Methodology

The Compound Method

A Learning Science Framework for Financial Education
Published by Compound  ·  compoundpower.netlify.app  ·  May 2026

Most financial literacy content fails not because it is wrong, but because it is not built to be remembered.

A 10-minute explainer video, a Wikipedia article, a podcast episode: each transfers information. None builds durable knowledge. The distinction matters enormously in financial education, where the gap between knowing a concept and applying it under real market conditions costs people actual money.

The Compound Method is the instructional design framework that governs every structural decision in the Compound curriculum. It is not a set of stylistic choices. It is a system of constraints derived from peer-reviewed cognitive science, applied to the specific problem of teaching long-term investing to adults who have never been formally taught it.

This document names the framework, cites its foundations, and explains why each design decision was made the way it was. It exists because pedagogical provenance matters: when an approach works, that approach should be documented, attributed, and held to account.

The Problem

Why Financial Education Fails

The dominant model for financial literacy education is: explain the concept, show an example, maybe ask a question. This is the model used by Investopedia, Khan Academy's finance modules, most YouTube channels, and most fintech onboarding flows. It is the wrong model for three compounding reasons.

First, it front-loads abstraction. Presenting a definition before a concrete problem violates what cognitive scientists call anchored instruction: people understand abstractions most durably when they encounter the problem the abstraction solves before they are told what the abstraction is called.

Second, it ignores working memory limits. Working memory can hold roughly four discrete items simultaneously (Cowan, 2001). A typical financial explainer presents a concept, its formula, an example, a counterexample, and a graph simultaneously. The result is cognitive overload.

Third, it mistakes exposure for encoding. Reading about compound interest is not the same as retrieving an understanding of compound interest under testing conditions. Passive exposure produces recognition. Active retrieval produces recall. Financial decisions require recall.

1. Anchored Instruction 2. Cognitive Load 3. Generation Effect 4. Spaced Retrieval 5. Corrective Feedback 6. Emotion as Anchor
Principle 1

Anchored Instruction

CTGV, 1990  ·  Bransford et al., 2000

Every lesson in Compound begins with a concrete, narrative problem before any concept is named. Learners follow Amara, a first-time investor, through a series of decisions that make the problem vivid and personally meaningful before any abstraction is introduced.

In Lesson 12 (Discounted Cash Flow), Amara needs to decide whether the future cash flows of a logistics company are worth paying for today. She confronts the problem before the term "discount rate" appears anywhere on screen. This is not storytelling for engagement. It is the specific sequencing that the CTGV demonstrated produces deeper comprehension and better transfer to novel problems than definition-first instruction.

Transfer, the ability to apply a concept to a situation you have never seen before, is the only outcome that matters in financial education.

Design Constraint
No concept is named in Compound before the problem that concept solves has been concretely established. The name is always the answer to the problem, never the starting point.
Principle 2

Cognitive Load Management

Sweller, 1988  ·  Cowan, 2001

Working memory is not expandable. Every element placed on a screen simultaneously competes for the same limited processing capacity. When capacity is exceeded, learning stops.

Compound enforces a strict one-idea-per-screen rule. Each step in the lesson sequence isolates a single cognitive demand. The Hook establishes the problem. The Concepts step introduces exactly three ideas, one at a time, through a sequential reveal mechanic that prevents parallel processing. The Simulator asks the learner to engage with only one variable relationship at a time.

The consequence is that each step feels simple. This is not a failure of ambition. It is the design working as intended. Simplicity at each step enables complexity at the level of the accumulated schema.

Design Constraint
No lesson screen in Compound presents more than one core concept simultaneously. Every additional element on a screen is evaluated for whether it adds to or subtracts from the active cognitive demand.
Principle 3

The Generation Effect

Slamecka and Graf, 1978  ·  Bjork, 1994

Memory encodes more deeply when a learner is required to generate an answer than when they passively receive one. This is true even when the generation attempt is incorrect. The effort of retrieval triggers a deeper encoding process than reading equivalent content.

In Compound, the three concept cards never display the concept name on the front. The front presents the problem the concept solves, as a question: "How do you instantly find someone willing to buy your exact number of shares right now?" The learner must sit with that question, attempt an answer mentally, and then tap to reveal. The reveal, "The Order Book," lands on a mind that has already worked on the problem.

Bjork calls this "desirable difficulty": constraints that feel like friction but produce superior long-term retention.

Design Constraint
No concept in Compound is passively presented. Every concept is the answer to a question the learner was asked to consider before the answer appeared.
Principle 4

Spaced Retrieval

Ebbinghaus, 1885  ·  Roediger and Karpicke, 2006

Without active retrieval, 70% of new information is lost within 24 hours. This is the Ebbinghaus forgetting curve, one of the most replicated findings in all of cognitive psychology. Retrieval resets and strengthens the retention curve. Each retrieval event pushes the decay curve flatter and further out.

Compound implements this through a structural constraint in the quiz design: the second quiz question in every lesson deliberately reaches back to a concept from a prior lesson. In Lesson 7, the second quiz question references the cash flow distinction from Lesson 6. In Lesson 12, it references the margin of safety concept from Lesson 11. This forces a retrieval event for prior material at the moment it is most useful, when the learner is actively building a connected schema.

Module checkpoint quizzes serve as a second-order spaced retrieval mechanism, requiring integrated recall across the full module.

Design Constraint
The second quiz question in every lesson is drawn from prior lesson material, not from the current lesson. No exceptions.
Principle 5

High-Value Corrective Feedback

Hattie and Timperley, 2007  ·  Kulhavy and Stock, 1989

Feedback that says "incorrect" creates no new neural pathways. Hattie and Timperley's meta-analysis of 500+ studies found that feedback quality, specifically whether it identifies the structural flaw in the learner's reasoning, is the primary determinant of whether feedback improves performance.

In Compound, every incorrect answer option has its own bespoke feedback. The feedback is not generic. It validates the reasoning behind the mistake, acknowledging why the wrong answer was tempting, and then isolates the specific structural error before redirecting to the correct mental model. A learner who selected "Company X is larger because it has a higher share price" receives feedback that addresses exactly why share price alone is insufficient, not simply a restatement of what market capitalization means.

This approach requires writing four to five unique feedback responses per quiz question rather than one. That is the cost of effective corrective feedback.

Design Constraint
Every incorrect option in every quiz question has individually written feedback that addresses the specific reasoning error that option represents, not the correct answer restated.
Principle 6

Emotion as Memory Anchor

Cahill and McGaugh, 1998  ·  McGaugh, 2000

The amygdala tags emotionally significant events for preferential encoding in long-term memory. Information paired with mild surprise, tension, or awe is recalled two to three times more reliably than equivalent information presented in neutral affect.

Compound uses Prudie (Pecunia Prudence), a character delivering one precise, verified data point per lesson, as an engineered emotional anchor. These are not trivia. They are selected for their capacity to produce a specific affective response: the mild surprise of encountering a fact that recontextualizes the concept just taught.

The character form matters because it provides relational continuity across lessons. Prudie is not a mascot. She is a recurring emotional anchor whose presence signals "this is the moment of highest encoding opportunity in this lesson."

Design Constraint
Every lesson contains exactly one Prudie data point, placed at the structural peak of the lesson. The fact must produce mild surprise in a naive learner and must directly reinforce the lesson's central concept without introducing new conceptual content.
Curriculum Architecture

The Sequencing Logic

The six principles govern individual lesson design. The sequence of lessons governs the architecture of the schema learners build. The 25-lesson curriculum is organized into five layers, each of which is a prerequisite for the next.

1
Foundation Layer (Lessons 1–5)
What money is, how companies are structured, how markets price assets, and how investors measure value at the most basic level. No learner can reason about intrinsic value without first understanding what equity represents.
2
Reading Layer (Lessons 6–10)
How to extract meaning from financial statements and annual reports. A valuation method applied to a company whose financials the investor cannot read is not analysis. It is guessing with extra steps.
3
Valuation Layer (Lessons 11–15)
Intrinsic value, DCF modeling, ratio analysis, margin of safety, and competitive moats. These concepts require both the foundational mental models and the ability to read financial data before they can be meaningfully applied.
4
Portfolio Layer (Lessons 16–20)
Diversification, asset allocation, index funds vs. active management, rebalancing, and risk-adjusted return measurement. A single-stock valuation skill without portfolio discipline produces volatile, non-compounding outcomes.
5
Optimization Layer (Lessons 21–25)
Tax-advantaged accounts, tax-loss harvesting, dividend strategy, compounding mechanics, and behavioral finance. These concepts produce value only after the preceding four layers are in place.
Clarifications

What The Compound Method Is Not

It is not gamification. XP points and streaks are motivation maintenance tools. They address the engagement problem, not the learning problem. They are present in Compound because motivation maintenance is real and necessary. They are not the pedagogy.

It is not personalization. The Compound Method is a fixed sequence built on the conviction that the sequencing itself is the pedagogical value, not that different learners need different sequences, but that there is one correct sequence for building this particular schema from scratch.

It is not just "evidence-based." That phrase has been diluted to mean almost nothing. Every design decision in Compound traces to a specific mechanism in a specific peer-reviewed study. The citation is not decoration. It is the accountability structure that keeps the design honest.

How to cite this document
Compound. (2026). The Compound Method: A Learning Science Framework for Financial Education. compoundpower.netlify.app/method
References
Cognition and Technology Group at Vanderbilt (1990). Anchored instruction and its relationship to situated cognition. Educational Researcher, 19(6), 2–10.
Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257–285.
Slamecka, N. J., & Graf, P. (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 592–604.
Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. Shimamura (Eds.), Metacognition: Knowing about knowing. MIT Press.
Roediger, H. L., & Karpicke, J. D. (2006). The power of testing memory: Basic research and implications for educational practice. Perspectives on Psychological Science, 1(3), 181–210.
Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112.
Cahill, L., & McGaugh, J. L. (1998). A novel demonstration of enhanced memory associated with emotional arousal. Consciousness and Cognition, 4(4), 410–421.
Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage capacity. Behavioral and Brain Sciences, 24(1), 87–114.