rating a lesson. How will you apply this reviewer's lens to diagnose the very next module you encounter?
Back to the bank
At the start of this lesson, a director of branch operations asked for a 20-minute eLearning module with a 10-question quiz to teach new branch managers how to coach. Now you can name what felt off.
The request leans on behaviorism: present content, then check recall with a quiz. Behaviorism fits when the correct behavior is known and has to become automatic. Coaching is not that. What a good coach says depends on the person in front of them and the situation, which is judgment, and judgment is where drill strains.
A better fit is constructivist: short, realistic coaching situations where a new manager decides what they would say, practices the conversation, gets feedback, and then compares their approach with how an experienced manager handled it. That is the same shape as example A. Because new managers are novices at coaching, open with one worked example of a good conversation before they try their own. That's prompt 2 at work: novices need more structure first.
It can still roll out quickly: a few short scenarios online, then one practice conversation with their own manager. You are not refusing the director. You are keeping what she needs, speed and reach, and changing what the learners do.
Compare this with the sentence you wrote at the start of the lesson.
You wrote:
3.4 Two frames, two audiences
24/7 Teach runs two frames, and you will meet both. With younger learners, in the teen programs and in the schools we work with, the frame is Learn-Do-Be: know it, do it, become the kind of person who does it well. It is pedagogy in the sense of 3.2, structure to chaos: the knowledge comes first, the practice is guided, and identity grows out of that practice. With adults, in the Instructional Design Bootcamp and the professional programs, the frame is the learning model you read in Lesson 2: three axes, the eight-stage project cycle, four criteria scored from the work. It is andragogy, chaos to structure: you start at Create with a real brief and a real deadline, and the theory arrives when the project makes you need it.
The two are not rivals, and one is not the junior version of the other. They are the same taxonomy read in opposite directions, for the reason 3.2 gives: children and novices need the scaffold first, and adults with experience learn by building first and naming what they built afterwards. What changes between them is what is written down up front, and what is judged.
| Learn-Do-Be (younger learners) | The learning model (adults) | |
|---|---|---|
| Audience | Children and teens; adults who are complete novices in a domain | Working adults with experience to build from |
| Theory base | Pedagogy: structure to chaos | Andragogy: chaos to structure |
| Where it starts | Learn: the knowledge, taught first and scaffolded | Create: a real project, attempted first |
| Shape of growth | Three phases in order. Identity (Be) is the destination, and it is written as an objective before the unit begins | Rounds of the project cycle. Identity is never written as an objective; it is read from the work across rounds, and the Iteration criterion and the mindset it describes are where it shows |
| What is written up front | A Be objective, then Learn and Do objectives derived from it (Mager's three parts, the ABCD pattern, Bloom's domains) | Success criteria for the project, checkable at Demo; the five questions of minimum viable analysis |
| What is judged | Knowledge (quizzes), skill (performance tasks) and disposition (observation, reflection), each against its objective | Four criteria, Analysis, Alignment, Evaluation and Iteration, at four levels, from the work |
| Where to read it | Handbook chapters 4 and 5 | Lesson 2; the learning model itself |
Two things follow for you. First, when you read a 24/7 Teach youth curriculum in QA and see Be objectives written up front, that is not verb inflation or a category error; it is the frame doing its job for its audience, and you judge it on whether the Learn and Do objectives actually serve the Be. Second, when you design for adults and find yourself writing a "Be" objective, stop and ask whether you are about to teach adults the way you would teach children. The learning model's answer is that an adult's identity as a designer is not an objective anyone writes; it is what the rubric reads off round three that was not there in round one.
Read for contrast: Handbook chapter 4, Connecting Frameworks to Objectives and Assessments in the Learn, Do, Be Model (11 minutes) and chapter 5, Crafting "Be" Objectives (19 minutes). Both are written for the youth frame and open with a note saying so. Read them with the table above beside you, and notice which of the two frames every example belongs to.
Check yourself: A summer program for rising ninth graders writes, before the first session, "Students will become dependable teammates, as shown by the task log and the mentor's rubric." A bootcamp for working designers writes, before the first project, "Participants will become confident instructional designers." Same move?
No. The first is Learn-Do-Be doing its job: a Be objective for younger learners, written up front, with the evidence named. The second is the youth frame applied to adults. The learning model would not write that sentence at all; it would state the project's success criteria and let confidence, if it comes, show up in the work at round three. The tell is the audience, not the verb.
Before the checkpoint
Eight questions on the four objectives. If you can fill in the table in 3.1 from memory, explain chaos-to-structure in two sentences, and say which frame a described program belongs to and why, you are ready.
Sources named in this lesson: Malcolm Knowles on andragogy and self-directed learning; George Siemens on connectivism; the linked 24/7 Teach article; chapters 4 and 5 of the handbook. Publication years are omitted here; check them against the primary source before citing.