Imagine a student opening an asynchronous course at 11:47 p.m. They watch a video, click through a chapter, complete a quiz, and submit an assignment. The module is marked complete.

The next afternoon, the student remembers that the video was long, one quiz question was confusing, and the assignment was due before midnight. What they cannot clearly remember is the central idea the chapter was supposed to teach.

This is one of the quiet problems of online education. Students can complete every course activity without building a durable understanding of the subject. The course records their movement through the material, but movement is not the same as learning.

This is not necessarily a failure of effort, motivation, or intelligence. It is often a failure of instructional design.

Human attention is limited, and memory is selective. Students do not preserve a complete recording of everything they encounter. They retain fragments, patterns, emotional moments, useful connections, and ideas that they have been asked to retrieve or apply. In an asynchronous class, where the instructor is not physically present to emphasize an important point, notice confusion, or bring an earlier idea back into the conversation, these memory-building moments must be designed into the course.

Memory researcher Charan Ranganath describes four approaches for making communication more memorable: chunking information, making it concrete, providing callbacks, and creating curiosity. These four principles can be especially useful in asynchronous education, where students need more than uploaded information. They need carefully designed encounters with the ideas that matter most.

Artificial intelligence can support this process, but only when its role is clearly defined. AI should not become a vending machine for assignment answers. It should help students revisit, question, translate, rehearse, and connect course material.

1. Chunk the Chapter Around One Important Question

A chapter may introduce a dozen terms, several models, multiple examples, and a collection of supporting facts. To an expert, these pieces form a recognizable system. To a student encountering the material for the first time, they may look like a vocabulary parade wearing identical shoes.

Working memory can hold only a limited amount of unfamiliar information at once. When everything appears equally important, students often remember very little.

Each chapter should therefore be organized around one central question.

A chapter on customer discovery might ask:

How can an entrepreneur distinguish a real customer problem from an interesting assumption?

A chapter on pricing might ask:

How does a business choose a price when customers, competitors, and costs point in different directions?

The question becomes a container for the chapter. Definitions, cases, activities, and assessments can then be organized beneath three or four key ideas that help answer it.

Before entering the details, students should see a brief roadmap:

• This chapter addresses one central question.
• You should remember three major ideas.
• You will use those ideas to make one decision.

AI can help an instructor produce summaries, outlines, glossaries, and alternative explanations, but the instructor must still decide what deserves to be remembered. That is a teaching decision, not a software setting.

The goal is not to make the chapter smaller. The goal is to give its parts a recognizable shape.

2. Make Abstract Ideas Concrete

Students rarely struggle because a definition contains too few words. They struggle because the words are floating above the ground.

Terms such as value proposition, competitive advantage, organizational culture, opportunity cost, and customer segmentation can sound understandable while remaining difficult to use. A student may repeat a definition correctly but still be unable to recognize the concept in a business situation.

Every chapter therefore needs a concrete decision, situation, person, or organization.

Instead of asking students only to define a value proposition, present a neighborhood restaurant with declining lunch sales. Ask what the restaurant is actually offering office workers besides food.

Instead of explaining fixed and variable costs only through formulas, show a student entrepreneur deciding whether to rent equipment, purchase it, or outsource production.

Instead of discussing target markets as categories on a slide, ask students to decide which customer a small company should stop trying to serve.

A concrete case gives the abstract idea somewhere to live.

The case does not have to be elaborate. A photograph, short paragraph, customer quotation, local business example, or 90-second video may be enough. What matters is that students must use the chapter concept to interpret something recognizable.

Source-grounded AI tools can expand this practice. Google recently renamed NotebookLM as Gemini Notebook, although many educators and students will continue to recognize the earlier name. The tool can work with instructor-selected PDFs, websites, videos, audio, Google Docs, and slides. It can then produce source-grounded explanations with inline citations, along with study guides, briefings, audio overviews, mind maps, and other learning formats.

In an asynchronous course, a student might use the chapter notebook to ask:

• Explain this idea using the business case.
• What evidence in the chapter supports that explanation?
• Give me a second example from a different industry.
• What is a common misunderstanding of this concept?
• Compare this concept with the one from the previous chapter.

The AI is not completing the assignment. It is helping the student walk around the concept and see it from several angles.

3. Provide Callbacks Across the Course

Many asynchronous courses move forward relentlessly. Chapter 1 disappears when Chapter 2 opens. Chapter 2 is buried beneath Chapter 3. By the final week, the course resembles an archaeological site with deadlines.

Learning becomes stronger when students are asked to retrieve earlier ideas and use them again.

A callback can be as simple as one question:

How would the customer evidence collected in Chapter 3 affect the pricing decision in Chapter 6?

It might appear at the beginning of a video, inside a Brightspace self-assessment, in a discussion prompt, or at the end of an assignment.

The purpose is not to catch students forgetting. The purpose is to make remembering part of learning.

A strong callback does three things. It asks students to retrieve an earlier concept, connect it to the current material, and use it in a new situation. That small intervention turns separate chapters into a developing system of knowledge.

AI can support callbacks by comparing selected chapter sources, but students should attempt the connection before consulting the tool. A useful sequence is:

1. Answer from memory.
2. Consult the chapter sources or AI companion.
3. Revise the answer.
4. State what changed.

This makes AI part of a feedback loop rather than an answer-replacement loop.

4. Create Curiosity Before Delivering the Answer

Many courses begin a chapter by presenting objectives, definitions, and instructions. All of that may be necessary, but it does not necessarily create a reason to care.

Curiosity begins when students notice a gap between what they think they know and what they can confidently explain.

Before students read a chapter, give them a question that appears simple but contains a complication:

• Why might a company with strong sales still run out of cash?
• Can listening to customers ever lead a business in the wrong direction?
• Why might lowering a price reduce demand?
• When could rapid growth make a business weaker?
• Can a profitable idea still be a poor opportunity?

Students can submit a prediction through a one-question Brightspace quiz, poll, or discussion. The response does not need to be graded for correctness. Its purpose is to create an intellectual loose thread that the chapter will later pull.

At the end of the chapter, students return to the same question and revise their position using evidence.

This before-and-after structure gives the chapter a small narrative. Students begin with uncertainty, encounter ideas and evidence, and end with a more informed judgment.

AI can deepen the curiosity by presenting competing explanations or identifying what evidence would be needed to resolve the question. But the instructor should create the original knowledge gap. Curiosity is more powerful when it is connected to the learning outcome rather than generated as decorative trivia.

AI as the Asynchronous Learning Companion

The most productive role for AI in an asynchronous course lies between the course material and the graded assignment.

Students often need help during that middle space. They may need a concept explained differently, a reading summarized, a term connected to an example, or a question they can use to test their understanding. In a face-to-face classroom, an instructor might provide this support through conversation. In an asynchronous course, the student may encounter the problem hours or days before receiving a response.

A source-grounded notebook can offer limited, immediate assistance based on materials selected by the instructor. It can generate audio explanations for students who learn while commuting, study guides for students who need structure, flashcards for retrieval practice, and alternative explanations for students who did not understand the first presentation.

However, AI-generated materials can contain errors. Students should be instructed to compare important claims with the assigned sources and use the citations provided by the tool.

The boundary should be unmistakable:

AI may help students understand the course material. It may not replace the student’s observation, judgment, evidence, or final response.

A useful AI interaction should lead back to the student. After using the notebook, the student might submit:

• One idea that became clearer.
• One passage or source that supported the explanation.
• One remaining question.
• One way the concept applies to a business or community.
• One point on which the student disagrees with the AI response.

This preserves human thinking as the center of the activity.

The Chapter Memory Loop

An effective asynchronous chapter does not require a small carnival of applications, videos, quizzes, bots, badges, and blinking buttons. It needs a repeatable learning rhythm.

Each chapter can follow a five-part memory loop:

Question: Begin with a problem or prediction that creates curiosity.

Map: Identify the three ideas students should remember.

Example: Place those ideas inside a concrete situation.

Callback: Connect the current chapter to something previously learned.

Application: Ask students to use the concept in a decision, observation, explanation, or recommendation.

AI can sit beside this loop as an optional or directed learning companion. It can help explain, rehearse, compare, and review. Brightspace remains the course home, the location of record, and the place where students submit their own work.

Designing for Memory, Not Merely Completion

Making a course memorable does not mean adding more material. In many cases, it means asking students to encounter fewer ideas more deliberately.

A memorable course repeatedly signals what matters. It provides a concrete place for abstract ideas to land. It brings earlier learning back into view. It creates questions before supplying answers. It gives students opportunities to retrieve, connect, and apply.

AI can complement this design by making course sources more conversational and accessible. It can provide another doorway into the material, particularly when the instructor and student are separated by time. But it should remain a doorway, not the destination.

The lasting value of a course is not measured by how much content students clicked. It is revealed by what they can recognize, retrieve, question, and use after the module has closed.

The simplest Brightspace model

For a five-week, ten-chapter course, I would create five weekly AI notebooks, not ten separate notebooks. Each notebook would contain the two chapters assigned that week, your chapter guide, any expert-session material, and one concrete business example. This keeps the system from becoming a filing cabinet with a chatbot trapped inside.

This builds naturally on your existing online and hybrid SBE100 development and your prior work designing AI-enhanced entrepreneurship modules.

Use the same structure for every chapter:

Memory layer	Student experience	Brightspace object
Curiosity	Answer one prediction question	Ungraded quiz or discussion
Chunk	Review three essential ideas	One HTML content page
Concrete	Examine one short business case	Page, image, or brief video
AI companion	Ask one approved source-based question	External link or uploaded AI artifact
Callback	Retrieve an earlier concept	Three-question self-check
Application	Make and justify one decision	Discussion or short assignment

Brightspace pages can contain uploaded files, multimedia, external links, embed code, and Quicklinks to quizzes, assignments, and discussions. Its Question Library also lets you store and reuse retrieval questions across quizzes and self-assessments.

Do not make the external notebook the course doorway

There is an important account wrinkle. Consumer Google accounts can make a notebook accessible to anyone with the link who has a Google account. However, public sharing is currently disabled for Workspace Education and Enterprise accounts.

For CUNY, the most reliable approach is:

1. Build the notebook as the instructor.
2. Generate the chapter briefing, audio overview, FAQ, or quiz.
3. Download or copy the useful artifact.
4. Place that artifact directly inside Brightspace.
5. Offer the interactive notebook link only after testing student access.

This prevents a Google-account problem from becoming an academic-access problem. Keep all required instructions, graded activities, and essential course materials inside Brightspace.

One master prompt for producing each intervention

Place the two chapter sources into the weekly notebook and use this prompt:

Using only the selected sources for Chapter [number], create a compact asynchronous learning intervention. Organize it around one central question. Identify three essential ideas in plain language, one concrete small-business example, one common misconception, three retrieval questions, and one callback connecting this chapter to Chapter [previous number]. Do not complete or imitate the graded assignment. Cite the source supporting each major explanation.

You can then copy the result into the following Brightspace page.

Chapter [Number] Memory Lab: [Chapter Title]

The Question

Before beginning the chapter, answer this question from your current understanding:

[Insert an intriguing chapter question.]

Record your initial answer in the Chapter [Number] Pre-Learning Check. You are not graded on whether your first answer is correct.

Three Ideas to Remember

By the end of this chapter, you should be able to explain and use these three ideas:

1. [Essential idea one]: [One-sentence explanation.]
2. [Essential idea two]: [One-sentence explanation.]
3. [Essential idea three]: [One-sentence explanation.]

Do not try to memorize every sentence in the chapter. Use these three ideas to organize the details.

See It in Practice

Read the following situation:

[Insert a 100- to 150-word business, community, or entrepreneurial case.]

As you review the chapter, consider:

Which of the three essential ideas best explains what is happening in this situation?

Use the AI Chapter Companion

The AI Chapter Companion works only with the sources selected for this module. Use it to clarify and test your understanding, not to complete your graded work.

Ask one of the following:

• Explain [chapter concept] using the chapter case.
• What evidence in the chapter supports this explanation?
• Give me another example from a different kind of business.
• What is a common misunderstanding of this idea?
• How does this chapter connect with Chapter [previous number]?

After using the companion, write down one idea that became clearer and identify the source that supported it.

Memory Callback

Without reopening the previous chapter, answer:

[Insert one question connecting the current chapter with an earlier idea.]

Then check the earlier material and revise your answer if necessary.

Three-Minute Self-Check

1. What is the chapter’s central question?
2. What are the three most important ideas?
3. How would you apply one idea to the business case?

Complete the ungraded Chapter [Number] Self-Check in Brightspace.

Apply the Learning

Submit a response of approximately 100 to 150 words:

1. Make one decision about the business case.
2. Use one chapter concept to justify the decision.
3. Identify the evidence that supports your reasoning.
4. State one uncertainty or limitation.

Your response must reflect your own judgment. AI may help you understand the material, but it may not make the decision or write the submission for you.