What Have We Learned About Better Explanations?
A summary of tips and ideas that connect cognitive science to clear communication.
This newsletter is based on a big idea: We can become better communicators by understanding the mechanics of the mind.
Since January of 2026, I’ve published 24 posts with this goal. Most of those posts focused on science: the researchers and experiments that led to a greater understanding of how we think and learn.
Many posts ended with a “Tips for Explainers” section where I applied the research to the real world of communication. For example, we learned the science behind why context is more powerful when it comes first.
Now that 6 months have passed, I’m taking stock of what we’ve learned about explanation as we follow the path an idea takes toward memory.
DISCLOSURE: The points below were distilled and written by an AI tool called Gemini Notebook (formerly NotebookLM). It looked at all 24 posts and distilled these points, with edits and videos by me. Details on the tool below.
What Science Has Taught Us About Explanation So Far
1. Preparation and Audience Strategy
Treat explanation as a “prediction problem”: People don’t share knowledge directly. Everything we communicate is a representation of what we know. How we represent our knowledge, and how it is perceived by others, is the basis for explanation failure or success. This requires making accurate predictions about what will work for others.
Acknowledge unique “Sensory Bubbles”: Everyone perceives a unique version of reality based on their biology and experience. Successful explainers work to create a shared baseline so they aren’t talking past their audience.
Avoid the “Curse of Knowledge”: The more you know, the harder it is to imagine not knowing. Don’t assume your audience shares your context; make the invisible visible for them.
Source: Inside the Sensory Bubble
Source: The Baseline Problem
Source: What Happens When We Explain (Original video included)
2. Earning Initial Attention
Nothing is possible without first earning attention.
Lead with relevance: In your first few sentences, signal why the information is urgent or relevant to the audience’s goals, threats, or emotions.
Prioritize signal strength: The mind filters physical characteristics first. Ensure your explanation is loud, clear, visible, and easy to hear to break through the internal “bottleneck” that limits what we can perceive.
Reduce competition: The mind can only process one channel at a time. Try to make your message the only available signal to prevent multitasking.
Set the tone immediately: An audience gets the conceptual “gist” of a scene in about 100 milliseconds. This initial impression happens before you even speak. Think about your first impression.
3. Building Context and Meaning
Provide context BEFORE content: Sequence is critical. Providing context first can help an audience remember twice as much information as providing it later.
Lower the “cost of understanding”: Understanding a new idea comes with costs: time, mental effort, etc. Give your audience a reason to stay engaged by focusing on context, meaning, and big ideas more than details.
Connect to personal context: When we learn, we take in raw information and construct understanding that’s based on our knowledge and experience. This personal context is how we find meaning and make sense of the world around us. The best explanations consider the personal contexts of the audience and what is meaningful to them.
Fill the gaps: Often, the key to explanation is building context that bridges the gap between the big idea and the personal context of the audience.
Source: Your Mind on Context
4. Working with Mental Limits
Respect the “Magic Number Four”: Working memory is limited; when isolated, the mind typically only holds about four chunks of information at once.
Use “Chunking”: Group related ideas together into small, related packages. This allows the audience to think through multiple complex ideas simultaneously.
Translate for novices: Experts think in highly compressed chunks. To be effective, experts must provide the details that allow novices to build their own chunks.
Encourage “Slow Thinking”: Design your explanation to move past the fast, intuitive reactions of System 1 and survive the deliberate reflection of System 2, which is where understanding becomes knowledge.
We’re Just Getting Started
The list above is incomplete because I continue to research and learn about how the mind processes information. We haven’t yet encountered ideas like reasoning, mental models, encoding, and more.
My plan is to keep following the path an idea takes as it becomes a memory. Once that path is complete, it will become a powerful framework for thinking about communication and explanation. I call it the gauntlet of the mind.
But there is another level to this. I am also fascinated with neurodiversity and how we can account for minds that process information differently. Once we have a neurotypical path, we can start to think about how to account for differences we encounter in everyday life.
About My Use of Gemini Notebook
I’m a user of an AI tool called Gemini Notebook (formerly called NotebookLM) by Google. Unlike ChatGPT or Claude, it only knows about what you provide it. For example, you can upload a dozen PDFs to a Notebook and use the chat interface to explore what’s in them.
I created a specific Notebook for collecting links to all Science of Explanation posts. This way, I can ask it to analyze and summarize what I’ve written.
For this post, I made this request:
It dutifully got to work and produced the points you see above, all based purely on the content I provided.
If You Enjoyed This Post
Please consider hitting the ♡ button below (it helps others find this post)
I’d love to hear from you. Leave a comment, and we can chat.
Know someone who would enjoy this? This post is public, so feel free to share.