Week 1: When information is cheap
Use a familiar Wi-Fi problem to distinguish a correct answer from evidence of understanding, with explanations and activities you can work through independently.
22 February 2027
You found an explanation. How would you tell whether you understand it?
This is a self-guided lesson. Both these notes and the slide deck contain the explanations you need; no presenter is assumed. Allow about 30 minutes, with another 20 minutes if you complete the extended investigation. Use paper or a text document for your answers. No networking background, working router, or AI account is required. The people and incidents below are fictional.
What you will learn#
By the end, you should be able to separate an observation from an inference, explain why a relevant answer can have more than one origin, and choose a follow-up question that gives you more useful evidence. You should also be able to explain why using a reference does not automatically disqualify a performance.
1. Connected to Wi-Fi, but websites will not load#
Your laptop is connected to your home Wi-Fi. Yet a website will not load. You search for an explanation and find: “A Wi-Fi connection does not guarantee internet access.” The sentence sounds useful. But what would it mean to understand it, rather than recognise or repeat it?
Start with the technical distinction. Wi-Fi provides a wireless connection to a local network. At home, the Wi-Fi access point is often part of the router. Internet access also requires a working route beyond that network. One part of the journey can work while another fails. A website failing to load alone does not identify the failed part.
For example, the home Wi-Fi equipment could remain powered on while an internet-provider outage interrupts the onward connection. This explains how local connectivity without internet access is possible. It does not diagnose every situation in which a website fails to load.
Technical background: Microsoft’s Wi-Fi guidance and Cloudflare’s introduction to routers. Our cases simplify the technology so we can focus on judging understanding.
2. A correct sentence is a starting point#
Imagine Sam reads that explanation and says, “Wi-Fi and internet access are different.” You are the examiner: the person deciding what this performance shows. You might be a teacher, a colleague checking an explanation, or someone examining your own learning.
An observation describes what happened: “Sam repeated the distinction while the explanation was visible.” An inference is a conclusion drawn from it: “Sam understands why local connectivity can work without internet access.” The second statement goes beyond what was directly observed.
Why does this matter? Sam could understand the distinction and have consulted the source to check the wording. Sam could also repeat it without understanding what a local network is. Both explanations fit the observation. That uncertainty does not establish that Sam lacks understanding; it means we need more evidence to distinguish the possibilities.
Familiarity is the feeling that something is recognisable or easy to follow. A clear explanation can feel convincing before you have tried using the idea. Make It Stick supplies a useful caution about relying on that feeling. Our application is to ask what someone can do with an explanation, while avoiding the opposite mistake of treating successful recall as proof of understanding.
3. Pause and write: use the distinction#
Allow two minutes. Imagine an internet-provider outage. The router and its Wi-Fi access point remain on. Could your laptop still be connected to the home Wi-Fi? Explain why in two sentences. You may consult the notes; record whether you did.
Compare your answer with a worked response
Yes. The laptop could retain its local wireless connection even though the onward internet connection is interrupted. These are different parts of the journey, so losing one does not necessarily remove the other.
The reasoning matters because “yes” alone leaves us unsure how the answer was chosen. An answer that separates the two connections is more inspectable. If you copied or adapted the explanation, that condition also belongs in the record.
4. Why keeping the source can be legitimate#
If the question is “Can Sam explain this without a reference?”, then removing the reference makes sense. But suppose the question is “Can Sam accurately interpret a support article for someone else?” Consulting the article belongs to that task. Removing it introduces a demand on memory and changes what is being examined.
Clark and Chalmers’ The Extended Mind challenges the assumption that thinking must be understood entirely within the individual. It argues that external resources can participate in cognition. This philosophical argument gives us a reason to examine the role of a resource, rather than reject it merely for being external. It does not show that every assisted answer is competent.
For this course, record the task and its conditions before interpreting the result. “Explained with notes available” is a useful description, not an apology. Later weeks will examine the difficulties of separating a person’s contribution from a collaborator’s, including AI. For now, preserve the distinction.
5. Choose a follow-up that answers your uncertainty#
Suppose you want to know whether Sam can use the idea rather than repeat its wording. Ask about a changed situation:
Another device freshly loads the same website at the same time, using the same home Wi-Fi, with mobile data off. Does this prove the laptop’s hardware is broken?
Pause for three minutes. Give a reason and name one further observation you would want. Keep the source available so that the task still concerns reasoning with support, not unaided recall.
Worked response: what the new evidence establishes
No. The successful device shows that the website is reachable through this home connection for at least one device at that time. It does not establish why the laptop failed. Its browser or network configuration could be involved; hardware failure is only one possibility.
Trying another browser on the laptop could provide additional evidence, but would not settle every possibility. This is an example of reasoning from limited information, not a complete troubleshooting guide.
The follow-up has a purpose: it asks whether Sam can limit a conclusion in light of a new observation. A question containing harder networking vocabulary would not necessarily help us distinguish comprehension from repetition.
A good response gives us more evidence, but the follow-up may also teach Sam something. An explanation produced after discussion should not be treated as proof that the same explanation existed before the discussion.
6. Keep the two questions separate#
We have uncertainty about the network: what actually caused the failure? We also have uncertainty about Sam: what does this response show about Sam’s understanding? Naming a plausible fault is not the same as demonstrating the reasoning that makes it plausible.
A careful examiner might record:
| Part of the record | Example |
|---|---|
| Observation | Sam rejected the hardware diagnosis and gave another possible explanation, while consulting the notes. |
| Interpretation | Sam used the supplied distinction to explain why this evidence does not identify a hardware fault. |
| Limit | This does not establish independent diagnosis; Sam may have developed the explanation during the probe. |
This record is our teaching tool, not a validated psychological scale. Its value is in making your reasoning inspectable and leaving room for correction.
7. Apply the lesson yourself#
Use the Week 1 investigation to retain your own answers and a short evidence record. Work alone or with a partner. The goal is not to catch someone failing. If an answer is strong, say what it supports and what further claim would still require evidence.
Before you finish, explain these distinctions in your own words:
- What you observed versus what you inferred.
- Using a reference competently versus demonstrating unaided recall.
- Finding a possible network explanation versus judging a person’s understanding.
If you cannot explain one yet, revisit that section and add an example. The revised explanation is evidence of what you can do now; preserve the earlier answer if you want to examine what changed.
Reading with a purpose#
Book companion: Brown, P. C., Roediger, H. L. III, and McDaniel, M. A. (2014). Make It Stick: The Science of Successful Learning. Belknap Press of Harvard University Press. Begin with Chapter 1, “Learning Is Misunderstood,” if you have access. Ask: why might an experience that feels like learning give an unreliable indication of what someone can later do? Author resources include the book’s endnotes. Those are supporting references, not the chapter itself. Buying the book is not required to complete this lecture.
Primary argument: Clark, A., and Chalmers, D. (1998). “The Extended Mind.” Analysis, 58(1), 7–19. Read the authors’ text, sections 1–3. Ask: when does a resource become part of doing the thinking? Bring one objection. We use this paper because it challenges the assumption that relevant competence must always reside inside an unaided individual.
What this leaves unresolved#
An accessible answer can be useful without settling what its user understands. Describe how it was produced, consider alternative explanations, and choose a follow-up that addresses the uncertainty. Next week asks whether fluent explanation is enough—or whether it can also persuade an examiner too quickly.