Seven stages of achieving goals, Psychology in good design, Working hard is a skill

Weekly I/O #105: Seven Stages of Action Cycle, Six Psychological Concepts in Design, Working Hard is Skill, Distributed Cognition, Four Factors for Fair Use

Jun 29, 2025

Weekly I/O is now available in Chinese! If you're a paid subscriber, you can receive all future Chinese translations directly and for free. Simply reply to this email, and I'll add you to the list. Here's the first translated post: 中文翻譯版本. I hope you find it helpful!
This week's edition of the newsletter is free to everyone! But if you'd like full access to more posts like this and all the Weekly I/O archives, consider subscribing!

Hi friends,

Here's a list of things I enjoyed learning this week:

This week's inputs are very useful but quite complex. It's worth exploring them further to get the most out of it, hence the longer content.

Let me know your thoughts!

Input

Here's a list of what I learned this week.

1. How does design help users achieve their goals? Examine your design using the framework of the Seven Stages of the Action Cycle: Goal, Plan, Specify, Perform, Perceive, Interpret, and Compare.

Book: The Design of Everyday Things

I oftentimes get annoyed by poor design in everyday life, whether it's a website form, an electronic controller, or door handles. I hate the confusing one more than the ugly one because design is how things work.

Cognitive scientist and usability engineer Donald Norman proposed the Seven Stages of the Action Cycle, a framework designers can use to understand how people interact with their design, particularly in the context of achieving goals. The seven stages are:

Goal: Decide what you want
Plan: Plan how to achieve the goal
Specify: Determine the exact steps
Perform: Execute your plan
Perceive: Notice what changes occur
Interpret: Understand what the changes mean
Compare: Check if the result matches your goal

Use watching a movie at home as an example. Initially, you form a goal: to enjoy a film. Next, you translate that goal into a clear plan: "I'll press the play button." Then, you determine precisely how to carry out this intention and pinpoint the right action. Finally, you physically execute the action by pressing the button.

These are the first four stages: Goal, Plan, Specify, and Perform.

Now, after the execution, you perceive the results, see the TV screen change, and then interpret what this means; perhaps the movie starts playing immediately, or a confusing error message appears. Finally, you evaluate whether your goal has been successfully met.

These are the last three stages: Perceive, Interpret, and Compare.

When a product is well-designed, the first four stages should enable users to easily figure out how to perform actions (Execution), and the last three stages should help users effortlessly understand the outcome (Evaluation).

Poor design fails because it leaves users confused about which actions to perform (execution failure) or unsure whether their actions succeeded (evaluation failure). This is tightly related to the six fundamental psychological concepts for good design.

Another framework I noted before for evaluating design is Discoverability and Understanding.

2. Six fundamental psychological concepts for designing good products: Affordances, signifiers, constraints, mappings, feedback, and conceptual models.

Book: The Design of Everyday Things

In Discoverability and Understanding from I/O#62, we learned one of the essential characteristics of good design: discoverability (how easy it is to identify what actions are possible and know where and how to perform them).

Good discoverability relies on the following six fundamental psychological concepts:

Affordances indicate possible actions.

An affordance is a relationship between the properties of an object and the capabilities of the agent that determine how objects can be used.

For example, chairs afford sitting, buttons afford pressing, and handles afford pulling. A good affordance clearly shows the intended use without instructions.

Affordance is not a property but a relationship that depends on both the object and the agent. If an affordance is not perceivable, a signifier is needed to indicate its presence.

Signifiers guide you exactly where to perform an action.

For example, a flat door plate signals "push". Signifiers convey information about possible actions and how they should be done.

For signifiers to be effective, they must be easily perceivable. Otherwise, they fail to serve their purpose.

Distinguishing between signifiers and affordances can be challenging. The way I understand it is that "affordances define what actions are possible, while signifiers indicate where those actions should take place."

Constraints limit mistakes by restricting actions.

They limit our choices to guide us effortlessly. There are roughly four types of constraints:

Physical Constraints: USB connectors fit only one way.
Cultural Constraints: Red traffic lights universally mean "stop."
Semantic Constraints: Motorcycle windshields imply riders face forward.
Logical Constraints: Switches placed logically next to corresponding lights.

Mappings connect controls and their effects.

For example, imagine a classroom or auditorium with numerous ceiling lights and a row of light switches located on the front wall. Each switch controls a specific light, and this mapping defines which switch operates which light.

A good mapping to identify which control corresponds to each light in a large space is to arrange the controls in the same pattern as the lights. This also applies to stove knobs: If the knobs match the burner positions, turning on the correct burner is easy.

Feedback provides immediate clarity after actions.

For example, hearing a door shut, clicking a mouse, or seeing an icon change provides clear feedback. This feedback should be immediate because even a delay of just a tenth of a second can be disorienting. If the delay is too long, users may lose interest and move on to other activities.

Moreover, feedback must be informative. Feedback that is either vague or incorrect does not help the user.

However, it's important to note that too much feedback can be just as frustrating as too little. The annoyance caused by a "backseat driver" is widely recognized.

While backseat drivers may be correct, their constant comments can become distracting rather than helpful. Similarly, machines that provide excessive feedback can act like backseat drivers, ultimately hurting the user's experience.

Conceptual Models help users understand systems intuitively.

A conceptual model is a simplified explanation of how something functions.

It doesn't need to be complete or entirely accurate as long as it is useful.

For example, the files, folders, and icons displayed on a computer screen help users create a mental model of documents and folders stored on the computer. However, in reality, there are no physical folders inside the computer. These are just effective representations designed to enhance usability.

Designers must ensure that their conceptual models align with those of their users. That's why designers should always talk directly to the user to make sure their conceptual models are aligned.

3. Working hard isn't just about attitude. Working hard and trying your best every time is a skill that can be trained to differentiate players and sustain successful careers.

YouTube: T.J. McConnell Is Still Trying to Prove He Belongs in the NBA - YouTube

I became a fan of T.J. McConnell for his performance in the 2025 NBA Finals. Despite not being the most physically gifted, this 6'1" guard displayed an incredible level of energy and hustle throughout the series.

In his words in a recent interview:

"I genuinely believe that playing hard is a skill because if it wasn't, everyone would do it."
"When I came in, and I'm trying to tell you about this guy, Chris Babcock, he's like, 'You need to stand out.' He's like, 'Nobody picks up full court. And you need to make that your calling card.' And so we got in the gym three times a day, got in elite shape, and he's like, 'You're going to pick everyone up full court, and that's what you're going to hang your hat on.' And it's something that I've created an identity about."
"If you play really hard, and make life hell for the other team, I mean the other stuff, the offense, that sh– will fall into place. But if you're going and playing to exhaustion every time you're out there, it gives you a chance to make a roster and get more minutes. I mean, I feel like I've made a 10-year NBA career on just outplaying people,"

I believe working hard is also a skill. If it weren't, everyone would do it. It's just not the traditional skill that you have a set of very tangible practice.

Instead, it's a skill that requires training your mental muscles to resist short-term gratification, curate an environment that prevents distractions, and cultivate the habits necessary for hard work.

4. Distributed cognition theory shows how our thinking naturally extends into our external tools, environments, and interactions. This connection happens in three dimensions: material, social, and temporal.

Paper: Digital Humanities and Distributed Cognition: From a Lack of Theory to its Visual Augmentation | Published in Journal of Cultural Analytics

Why it's easier to remember something if you write it down? Why it's easier to solve a tough problem after talking it over with someone else?

These experiences illustrate Distributed Cognition (DC), a theory that expands the concept of thinking beyond the brain.

According to DC, our minds aren't confined to internal mental processes. Instead, cognition naturally spreads outward into external tools, collaborative interactions, and cultural practices.

For example, early sailors navigating oceans didn't just rely on memory. Instead, they integrated their thinking with external aids, such as maps, compasses, and teamwork, to turn navigation into a collective, externalized cognitive process.

Moreover, Distributed Cognition happens in three key dimensions:

Material distribution: We utilize external tools (such as calculators, computers, or even simple notepads) to handle tasks that our brains alone would find overwhelming.
Social distribution: We collaborate, sharing ideas and creating joint solutions that surpass what individuals achieve alone.
Temporal distribution: We leverage accumulated knowledge and cultural traditions stored externally, enabling us to learn from past generations.

Together, these dimensions illustrate how our intelligence naturally functions by interacting with external resources.

5. Fair use decisions in copyright law rely on four factors: purpose of use, nature of work, amount taken, and market effect.

Article: Training AI is Fair Use, Product Protection Versus LLM Liability, Piracy and Competition – Stratechery by Ben Thompson

Can an AI legally learn from copyrighted books? Recently, a federal court tackled this issue in a case involving Anthropic's AI training process. And their answer is yes.

Copyright encourages creativity by granting exclusive rights, but these rights must balance incentive and public benefits. From the Stanford Library, U.S. judges assess fair use through four key factors:

The Purpose and Character of Your Use

How is the work being used? Transformative use, where the new work changes the original purpose or adds new insights, typically favors fair use. For instance, parodies or educational critiques are transformative.

The Nature of the Copyrighted Work

Is the original work factual or creative? Factual content, such as biographies or news articles, usually receives less protection, making fair use more likely. Creative works, such as novels and movies, receive stronger protection.

The Amount and Substantiality of the Portion Taken

How much content was used, and how important is that content? Even a small portion that's central or crucial, such as a song's iconic chorus, might not qualify as fair. On the other hand, using a larger amount that's necessary for commentary or criticism can still be fair.

The Effect of the Use Upon the Potential Market

Does the new work negatively affect the original's market value? If the secondary work directly competes with or replaces the original, judges usually rule against fair use. But if it targets a different market or enhances interest in the original, fair use is more likely.

In evaluating the fair use of AI, judges emphasized the transformative nature of training.

AI doesn't reproduce books. Instead, it learns to generate original content, similar to human education. Judges also acknowledged the necessity of using vast amounts of content for training, distinguishing this from direct copying.

Recap

Try answering these five simple questions to review and reinforce what you've learned:

Thanks for reading!

One small favor: please share which input you found the most helpful or intriguing! Just comment on Substack with a number!