# Human Decision Making: How People Really Choose

> How fast and slow thinking work, why working memory is small, and how to design screens that match the way people decide.

**In short:** People decide with fast and slow thinking and a small working memory, so good screens match how people really choose.

From "UI/UX basics", a free course by Venkat R. Murty: https://designervenkat.online/learn/ui-ux/human-decision-making

## What you will learn

People do not decide like a calculator. We act fast, we get tired, and we forget things. This lesson explains how human decision making works and what it means for the screens you design. You will read about two thinking modes, the small size of working memory, and why showing options beats asking people to remember them.

- **Decision:** A choice between actions, such as tapping Buy or leaving the page.
- **System 1:** The fast, automatic way of thinking. It needs little effort.
- **System 2:** The slow, careful way of thinking. It needs attention and effort.
- **Working memory:** The small mental space where you hold information for a short time while you use it.
- **Chunk:** A group of items that your mind treats as one unit.
- **Recognition:** Seeing a clue and knowing what it is.
- **Recall:** Pulling information out of memory with no clues.
- **Cognitive load:** The mental effort needed to understand a screen and finish a task.

## What are the two ways the mind decides?

Many psychologists describe thinking as two modes. Daniel Kahneman made the idea popular in his 2011 book Thinking, Fast and Slow. He named the two modes System 1 and System 2. The Decision Lab describes System 1 as a near-instant process. System 2 is slower and needs more effort.

- **Fast and automatic:** System 1 sizes up a situation and jumps to a conclusion. It works well for familiar things, like spotting a red warning icon. Because it moves so quickly, it can also be biased and get things wrong.
- **Slow and deliberate:** System 2 works through a problem one step at a time. It uses attention, and researchers note that it costs a lot of mental energy. So people use it less often than the fast mode.
- **Usually both together:** The two systems are a helpful picture, not two real parts of the brain.

> **Tip:** Treat System 1 and System 2 as a planning model, not a medical fact. Design for quick, automatic reactions first. Then give careful thinkers the detail they need, in a clear order.

## Why is working memory small?

Working memory holds the few things you are using right now. In 1956, psychologist George Miller published a famous paper. It suggested that people can hold about seven items, plus or minus two. Later researchers refined this. Cowan proposed that young adults hold about four chunks. Miller himself said that memory is limited by chunks, not by a fixed count of items. Designers should not treat seven as a rule.

### How does chunking help fit the limit?

Chunking means grouping items into meaningful units. A ten-digit number is hard to hold in one piece. Short groups are easier to hold. The code below splits digits into groups of three.

```python
digits = '5551234567'
chunks = [digits[i:i + 3] for i in range(0, len(digits), 3)]
print(chunks)
print(len(digits), 'digits shown as', len(chunks), 'chunks')
```

Output:

```text
['555', '123', '456', '7']
10 digits shown as 4 chunks
```

The last group is shorter, and that is fine. What matters is that each group becomes one unit in the mind.

**Quick check:** A user must remember a 10-digit code across two screens. Which change best fits the limits of working memory?

A. Show all ten digits as one long string
B. Split the code into short labelled groups
C. Hide the code and ask the user to type it from memory
D. Use a smaller font so more digits fit on screen

**Answer:** B. Short groups turn the code into a few chunks, so the mind holds fewer units at once.

## Why does recognition beat recall?

Recognition is easier than recall. When a visible clue is present, people only need to notice it. When no clue is shown, they must search their memory alone.

1. **Show the options:** Display choices on screen, such as a list of recently used files. Do not make users type names from memory.
2. **Label the icons:** Add text to icons, at least for early use. One classic example describes a new app that used only unlabelled icons, which left new users confused.
3. **Keep the status visible:** Show the current filter, date range, or step on the screen. People should not have to remember what they chose.

## What is cognitive load in design?

Cognitive load is the mental effort a person spends to understand a screen and get a task done.

- **Intrinsic load:** The difficulty built into the task itself. People come to a site to get something done, so this load is not bad in itself.
- **Extraneous load:** Effort caused by poor design, such as clutter or unclear labels. This is the kind that designers should cut.
- **Germane load:** Useful effort that helps people learn and understand. Good design keeps room for it.

Good design does not remove all effort. It removes the unnecessary kind. When load gets too high, people miss details, make errors, and may give up the task.

**Quick check:** Which statement best describes System 1 thinking?

A. It is slow, careful, and step by step
B. It is fast and automatic, and needs little effort
C. It is used only by experts
D. It is a fixed part of the brain that scientists can measure

**Answer:** B. System 1 is the fast, automatic mode. System 2 is the slow, careful one. The two are a model, not separate brain parts.

## Sources

- [System 1 and System 2 Thinking (The Decision Lab)](https://thedecisionlab.com/reference-guide/philosophy/system-1-and-system-2-thinking)
- [There Are Two Types and Two Systems of Cognitive Processes (Psychology Today)](https://www.psychologytoday.com/us/blog/theory-knowledge/202204/there-are-two-types-and-two-systems-cognitive-processes)
- [Dual Process Theory: System 1 vs System 2 (Yu-kai Chou)](https://yukaichou.com/behavioral-analysis/dual-process-theory-system-1-system-2-kahneman/)
- [Vector Symbolic Algebras for the Abstraction and Reasoning Corpus (arXiv paper, section on dual process cognition)](https://arxiv.org/pdf/2511.08747)
- [The Magical Number Seven, Plus or Minus Two (Wikipedia)](https://en.wikipedia.org/wiki/The_Magical_Number_Seven,_Plus_or_Minus_Two)
- [Miller's Law: Why the Magic Number 7 Is a Myth (Yu-kai Chou)](https://yukaichou.com/behavioral-analysis/millers-magical-number-seven-plus-minus-two-working-memory/)
- [Miller's Law (Think Insights)](https://thinkinsights.net/leadership/millers-law)
- [4. Miller's Law, Laws of UX (O'Reilly)](https://www.oreilly.com/library/view/laws-of-ux/9781492055303/ch04.html)
- [Designing for Cognitive Load (Master CAWEB)](https://mastercaweb.unistra.fr/en/actualites/ux-ui-design-en/designing-for-cognitive-load/)
- [Cognitive Load in UX: How to Design Simpler Experiences (Equinet Academy)](https://www.equinetacademy.com/blog/cognitive-load-in-ux-how-to-design-simpler-experiences/)
- [Cognitive Load in UX: 7 Principles for Better User Experiences (Orbix Studio)](https://www.orbix.studio/blogs/cognitive-load-in-ux-design)
- [Cognitive Load: Why Less Is Often More in Interface Design (Inside UX Design)](https://www.insideuxdesign.com/en/blog/cognitive-load-why-less-is-often-more-in-interface-design)
- [What Is Cognitive Load? (NN/g video)](https://www.nngroup.com/videos/cognitive-load/)
- [Design for the Human Mind: How Cognitive Science Shapes UX Design (Design Bootcamp)](https://medium.com/design-bootcamp/design-for-the-human-mind-how-cognitive-science-shapes-ux-design-f0b83e3da6b6)
